Professional Documents
Culture Documents
Alcohol Textbook 4th Ed
Alcohol Textbook 4th Ed
TEXTBOOK 44TH
TH
E
EDITION
DITION
A reference for the beverage, fuel and industrial alcohol industries
NOTTINGHAM
ISBN 1-897676-13-1
Contents
Foreword ix
T. Pearse Lyons
Presient, Alltech Inc., Nicholasville, Kentucky, USA
1 Ethanol around the world: rapid growth in policies, technology and production 1
T. Pearse Lyons
Alltech Inc., Nicholasville, Kentucky, USA
2 Grain dry milling and cooking procedures: extracting sugars in preparation for
fermentation 9
Dave R. Kelsall and T. Pearse Lyons
Alltech Inc., Nicholasville, Kentucky, USA
12 Understanding near infrared spectroscopy and its applications in the distillery 145
Don Livermore1, Qian Wang 2 and Richard S. Jackson2
1
Hiram Walker & Sons Ltd., Walkerville, Ontario, Canada
2
Bruker Optics Inc., Billerica, Massachusetts, USA
14 Production of Scotch and Irish whiskies: their history and evolution 193
T. Pearse Lyons
Alltech Inc., Nicholasville, Kentucky, USA
15 Tequila production from agave: historical influences and contemporary processes 223
Miguel Cedeño Cruz
Tequila Herradura, S.A. de C.V. Ex-Hda San Jose del Refugio Amatitán, Jalisco, México
18 From liqueurs to ‘malternatives’: the art of flavoring and compounding alcohol 265
Andy Head and Becky Timmons
North American Biosciences Center, Alltech Inc., Nicholasville, Kentucky, USA
Foreword vii
21 Managing the four Ts of cleaning and sanitizing: time, temperature, titration and
turbulence 299
Jim Larson and Joe Power
North American Biosciences Center, Alltech Inc., Nicholasville, Kentucky, USA
Recovery
25 Understanding energy use and energy users in contemporary ethanol plants 355
John Meredith
Ro-Tech, Inc., Louisville, Kentucky, USA
Index 441
Foreword ix
Foreword
Alcohol production: a traditional process changing rapidly
T. PEARSE LYONS
President, Alltech Inc., Nicholasville, Kentucky, USA
In its simplest form, alcohol production is the liters/tonne. In fact, we predict yields of 3.2
process of preparing starch- or sugar-containing gallons in the near future. For molasses, the
raw materials for fermentation by yeast, which advent of detranases can ensure maximum
is currently the only microorganism used for yields. Every aspect of cooking and fermentation
converting sugar into alcohol. The ethanol is must be controlled, however.
then concentrated and recovered in a process
called distillation. Though essentially a simple
process, making it happen with maximum The cooking process
efficiency on a very large scale is a remarkable
combination of microbiology and engineering. Much debate exists at the moment regarding the
Over the past 25 years progress has been made cooking process. It is a sad reflection perhaps
in virtually all aspects of the process, but what on our industry that since the pioneering PhD
are the challenges facing alcohol in the future? work by Dr. John Murtagh in 1972 no further
Beginning with raw material processing, here scientific thesis has been published on cooking.
are the areas that we feel need to be addressed: Should grain be finely ground? Should cooking
involve high temperature α-amylase or should
an entirely different type of enzyme be used?
Raw material reception and processing Often the enzyme selected is one developed
based on experience in the corn wet milling
Despite many declarations to the contrary, it is sector, which may be far from ideal for dry mill
virtually impossible for a distillery to accurately alcohol production.
calculate true yields. A sampling of grains is Our industry needs a new enzyme, one tailored
either done in very cursory fashion or not at all. to our specific needs: converting grain to
The distiller must see himself first as a processor fermentable sugar. Raw materials contain not
of grain rather than a producer of alcohol. He only starch, but also hemicellulose (a polymer
must know accurately how much starch is of 5-carbon sugars) and cellulose. The cooking
present in his raw material and he must be able process must be designed to loosen these
to measure that immediately. NIR spectroscopy materials such that they release bound starch and
has much to offer in terms of raw material also become available in turn for fermentation.
applications; and this topic is detailed in this
volume. However, if we control the process and
use the appropriate enzymes from solid state Yeast and fermentation
fermentation, such as Rhizozyme™ and
Allzyme™ SSF, which are capable of releasing Despite all we know about it, yeast remains
not just bound starch but also hemicellulose and underrated and misunderstood. It is the
cellulose, we can and indeed some plants do, workhorse of the distillery, yet ‘economies’ on
get 2.9+ gallons per bushel (116 gallons or 420 quality and quantity of yeast are made; and we
x T.P. Lyons
STARCH
Corn Rye
Wheat LIQUIFACTION Potatoes
Barley Milo (sorghum)
Enzymes
DEXTRINS Cellulose
Molasses Agave
Sugarcane SUGAR Sugarbeets
Whey
CO-PRODUCTS
Yeast CO2
DDGS BIOREFINERY
Wet cake
Value-added
PRIMARY ALCOHOL Solubles
co-products
MATURATION SPIRITS
often force it to grow and operate under uncontrolled, will prevent the yeast from
unsuitable conditions. As Professor Mike performing. Operating a fermentation at 23%
Ingledew has pointed out many times, the ethanol is totally different than operating a
fermentor is not a garbage can! We must make distillery at 8% ethanol. Any minor change,
sure that conditions in fermentation are optimum whether in mycotoxin level, salt level or nutrient
for yeast. level can have a major impact on performance.
Like our need for appropriate enzymes, ethanol
production also requires yeast with higher
ethanol and temperature tolerance if we are to Microbial contamination
keep progressing to higher levels of alcohol in
the fermentor. Today, 17-20% abv is becoming A distillery is never going to be a sterile
standard in our industry as we move to high fermentation and even those of us who have had
temperature Thermosacc TM -like yeast. When the good fortune to operate in sterile fermentation
used in combination with enzymes capable of conditions know how easily infection can take
extracting increasing amounts of sugar, such as hold. Since the yeast is a relatively slow-growing
those produced by surface culture fermentation, microorganism compared to most of the
yeast can bring us to a whole new level. infectious microorganisms (Lactobacillus, etc.),
While 23% ethanol by volume in the fermentor it is critical that we have an arsenal of
without loss of performance is possible given antimicrobial agents and effective cleaning
the right yeast and enzyme combination, it can products and programs ready to protect the
only be done by careful management of the so- fermentor environment.
called stress factors. Factors which, if left
Foreword xi
Recovery and utilization of co-products with PhD and masters degrees, our industry has
very few. Unlike the brewing industry, which
As we review our overall process, it is obviously has no less than five brewing schools around
critical that we maximize utilization of co- the world, the annual Alltech Alcohol School
products. Assuming the feedstock is corn, the remains the only venue for industry-wide
major cost factors are the raw material, energy, training.
enzymes, processing chemical cost (water The Alltech Institute of Brewing and Distilling,
treatment, etc.) and labor. Much of these costs established in 2000, set out to address this
can be offset with 17 lbs of distillers grains shortfall by working in conjunction with the
coming from each bushel of corn; but we must Heriot-Watt University in Scotland. Perhaps our
maximize the return on this essential product. industry should use the brewing industry as a
Distillers grains with solubles (DDGS) are model. Scholarships should be funded and
currently (July 2003) selling for around $80/ton perhaps a bioscience center concept established.
and corn for $2.20 per bushel. At $0.04 per lb Bioscience centers bridge the gap between
($80/ton), 17 lbs of DDGS gives us a credit of university and industry by creating an
$0.68. This is set against the $2.20 bushel of environment where students are encouraged to
corn or the $0.03-0.04 for enzymes. If on the complete higher degrees while doing industry-
other hand, new or novel products could be focused research.
made from distillers grains, (i.e. improved by-
pass proteins for cattle, pre-hydrolyzed DDGS
for human foods), the financial impact could be Conclusions
substantial. The future will see many new
products built around DDGS. This is the subject The alcohol industry, therefore, is alive and well,
of a separate chapter in this volume. and we have been given an opportunity to
As dry mills shift toward biorefining, instead achieve something that many of us thought
of simply grinding, corn, yield of ethanol would have been achieved by the mid-1980s.
becomes only one part of the economic We have a superb oxygenate that will help
equation. Perhaps in the future, further reduce global warming and at the same time
processing of distillers grains will create products enable us to add value to our grains. The world
that make spent grains the single most important is not short of starch and sugar, the world is short
raw material from a distillery. After all, they are of protein. The fuel alcohol industry and
rich and valuable protein sources, a commodity beverage alcohol industries therefore are in fact
in very short supply across the world. generators of protein; it only remains for us to
make sure the proteins we generate are the most
advantageous possible to man and beast.
Education The chapters in this book we hope will help
the reader realize the complexity and at the same
Perhaps the biggest surprise, as one surveys the time the simplicity of the process of converting
progress that has been made in the last 20 years sugars to ethanol. If we make your search for
is how little progress has been made in the area information just a little easier, we will have
of education. Unlike the brewing industry, which achieved our objective. If we can encourage
seems to have an abundance of professionals more research, then we will be well-rewarded.
Ethanol industry today
Ethanol around the world: rapid growth in policies, technology and production 1
Chapter 1
T. Pearse Lyons
President, Alltech Inc., Nicholasville, Kentucky, USA
nowhere near the forecasted 7 billion liters (1.85 UNPRECEDENTED GROWTH IN THE US
billion gallons), and instead languished around
1.5 billion liters (0.4 billion gallons). Beverage In the United States, domestic events have given
ethanol remained the dominant ethanol product. the industry a big push. The US Clean Air Act
To most, it looked like the gasohol boom was mandated an average of 2% oxygen be present
over. in what was called ‘reformulated gasoline’
While smaller operations disappeared, other (RFG). Representing some 30% of the vast US
larger companies, well funded and with well- gasoline market (100 billion gallons), it created
designed plants, continued to grow. Corporations a clear demand for an oxygenate. MTBE (methyl
such as Cargill, Staleys and ADM became tertiary butyl ether), with approximately 18%
industry leaders. Today, one of these (ADM) is oxygen, and ethanol with 30% oxygen were
responsible for processing 45% of the corn that chosen. Both were excellent oxygen boosters.
is converted into ethanol in the US. This MTBE, a petroleum derivative, was the early
company and others of this size knew the ethanol choice, but a series of scares regarding its
industry had long term prospects and they possible carcinogenic nature and its detection
planned survival strategies accordingly. in groundwater led to its planned phaseout. The
Meanwhile in Brazil during the 1970s and alcohol industry, meanwhile, was beginning to
1980s, a similar revolution was occurring. In emerge, particularly in the Corn Belt (Nebraska,
1975 the so-called ‘Proalcohol’ program was Iowa, Minnesota and Illinois). Its capacity had
launched, with massive support by the reached 10 billion liters (2.7 billion gallons) by
government. Initially, the program focused on 2003 (Figure 2) and in anticipation of the
the production of anhydrous ethanol, which was phaseout of MTBE (not just in California, but
blended with gasoline to produce the Brazilian around the country) some 73 plants had been
gasohol. The emphasis quickly changed to built. At the time we are going to press, these 73
hydrous ethanol (95% ethanol/5% water), which plants, located in 20 states, have the capacity of
could be used in its pure form in cars with 11 billion liters (Figure 3). A further 13 plants
specially designed engines. By the mid-1980s, are under construction, which will add some 2
such cars represented the vast majority of the billion liters (500 million gallons) of capacity.
marketplace and it seemed the Proalcohol The ethanol industry also continues its streak
product was going to replace gasoline on a 1:1 of monthly production records. An all-time
basis. monthly production record of 181,000 barrels
The system, however, was very inflexible. per day (b/d) was set in June of 2003, according
When ethanol became difficult to get in the late to the US Energy Information Administration
1980s, consumers began to switch back to (Figure 4). The previous all-time record was
conventional cars in which gasoline (still 179,000 b/d, set in April of this year.
blended with a minimum level of ethanol) could
be used. Was the fuel ethanol program also Driving industry growth: Farmer-owned ethanol
going to falter in Brazil? plants
2500
2000
Millions of gallons
1500
1000
500
0
80
88
90
98
00
82
86
92
96
02
84
94
19
19
19
19
20
19
19
19
19
20
19
19
Figure 2. Historic US fuel ethanol production.
Under constructioon
190
Thousands of barrels/day
180
170
160
150
140
130
120
110
Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun
1000
800
Millions of bushels
600
400
200
0
80
88
90
98
00
82
86
92
96
02
84
94
19
19
19
19
20
19
19
19
19
20
19
19
processed into ethanol in 2003, or about 10% to December 31, 2003, all major refiners
of the total US harvest. Since 1999, farmer- voluntarily switched to ethanol in early 2003,
owned ethanol plants have increased their resulting in a new market for more than 600
percentage of total production capacity from million gallons. In 2004, with the MTBE ban in
20% to nearly 35%. As a whole, farmer-owned place, California’s ethanol consumption will
ethanol plants are the largest ethanol producer increase to 750-800 million gallons annually.
in the US. MTBE bans in New York and Connecticut are
also scheduled to take effect on January 1, 2004.
New marketplace opportunities These two states combined represent a market
for more than 500 million gallons of ethanol. If
Following concerns regarding water all of the northeast were to replace current MTBE
contamination with MTBE, the state of California use with ethanol, a market for more than a billion
passed landmark legislation phasing out the use gallons of ethanol would be created. In total, 17
of MTBE in the state’s motor fuel supply. While states have passed legislation to phase out MTBE
the original deadline was delayed by one year use; and the future for ethanol looks bright.
Ethanol around the world: rapid growth in policies, technology and production 5
Comprehensive energy legislation offers higher than 2002, we can anticipate a parallel
opportunities for ethanol increase in production of feed co-products. A
waiting market exists in the world’s animal feed
The nation’s increasing reliance on imported industry (>600 million tonnes) where
energy, particularly from unstable regions of the traditionally used sources of protein such as
world, combined with erratic natural gas and animal by-products and fish meal have been
world oil markets has prompted the either eliminated due to concern surrounding
administration and US Congress to debate Mad Cow Disease (BSE) or reduced availability
comprehensive energy policy legislation. As part and higher prices. The combined protein and
of this legislation, an historic fuels agreement energy value of ethanol by-products gives them
supported by an unprecedented coalition of tremendous potential in animal feeds across the
agriculture, oil and environmental interests world.
would provide a growing market for renewable
fuels while addressing concerns regarding
MTBE water contamination and providing BRAZIL AND SOUTH AMERICA
needed flexibility in the fuels marketplace. This
agreement would establish a Renewable Fuel The US, however, is not the only country
Standard (RFS), which would set an annual and experiencing an increase in alcohol production.
increasing minimum level of renewable fuels In Brazil, new initiatives are underway and the
such as ethanol and biodiesel permitted in the inflexible Proalcohol program has been changed.
marketplace, growing to 5 billion gallons by Flex-fuel engines, which run equally well on
2012. either pure hydrous alcohol or gasoline, have
been launched. This technology can cope with
New uses for ethanol anything from 100% hydrous ethanol to a
gasohol mix of 80% gasoline and 20%
While ethanol today is largely used as a blending anhydrous ethanol, adjusting to any change
component with gasoline to add octane and within a fraction of a second. This compares very
oxygen, opportunities abound for ethanol to favorably to the engines in the US, which would
play a role in advanced vehicle technology and normally handle up to 85% anhydrous in a blend
alternative fuels markets. For example, as a and cannot use pure hydrous ethanol.
renewable fuel with an established fueling Furthermore, since ethanol sells in Brazil for less
infrastructure in the US, ethanol is an attractive than gasoline at the pumps (50-60% of the
fuel option for the fuel cell market in both power gasoline price) even taking into account that 30%
generation and transportation. Research is also more ethanol is required to travel the same
underway on E dieselTM, a mixture of ethanol, distance, the new cars are bound to be a success
diesel and a blending agent. And there is a and will drive up ethanol demand.
largely untapped market for E85, a blend of 85% It is expected that a million cars with Flex-fuel
ethanol and 15% gasoline, to power the growing engines could be sold each year. With an average
number of available flexible fuel vehicles. 240 liters (63 gallons) of ethanol per month
consumption, this could mean an additional 3
Valuable feed co-product production increasing billion liters would be needed over Brazil’s
current 12-13 billion liters of production.
Record ethanol production means record
production of valuable feed co-products for Table 1. Projected production of ethanol (billion liters)1.
livestock. In 2002, US dry mill ethanol facilities,
2003 2005
which account for approximately 65% of ethanol
production, produced about four million short USA 10 18
tons of distillers dried grains with solubles Brazil 13 16
(DDGS). Ethanol wet mills produced European Union Unknown 8
India 1.5 1.8
approximately 450,000 US tons of corn gluten
Thailand Unknown 0.7
meal, 2.5 million tons of corn gluten feed and Australia .04 .4
germ meal, and 530 million pounds of corn oil. China Unknown 2
With estimates for 2003 production nearly 30% 1
F.O. Licht, July 2003
6 T.P. Lyons
While few people will say exactly how much it bioethanol as a way to reduce oil imports. It was
costs to produce a liter of fuel ethanol in Brazil, estimated that 0.7 billion liters of ethanol would
it is believed to be between $0.10 and $0.15. be required should they decide to introduce the
This puts the industry in a very favorable oxygenate at 10% of gasoline. Both cassava
position to export ethanol, since a similar cost (tapioca) and molasses were suggested as raw
in the US from corn would be more like $0.20. materials and eight projects were approved. At
Fuel ethanol emergence in South America is this time only one project is under construction.
not limited to Brazil. In Peru and Colombia Two distinct areas for ethanol production exist
interest is also growing. Colombia estimates an in China: the maize-rich north and the sugar-
annual need for almost one billion liters and 12 cane rich south. Both will feature in the country’s
production plants. ethanol future. The government sees ethanol as
part of its program to reduce pollution –
important with the looming 2004 Olympics –
THE EUROPEAN UNION and at the same time encourage rural
employment. China’s largest fuel ethanol
Driven by the Kyoto Agreement, the European facility, costing almost $300 million to construct,
Commission proposed in 2001 to increase the is in Jilin. With a capacity of some 600 million
use of crop-based, non-polluting renewable liters (165 million gallons), it will use 2 million
fuels. The target was for biofuels to account for tons of maize a year (13% of production in that
2% of all fuels in EU transport by 2005. It was province). By October 2003, all vehicles in the
mandated this would increase to 5.75% by 2010 Province of Jilin will use fuel containing at least
(from 0.3% in 2001). 10% ethanol.
Both ethanol and biodiesel (from canola) are
considered biofuels. In 2003 the targets were
made voluntary, but a direction on tax relief for AUSTRALIA
renewable energy was expected to be put into
place. These would augment the existing tax Uncertainty also exists in Australia, where a
relief schemes, which vary from country to target of 350 million liters was encouraged with
country. Spain, Germany, and Sweden, for tax exemptions and grants for up to six years.
example, have 100% tax relief on renewable Local governments have tried to boost ethanol
fuels, while France has 60% and Britain 40%. consumption. For example, in Cairns,
Across Europe, from Poland to Ireland, interest Queensland’s largest city, government-owned
is rising as the new legislation begins to take vehicles were mandated to use E10 (90%
effect. gasoline, 10% ethanol). Using some 20,000 liters
per day, the government hoped in this way to
avoid some of the negative publicity being
ASIA generated by the anti-ethanol lobby.
India’s transport sector accounts for over half
of the country’s oil consumption. It is not Conclusions
surprising, therefore, that interest is growing in
replacing dependence on fossil fuels. By January Ethanol, as a global product, is experiencing a
2003, nine states had mandated at least 5% significant jump in production and production
ethanol (E5) in transport fleets. Local distilleries capacity. Whether it be used domestically or
with a capacity of some 3.2 billion liters (900 exported to areas around the world (such as
million gallons) are currently only operating at Japan, Singapore and even California) where
50% capacity. The demand for an additional 400 production is not sufficient or not possible, it is
million liters under this new mandate is therefore clear it will become a global commodity. The
immediately being met. Although some political future, therefore, is very positive; and this future
maneuvering is still needed, the fuel ethanol will not be one dependent on subsidies. A 1997
program looks sound. study by the Kellogg School of Management at
In Southeast Asia, Thailand’s government, like Northwestern University concluded the net effect
many in the region, had an early interest in
Ethanol around the world: rapid growth in policies, technology and production 7
Chapter 2
This chapter deals with the milling and cooking facilitate subsequent penetration of water in the
stages of alcohol production from whole cereals. cooking process. The optimum size of the
In brief, in this dry milling process the whole ground particle is the subject of disagreement.
cereal, normally corn (maize), is ground in a mill Some engineers believe the particle should be
to a fine particle size and mixed with liquid, as fine as possible to allow the water maximum
usually a mixture of water and backset stillage. access for hydrolysis of starch while others
This slurry is then treated with a liquefying believe a better yield is obtained if the particles
enzyme to hydrolyze the cereal to dextrins, which are larger and the jet cooker can act on the
are a mix of oligosaccharides. The hydrolysis particles. The key is simply to expose the starch
of starch with the liquefying enzyme, called α- without grinding so fine as to cause problems in
amylase, is done above the temperature of co-product recovery. A wide variety of milling
gelatinization of the cereal by cooking the mash equipment is available to grind the whole cereal
at an appropriate temperature to break down the to a meal. Normally, most distilleries use hammer
granular structure of the starch and cause it to mills, although some may use roller mills,
gelatinize. Finally the dextrins produced in the particularly for small cereal grains.
cooking process are further hydrolyzed to
glucose in a saccharification process using the
exoenzyme glucoamylase, and possibly another HAMMER MILLS
enzyme (Rhizozyme™). These enzymes may
also be added to the yeast propagation tank or In a hammer mill, the cereal grain is fed into a
the fermentor. These separate stages of milling, grinding chamber in which a number of
cooking and saccharification will be explained hammers rotate at high speed. The mill outlet
in more detail. This is called the dry milling contains a retention screen that holds back larger
process. Most of the new distilleries use this particles until they are broken down further so
process or a minor variation of it. that there will be a known maximum particle
size in the meal. The screens are normally in the
size range of 1/8-3/16 in. (2-4 mm).
Milling Sieve analysis (particle size distribution) shows
whether the hammer mill screens are in good
The purpose of milling is to break up the cereal order and whether the mill is correctly adjusted,
grains to appropriate particle size in order to and should be conducted regularly. Table 1
10 D.R. Kelsall and T.P. Lyons
shows a typical sieve analysis for corn. The two outside these specifications the mill should be
largest screens retain only 11% of the particles adjusted. Normally the hammers in a hammer
while the quantity passing through the 60-mesh mill are turned every 15 days, depending on
screen is also fairly low at 7%. For efficient usage; and every 60 days a decision should be
processing of the cereal starch into alcohol, the made as to whether or not to replace the hammers
particles should be as fine as possible. However and screens.
a compromise must be made such that the
Table 2. Typical alcohol yields from various cereals.
particles are not so fine that they cause balling
in the slurry tank or problems in the by-product Cereal Yielda
recovery process. Sometimes the sieve analysis (US gallons of anhydrous
will vary depending on the severity of the shaker alcohol/bushel)
and a consistent shaking should be done each Fine grind corn1, 3/16 in. 2.85
time the test is done. Coarse grind corn1, 5/16 in. 2.65
Milo 2.60
Barley 2.50
Table 1. Typical results of a sieve analysis of corn meal. Rye 2.40
a
Screen size Hole size (in) Corn on screen (%) Note that a distiller’s bushel is always a measure of weight.
12 0.0661 3.0 It is always 56 lb, regardless of the type of grain.
1
16 0.0469 8.0 Laboratory data using Rhizozyme™ as glucoamylase
20 0.0331 36.0 source.
30 0.0234 20.0
40 0.0165 14.0
60 0.0098 12.0 Since sieve analysis is critical, a case can also
Through 60 7.0 be made for recycling to the hammer mill any
grain not ground sufficiently fine. Fineness of
the grind also has an important bearing on
Fineness of the grind can be significant factor centrifugation of the stillage post-distillation. A
in the final alcohol yield. It is possible to obtain finer grind may yield more solubles and hence
a 5-10% difference in yield between a fine and place a greater load on the evaporator. However
a coarse meal. Table 2 shows the typical alcohol since the key is to maximize yield, dry house
yield from various cereals. It can be seen that considerations, while important, cannot override
the normal yield from corn is 2.85 gallons of yield considerations.
anhydrous ethanol per bushel (56 lbs). However,
the yield with coarsely ground corn may drop
to 2.65 gallons per bushel, a reduction in yield ROLLER MILLS
of 7.5%. This is a highly significant reduction
and would have serious economic consequences Some distillers use roller mills (e.g. malt whisky
for any distiller. Grinding is about expanding producers), particularly where cereals containing
the amount of surface area. When more surface substantial quantities of husk material are used.
area is exposed, water and enzymes are better In a roller mill, the cereal is nipped as it passes
able to penetrate the kernel. A particle 1 in. in through the rollers, thus exerting a compressive
diameter has a total surface area of 6 in2. If you force. In certain cases, the rollers operate at
grind the 1 in. particle into 1,000 particles, different speeds so that a shearing force can be
surface area becomes 6,000 in2. No amount of applied. The roller surfaces are usually grooved
heating or ‘exploding’ will overcome poor grind to aid in shearing and disintegration. Figure 1
as measured by yield. The key is to expose more shows the general configuration of a roller mill.
surface area, but to do so in a controlled manner. In Scotland the solids in whisky mash, made
Since grind is so important, it is recommended entirely from malted barley, are usually removed
that a sieve analysis of the meal be done at least by using a brewery-type lauter tun, which is a
once per shift. The distiller should set vessel with a perforated bottom like a large
specifications for the percentage of particles on colander. In this case, a roller mill should be
each sieve; and when the measured quantity falls used as the shear force allows the husk to be
Grain dry milling and cooking procedures 11
1st pair
Cooking is the entire process beginning with
of rolls mixing the grain meal with water (and possibly
backset stillage) through to delivery of a mash
Anti-explosion
device
ready for fermentation. Figure 2 shows the
components that make up a typical milling and
Cylindrical
cooking system. This schematic diagram could
screen
represent the processes involved in beverage,
Revolving
beater
industrial or fuel alcohol production, except that
sk
s nowadays only the whisky distillers use malt as
Hu
Fine grits Coarse grits 2nd pair a source of liquefying and saccharifying
and flour of rolls enzymes. All other alcohol producers use
microbial enzyme preparations. The fastest
enzyme systems have low calcium requirements
(2-5 ppm), lower pH optima (less buffering
required) and high temperature stability. They
To meal bin are designed to reduce viscosity, with less
Figure 1. Schematic of a roller mill. emphasis on DE. The key to cooking is to liquefy
the starch so it can be pumped to the fermentor,
hence viscosity is critical.
separated with minimal damage. The husk then
acts as the filter bed in the lauter tun for the
efficient separation of solids and liquid.
HYDROLYSIS OF STARCH TO FERMENTABLE
SUGARS
Cooking Step 1: gelatinization
The purpose of cooking is 5-fold: The purpose of cooking and saccharification is
to achieve hydrolysis of starch into fermentable
Sterilization. The mash must be sterilized so that sugars, which is accomplished by the
harmful bacteria are minimized. A brewer endoenzyme α−amylase, followed by the
achieves this by boiling mash (wort) while the exoenzyme glucoamylase (amyloglucosidase) to
distiller achieves it by cooking. release glucose. However in order for the α-
Solubilization of sugars. Sugars are solubilized amylase to gain access to the starch molecules,
to a certain point, typically 2-3% free sugars. In the granular structure of the starch must first be
this way yeast growth can occur rapidly but not broken down in the process known as
excessively as would happen if too much sugar gelatinization. When the slurry of meal and
was released at once. water are cooked, the starch granules start to
adsorb water and swell. They gradually lose their
Release of all bound sugars and dextrins (chains crystalline structure until they become large, gel-
of sugar). Extraction must occur such that filled sacs that tend to fill all of the available
subsequent enzymatic hydrolysis can ensure that space and break with agitation and abrasion.
all sugars are utilized. Starch (precursor of The peak of gelatinization is also the point of
sugars) is in large bound to protein and fiber, maximum viscosity of a mash. Figures 3, 4 and
which is are freed during cooking. 5 show the progressive gelatinization of
Protein breakdown to amino acids should be cornstarch, as viewed on a microscopic hot stage.
minimized. Amino acids and peptides can In Figure 3 the granules are quite distinct and
become bound to sugars in Maillard reactions, separate from the surrounding liquid. In Figure
which leaves sugar unavailable.
12 D.R. Kelsall and T.P. Lyons
Raw material
α-amylase
α-amylase (0.02%)
(0.04 - 0.06%)
(High T L120™)
(High T L120™)
Hammer mill
Fermentors
Still Mash cooler
32°C
4 these same granules have swollen in size and Table 3. Gelatinization temperature ranges of various
some of the liquid has entered the granules. feedstocks.
Figure 5 shows the granules as indistinct entities Gelatinization range (°C)
in which the liquid has entered to expand them
Corn
considerably. Standard 62-72
Gelatinization temperatures vary for the High amylose 1 67->80
different cereals (Table 3). Some distillers High amylose 2 67->80
consider it important for the slurrying Barley 52-59
temperature of the meal to be below the Rye 57-70
temperature of gelatinization. This avoids Rice (polished) 68-77
coating of grain particles with an impervious Sorghum (milo) 68-77
layer of gelatinized starch that prevents the Wheat 58-64
enzymes from penetrating to the starch granules
and leads to incomplete conversion. Many Step 2. Liquefaction: hydrolysis of starch to
distillers, however, go to the other extreme and dextrins
slurry at temperatures as high as 90°C (190°F).
At these temperatures starch gelatinizes almost Liquefaction is accomplished by the action of
immediately and with adequate agitation there α−amylase enzyme on the exposed starch
is no increase in viscosity and no loss of yield. molecules. Starch exists in two forms. One form
Grain dry milling and cooking procedures 13
Figure 3. Gelatinization of cornstarch. Starch granules viewed on microscope hot stage at 67°C (under normal illumination).
Figure 4. Gelatinization of cornstarch. Same granules as in Figure 3 at 75°C (under normal illumination).
is the straight-chained amylose, where the Amylopectin has a branched structure (Figure
glucose units are linked by α−1,4 glucosidic 7). It has the same α−1,4 glucosidic linkages as
linkages (Figure 6). The amylose content of corn in amylose, but also has branches connected by
is about 10% of the total starch; and the amylose α−1,6 linkages. The number of glucose units in
chain length can be up to 1,000 glucose units. amylopectin can be as high as 10,000. Corn,
The other form of starch is called amylopectin, wheat and sorghum (milo), the three most
which represents about 90% of the starch in corn. common feedstocks for ethanol production, have
14 D.R. Kelsall and T.P. Lyons
Figure 5. Gelatinization of cornstarch. Same granules as in Figures 3 and 4 at 85°C (under normal illumination).
Amylose
α-1,4 linkage
CH2OH CH2OH
6 6
O O
H 5 H H 5 H
4 H 1 1 4 H 1
O OH H O OH H O
3 2 3 2
H OH H OH
n
similar levels of starch, but percentages of Step 3. Saccharification: release of glucose from
amylose and amylopectin differ with grain and dextrins
with variety. Starch structures are reviewed in
greater detail in the chapter by R. Power in this Saccharification is the release of the individual
volume. glucose molecules from the liquefied mixture
The α−amylase enzyme acts randomly on the of dextrins. The dextrins will be of varying chain
α−1,4 glucosidic linkages in amylose and lengths. However, the shorter the chain length
amylopectin but will not break the α−1,6 the less work remaining for the exoenzyme
linkages of amylopectin. The resulting shorter glucoamylase, which releases single glucose
straight chains (oligosaccharides) are called molecules by hydrolyzing successive α−1,4
dextrins, while the shorter branched chains are linkages beginning at the non-reducing end of
called α-limit dextrins. The mixture of dextrins the dextrin chain. Glucoamylase also hydrolyzes
is much less viscous. α−1,6 branch linkages, but at a much slower
Grain dry milling and cooking procedures 15
α-1,6 linkage
Amylopectin
CH2OH
6
O
H 5 H α-1,4 linkage
4
H 1
O OH H O CH2 CH2OH
3 2 6 6
O
H OH H 5 H H 5
4
H 1 4
H
O OH H O OH
3 2 3
H OH H
rate. One real problem with HPLC analysis is that there are a variety of batch and continuous
that dextrins are given as total oligosaccharides cooking systems. For a batch system there is
with no differentiation between four glucose usually only one tank, which serves as slurrying,
units and 20 glucose units. The work of the cooking and liquefaction vessel. Live steam jets
glucoamylase is directly proportional to the are typically installed in the vessel to bring the
length of the dextrin chain; and there is currently mash to boiling temperature along with cooling
no way of knowing how well the α-amylase has coils to cool the mash for liquefaction. Figure 8
worked to produce small oligosaccharides. shows a typical batch cooking system.
In Scotch whisky production, malted barley is In the batch cooking system, a weighed
used as a source of both α−amylase and the quantity of meal is mixed into the vessel with a
exoenzyme ß-amylase. The fermentable sugar known quantity of water and backset stillage.
produced is maltose, a dimer made up of two These constituents of the mash are mixed in
glucose units. simultaneously to ensure thorough mixing. The
quantity of liquid mixed with the meal will
determine the eventual alcohol content of the
Cooking systems fermented mash. When a distiller refers to a ’25
gallon beer’, it means 25 gallons of liquid per
BATCH COOKING SYSTEMS bushel of cereal. For example, for a corn
distillery with an alcohol yield of 2.5 gallons of
In considering all the different processes that absolute alcohol per bushel, the 25 gallons of
make up cooking, it should first be explained
16 D.R. Kelsall and T.P. Lyons
Grain
Water
Backset
α-amylase
(High T L120™)
Cooling coils
To fermentors
30°- 32°C
Heat exchanger
liquid would contain 2.5 gallons of alcohol. The pH range for efficient α−amylase usage
Therefore it would contain 10% alcohol by is 6.0-6.5, although enzymes with good activity
volume (abv). Using the distillery alcohol yield, at pH 5.5-5.7 are now available. Therefore, mash
the distiller can determine the quantity of cereals pH should be controlled in this range from the
and liquid to use. Most distilleries traditionally first enzyme addition until the end of
operated with beers in the 10-15% abv range, liquefaction. The glucoamylase enzyme has a
although some beverage plants run at alcohol lower pH range (4.0-5.5), so after liquefaction
levels as low as 8%. More recently fuel alcohol the pH of the mash should be adjusted with either
plants run at alcohol levels as high as 19-20% sulfuric acid or backset stillage, or a combination
by volume, the average being nearer to16-17% of the two.
by volume. The quantity of backset stillage as a percentage
In the batch cooking system, a small quantity of the total liquid varies from 10 to 50%. On
of α−amylase is added at the beginning (0.02% one hand, the backset stillage supplies nutrients
w/w of cereal) to facilitate agitation in the high essential for yeast growth. However too much
viscosity stage at gelatinization. After boiling, backset stillage can result in the oversupply of
usually for 30-60 minutes, (sometimes under a certain minerals and ions that suppress good
slight pressure), the mash is cooled to 75-90°C fermentation. Especially noteworthy are the
and the second addition of α−amylase made sodium, lactate and acetate ions. Sodium
(0.04-0.06% w/w cereal). Liquefaction then concentrations above 0.03% or lactate above
takes place, usually over a holding period of 45- 0.8% or acetate above 0.03% can inhibit yeast
90 minutes. The mash should always be growth and can slow or possibly stop the
checked at this stage to make certain that no fermentation prematurely. Overuse of backset
starch remains. Starch produces a blue or purple (or even process condensate water) must be
color with iodine. Mash should not be transferred avoided to prevent serious fermentation
from the liquefaction hold until it is ‘starch- problems.
negative’.
Grain dry milling and cooking procedures 17
CONTINUOUS COOKING SYSTEMS Modern systems have 29-33% solids. From the
liquefaction chamber, the mash is pumped
Few distilleries outside of beverage plants use through a heat exchanger to fermentation.
batch cooking. Most fuel ethanol distilleries use
a continuous cooking system. In the continuous
cooking process (Figure 9) meal, water and CONTINUOUS U-TUBE COOKING SYSTEM
backset stillage are continuously fed into a
premix tank. The mash is pumped continuously The continuous U-tube cooking system (Figure
through a jet cooker, where the temperature is 10) differs from the columnar cooking system
instantly raised to 120°C. It then passes into the in that the jet cooker heats the mash to 120-
top of a vertical column. With plug flow, the 140°C prior to being transferred through a
mash moves down the column in about 20 continuous U-tube. The retention time in the U-
minutes and passes into the flash chamber for tube is only three minutes, after which it is
liquefaction at 80-90°C. High temperature- flashed into the liquefaction vessel at 80-90°C
tolerant α−amylase is added at 0.04-0.08% w/w and the enzyme is added (high temperature-
cereal to bring about liquefaction. The retention tolerant α-amylase 0.05-0.08% w/w cereal). The
time in the liquefaction/flash chamber is a residence time in the liquefaction vessel is a
minimum of 30 minutes, but should be at least minimum of 30 minutes.
60 minutes. The pH from slurrying through to The main advantage of this system is the
the liquefaction vessel must be controlled within relatively short residence period in the U-tube.
the 6.0-6.5 range. The greatest advantage of this If properly designed there is no need to add any
system is that no enzyme is needed in the α−amylase enzyme in the slurrying stage.
slurrying stage, leading to significant savings in However, because of the relatively narrow
enzyme usage. It is critical to have plug flow diameter of the tubes, some distillers add a small
through the chamber along with good enzyme amount of enzyme to the slurry tank to guarantee
dispersion. The mash should have a relatively a free flow.
low viscosity and dextrose level of 2-3%.
α-amylase
(High T L120™)
Grain Steam Vacuum
0.05 - 0.8 w/w
ejector
Water
Pressure relief
Backset
(up to 40%)
Glucoamylase + yeast
(Allcoholase II L400™ +
Column
Rhizozyme™+ Thermosacc™)
20 min
120°C
To fermentor
30-32°C
Heat exchanger
Vacuum
Grain ejector
Water
Steam
To fermentor
Retention 'U' tube 30-32°C
3 min retention
140°C Heat exchanger
Figure 10. High temperature, short time, continuous U-tube cooking system.
100
Holding time
80
60
40
20
0
0 25 50 75 100 125
Time (minutes)
50
18
Corn
Viscosity in poises
14
10 25
Figure 14. Activity of high temperature-tolerant α-amylase Table 5. Advantages of Rhizozyme™ over conventional
in relation to temperature. glucoamylase.
Conventional Rhizozyme™
The reaction time for enzyme-catalyzed glucoamylase
reactions is directly proportional to the
Temperature optimum, °C 60+ 30+
concentration of enzyme. Consequently, distillers pH optimum 4.0-5.5 3.5-5.0
wishing to minimize the quantity of enzyme Amylase activity,
used should design equipment to have long SKB units None 50,000
residence times to allow the reactions to be Cellulase activity
completed with minimal dosage of enzyme. CMC-ase units None 2500
Amylopectinase, AP units None 5000
chromatography systems (HPLC) that can demonstrates the diversity of approach possible,
measure sugars directly. Recent experience, yet points out many similarities (Table 6). It
however, shows that DE, provided it is above would appear that there are three schools of
10, is of no concern. In focusing on 23% ethanol, thought:
the key in any cooking and liquefaction process
is to liquefy, i.e., lower viscosity, so that mash a) A long liquefaction period (1-2 hrs at 90+ °C)
can be pumped through the heat exchanger to after the high temperature jet cooker and low
the fermentor (for SSF) or to the saccharification temperature slurry (no enzymes).
tank.
The functional characteristics of liquid b) A short high temperature slurry prior to a
glucoamylase prepared from the microorganism long liquefaction period (1-2 hrs at 90+ °C)
Aspergillus niger, can be seen in Figures 15 and after the high temperature jet cooker.
16. Two parameters, temperature and pH, dictate Enzymes used in slurry.
how enzymes can be used. While liquefaction c) Choice of saccharification stage or SSF.
is carried out at a pH of 6.0-6.5 and a
temperature of 90°C, this is not at all acceptable From the experience of the authors,
for saccharification. The pH must be in the 4-5 recommendations are for a short slurry at 140-
range for saccharification; and the optimum 150°F followed by jet cooking (10-15 min.) and
temperature for the glucoamylase activity is a long (1.5 hrs) liquefaction step. The mash
60°C. The mash, therefore, must be acidified should have a sugar profile similar to Table 7. It
with either sulfuric acid or backset stillage or should then be cooled to fermentation
both before addition of the glucoamylase. temperature on the way to the fermentor. This
Temperature must also be adjusted. As mentioned profile will ensure a rapid start by yeast with no
previously, normal mash saccharification sugar overload. In the presence of Rhizozyme™
temperature is 60-65°C; although for or glucoamylase, dextrins will continue to be
microbiological reasons 70-75°C would be ‘spoon-fed’ to the yeast during fermentation.
preferable. Lactobacillus can survive at 60°C; and Enzyme additions should include all the amylase
frequent infection of saccharification systems has during liquefaction and Rhizozyme™ during
caused many distillers to change to fermentation.
saccharifying in the fermentor.
Table 7. The target sugar profile following liquefaction.
Chapter 3
RONAN F. POWER
North American Biosciences Center, Alltech Inc., Nicholasville, Kentucky, USA
Introduction
Starch is the second most abundant carbohydrate Embden-Meyerhof-Parnas (glycolytic) pathway
in the plant world after cellulose. It is the (Figure 1). In this pathway, glucose is
principal plant storage polysaccharide and is phosphorylated, then split into two
similar in structure and function to glycogen, phosphorylated 3-carbon derivatives,
which is the main storage polysaccharide of the glyceraldehyde-3-phosphate and dihydroxy-
animal kingdom. Both starch and cellulose are acetone phosphate. Both products are
glucose-based polymers that differ in the subsequently converted into pyruvic acid, which
orientation of the linkages between the glucose under anaerobic conditions yields ethanol and
subunits in the respective macromolecules. carbon dioxide. It is obvious from Figure 1 that
Saccharomyces cerevisiae is the primary the yeast, S. cerevisiae, is very well equipped to
microorganism involved in the transformation ferment glucose and fructose. This is why wine
of starch to ethyl alcohol, and corn starch fermentations are technically very simple. Grape
remains the major raw material for industrial pressings deliver pure glucose and, of course,
alcohol fermentation although potato starch also the required yeast are already present on the
has limited use, particularly in Europe. For surface of the grapes. Likewise, molasses
example, in the 1999 US crop year, 526 million represents a readily assimilated feedstock for
bushels of corn (14.7 million tons at 15% yeast since the predominant sugar in molasses
moisture) were used in the fuel ethanol industry is the disaccharide, sucrose. Yeast contains the
(McAloon et al., 2000). enzyme sucrase, which cleaves sucrose into its
The starting compound for alcohol synthesis constituent monosaccharides, glucose and
by yeast is glucose (dextrose), which is a six- fructose.
carbon sugar. Once fermented, each molecule Starch, however, cannot be fermented directly
of glucose yields two molecules of ethanol and by S. cerevisiae, because the organism lacks the
carbon dioxide, respectively. requisite starch-degrading or amylolytic
enzymes to liberate glucose from this storage
C6H12O6 → 2C2H5OH + 2CO2 polysaccharide. Whether the starch is derived
from corn, potatoes, sorghum (milo), barley or
The sequence of biosynthetic reactions by which other cereal grains, the bonds between the
yeast converts glucose to ethanol is termed the glucose subunits in the starch chain must first
24 R.F. Power
GLUCOSE
ATP
Glucokinase
ADP
Glucose-6-phosphate
Phosphohexose
isomerase
Fructose-6-P
ATP
Phosphofructokinase
ADP
Fructose 1,6
bisphosphate
Aldolase
Glycolysis
Dihydroxyacetone Glyceraldehyde
phosphate Triosephosphate 3-phosphate
isomerase
NAD+
Glyceraldehyde-3-P
dehydrogenase
NADH
1,3 bisphosphoglycerate
ADP
Phosphoglycerate
kinase
ATP
3-phosphoglycerate
Phosphoglycerate
mutase
2-phosphoglycerate
Enolase
H2O
Phosphoenolpyruvate
ADP
Pyruvate
kinase
ATP
Pyruvate
be hydrolyzed to liberate free glucose molecules also be written (CH 2O) 6. There are four major
which yeast can utilize. Brewers achieve this by classes of carbohydrate that are of general
exploiting the endogenous starch-digesting interest: monosaccharides, disaccharides,
enzymes produced by seeds when they oligosaccharides and polysaccharides.
germinate. Grains, such as barley, destined for Monosaccharides, or simple sugars, consist of a
fermentation, are first sprouted to initiate enzyme single polyhydroxy aldehyde or ketone unit. The
synthesis within the seeds. This process is most abundant monosaccharide in nature is, of
termed malting. The malted grains are then course, the 6-carbon sugar D-glucose.
steeped in hot water to allow the enzymes to Disaccharides encompass sugars such as
convert starch to fermentable sugar. This sucrose, which is an important transport
mashing phase is then followed by a lautering carbohydrate in plants. Lactose, a disaccharide
or mash filtration stage to separate the solids of glucose and galactose, is commonly called
from the glucose-rich liquid ‘wort’ to which yeast milk sugar. Oligosaccharides (Greek oligos,
is added. ‘few’) are made up of short chains of
The first step in the industrial production of monosaccharide units joined together by
alcohol from starch also requires that the starch covalent bonds. Most oligosaccharides in nature
is saccharified; and this is generally recognized do not occur as free entities but are joined as
to consist of two phases: a) the dextrinization or side chains to polypeptides in glycoproteins and
liquefaction phase in which starch is partially proteoglycans. Polysaccharides consist of long
degraded to soluble dextrins, and b) the true chains of hundreds or even thousands of linked
saccharification or ‘conversion’ phase in which monosaccharide units and, as will be discussed,
the dextrins are hydrolyzed to fermentable exist in either linear or branched chain form.
sugars. The two phases of saccharification may Knowledge of the terminology associated with
be carried out by either acid or enzymatic the organizational structure of monosaccharides,
hydrolysis. In the past, acid combined with high such as D-glucose, is fundamental to an
temperatures and pressures was used to understanding of how polysaccharides differ and
accomplish the same reaction that enzymes how polysaccharide-degrading enzymes, such
perform today under relatively mild conditions. as amylases, function.
Acid hydrolysis was extremely hazardous, Monosaccharide structures are often drawn as
required corrosion-resistant vessels and resulted open Fischer projection formulae in which the
in a product with a very high salt content, due sugar is depicted in either its D- or L-
to neutralization. In addition, dextrose yield was configuration. Most naturally occurring sugars
limited to approximately 85% and off-colors and are D-isomers, but D- and L-sugars are mirror
flavors were often a result of the acid hydrolysis images of each other. The D or L designation
process. The change to enzymatic hydrolysis in refers to whether the asymmetric carbon farthest
the 1960s eliminated all the drawbacks of the from the aldehyde or keto group points to the
acid hydrolysis process and increased the yield right or the left as shown for D-glucose and L-
of dextrose to between 95 and 97%. This chapter glucose (Figure 2).
will focus on the physical and enzymatic steps The ring or cyclic forms of monosaccharides
required to convert starch to sugars that are are most commonly drawn as Haworth projection
utilizable by yeast. formulae, as shown for glucose in Figure 3.
Although the edge of the ring nearest the reader
is usually represented by bold lines, the six-
Carbohydrate chemistry: the basics membered pyranose ring is not planar, as
Haworth projections suggest. More importantly,
When most people think of carbohydrates, they however, the numbering system for the carbon
think of sugar or starch. Carbohydrates are thus atoms in the glucose structure should be
called because they are essentially hydrates of carefully noted. Carbon atom number one, also
carbon. They are composed of carbon and water known as the anomeric carbon atom, is the first
and most may be represented by the general carbon encountered when going in a clockwise
formula (CH 2O) n. For example, the empirical direction from the oxygen atom in the ring
formula for D-glucose is C 6H 12O 6, which can structure, and the other carbon atoms, C2 to C6,
26 R.F. Power
are numbered as indicated in Figure 3. When the anomeric carbon is below the apparent plane
the hydroxyl or OH group at the anomeric of the glucose subunit, the glucosidic linkage
carbon atom extends below the ring structure of formed is stated to be in the α-configuration and
D-glucose, the sugar is stated to be in the α the bond is termed α(1→4). Amylose chains
(alpha) configuration. When the hydroxyl group typically contain between 500 and 2,000 glucose
at C1 extends above the ring, the sugar is in the subunits (Lehninger, 1982).
ß (beta)-configuration. These seemingly minor In amylopectin, glucose subunits are also
structural changes represent the major difference linked in α(1→4) configuration in the linear
between key natural polysaccharides such as parts of the polysaccharide but branch points
starch and cellulose. occur approximately once every 25 glucose units
where the terminal glucose in a parallel amylose
chain is involved in an α(1→6) linkage. These
H O H O
1
C
1
C
branch points are resistant to hydrolysis by many
amylases. Some amylases are totally incapable
H
2
C OH HO
2
C H of hydrolyzing the α(1→6) bonds while others
hydrolyze the 1→6 bonds more slowly than the
3 3
HO C H H C OH 1→4 bonds.
Glycogen is a very closely related structure to
4 4
H C OH HO C H amylopectin and serves as the glucose storage
5 5
polysaccharide in animals. The main difference
H C OH HO C H is that the branch points in glycogen occur more
6 6
frequently, approximately once every 12 glucose
CH2OH CH2OH units. Cellulose is a linear, unbranched
homopolysaccharide of 10,000 or more D-
D-glucose L-glucose
glucose units linked by 1→4 glucosidic bonds.
It might be expected, therefore, to closely
Figure 2. Fischer projection formulae for D- and resemble amylose and the linear chains of
L- glucose. glycogen, but there is a critical difference: in
cellulose the 1→4 linkages are in the ß-
STARCH configuration (Figure 5). This apparently trivial
Native starch is a large polymer made up of D- difference in the structures of starch and cellulose
glucose molecules linked together to form a results in polysaccharides having widely
linear polymer called amylose or a branched different properties. Because of their ß-linkages,
polymer called amylopectin (Figure 4). the D-glucose chains in cellulose adopt an
In amylose, the glucose subunits are linked extended conformation and undergo extensive
together by bonds between the anomeric carbon side by side aggregation into extremely
of one subunit and carbon number 4 of the insoluble fibrils. Since no enzyme capable of
neighboring subunit. Because the OH group at hydrolyzing cellulose is secreted by the intestinal
tract of vertebrates, cellulose cannot easily be
6 6
CH2OH CH2OH
5 5
O H O OH
H H
4 1 4 1
OH H OH H
2
OH 3 OH OH 3 2 H
H OH H OH
α-D-glucose β-D-glucose
O O O O
H H H H H H H H
OH H OH H OH H OH H
O O O O
OH
H OH H OH H OH H OH
CH
H 2 OH
O O
OH
H CH
H H 2 OH
H
O O
OH OH
Amylopectin H
H
H
O α-(1-6) linkage
OH
O O O O H
H H H H H H H
OH H OH H OH H OH H
O O O O O
H OH H OH H OH H OH
Figure 4. Structure of amylose and amylopectin; the polysaccharides that comprise starch.
H
CH2OH OH CH2OH H OH CH2OH
O H O O
H O H H O
H OH H O OH H H
O OH H H O OH H H O OH H
H H O H H O H
H OH CH2OH H OH CH2OH H OH
digested and its D-glucose units remain source and has an impact on its gelatinization
unavailable as food to most non-ruminant properties (see below). For example, four classes
animals. of corn starch exist. Common corn starch
In contrast, the α-linkage of starch allows the contains approximately 25% amylose, while
molecule to be more flexible, taking on a helical waxy maize is almost totally comprised of
configuration in solution which is stabilized by amylopectin. The two remaining corn starches
hydrogen bonding. Furthermore, the are high-amylose varieties with amylose contents
bottlebrush-like structure of amylopectin lends ranging from 50 to 75%. Other typical amylose
itself to less densely packed associations contents are 20% for potato, 17% for tapioca and
between adjacent starch molecules and further 25% for wheat. Common rice starch has an
enhances water solubility and access by starch- amylose:amylopectin ratio of approximately
degrading enzymes. The ratio of amylose to 20:80 while waxy rice starch contains only about
amylopectin is a characteristic of each starch 2% amylose (Table 1).
28 R.F. Power
Table 1. Ratio of amylose to amylopectin (percent of that amylose molecules, because of their linear
total starch) in various starch sources. structure, will lie in closer proximity to each
other than the brush-like amylopectin molecules.
Starch source Amylose (%) Amylopectin (%) More extensive inter-molecular hydrogen
Common corn 25 75
bonding will occur between the closely bunched
Waxy maize 1 99 amylose polymers in high-amylose content
High amylose corn 50-75 25-50 starch sources than will occur in high-
Potato 20 80 amylopectin sources. Consequently, it requires
Rice 20 80 more energy to break the hydrogen bonds in
Waxy rice 2 98 high-amylose starch sources, which results in
Tapioca/cassava/manioc 17 83 higher gelatinization temperatures.
Wheat 25 75
Sorghum 25 75 Table 2. Typical temperature ranges for the
Waxy sorghum <1 >99 gelatinization of various cereal starches.
Heterowaxy sorghum <20 >80
Cereal starch source Gelatinization range (oC)
STARCH
amylose + amylopectin
Slurry
GELATINIZATION
Cooking
Heat and moisture solubilize starch
AMYLOSE
AMYLOPECTIN
α-amylase
α-amylase
α-amylase
LIQUEFACTION α-amylase
Production of dextrins endoenzyme
and α-limit dextrins hydrolyzes random α(1-4) bonds
by α-amylase
α-LIMIT DEXTRINS
DEXTRINS
Glucoamylase Pullulanase
(debranching)
Glucoamylase
α(1-4), α(1-6 ) bonds
SACCHARIFICATION
Hydrolysis of dextrins to
fermentable sugars
catalyzes the removal of maltose units from ends of dextrins causing the release of ß-D-
starch, it produces what is known as a Walden glucose. In other words, it is an exo-acting
inversion of the OH group at C-1, giving rise to enzyme that cleaves single molecules of glucose
ß-maltose (Belitz and Grosch, 1999). Thus, the in a stepwise manner from one end of the starch
products of bacterial and fungal α-amylases are molecule or dextrin. Reference has also been
all in the α-configuration, while the products of made to the fact that glucoamylases can
ß-amylases exist in the ß-configuration. hydrolyze α(1→6) bonds to a certain extent.
As mentioned, the true saccharification or However, these branches are cleaved at a rate
conversion of dextrins to glucose is conducted, approximately 20 to 30 times slower than the
in the vast majority of cases, by glucoamylase, cleavage of α(1→4) bonds by the enzyme. Extra
although other enzymes, including some α- glucoamylase can be added to compensate for
amylases, have significant saccharifying activity this slower rate but this can cause undesirable
in their own right. Glucoamylase is also known side reactions to take place, whereby glucose
as amyloglucosidase (glucan 1,4 –α-glucosidase; molecules repolymerise in a reaction termed
EC No. 3.2.1.3) and its main activity is the reversion, forming isomaltose and causing a
hydrolysis of terminal 1,4-linked α-D-glucose decrease in glucose yield.
residues successively from the non-reducing Glucoamylases are isolated from fungal
Enzymatic conversion of starch to fermentable sugars 31
sources such as Aspergillus niger and Rhizopus yield by liberating fermentable sugars from non-
sp. Fungal enzymes, by nature, are less starch polysaccharide sources. For example,
thermotolerant than their bacterial counterparts multi-enzyme complexes isolated from fungal
and therefore the temperature maxima for cultures grown on fiber-rich, semi-solid or
glucoamylases tend to be in the region of 60oC surface culture systems, contain a mixture of
while their pH optima normally lie between pH amylolytic, cellulolytic and other
4.0 and 4.5. These conditions of temperature and polysaccharidase activities which have
pH can cause significant problems in plants temperature and pH optima that quite closely
using dedicated saccharification tanks because match those found in distillery fermentors. One
they are quite conducive to the growth of certain commercial preparation, Rhizozyme™, is a
bacterial contaminants. The claimed benefit of surface culture enzyme gaining increasing use
pre-saccharification of mash prior to in the fuel ethanol industry. These enzymes
fermentation is that a high level of glucose is satisfy the requirement to saccharify dextrins to
made available at the start of fermentation. glucose but have the additional activities
However, in addition to infection problems, such required to attack, for example, ß-linked glucose
an approach can lead to the generation of non- polymers, if present.
fermentable reversion products, such as
isomaltose.
Because of the drawbacks of mash pre- Summary
saccharification, a popular approach is to
saccharify simultaneously with fermentation The objective of industrial alcohol production
(SSF or Simultaneous Saccharification and is to produce the highest ethanol yield possible.
Fermentation). The requirement for high glucose This requires the greatest attainable conversion
concentrations at the start of fermentation is met of starch to fermentable sugars because yeast
by the addition of glucoamylase during the cannot use starch in its native state. The use of
filling of the fermentor. Although the concern starch-degrading enzymes represented one of the
with such a system is that glucose may become first large-scale industrial applications of
limiting during the fermentation, this is not a microbial enzymes. Traditionally, two main
problem in practice and is merely a case of enzymes catalyze the conversion of starch to
metering in the correct amount of enzyme. glucose. In the first stage of starch conversion,
Glucose is used up as it is produced and thus α-amylase randomly cleaves the large α-1,4-
there is little risk of reversion reactions occurring. linked glucose polymers into shorter
An additional attraction to this approach is that oligosaccharides at a high reaction temperature.
it lends itself to the use of somewhat non- This phase is called liquefaction and is carried
conventional enzymes, which can boost ethanol out by bacterial enzymes. In the next stage,
References
Belitz, H.-D. and W. Grosch. 1999. Food
Chemistry. (M.M. Burghagen, D. Hadziyev, P.
Grain handling: a critical aspect of distillery operation 33
Chapter 4
DAVID J. RADZANOWSKI
Alltech Inc., Nicholasville, Kentucky, USA
Introduction
Until recently, sanitation in grain handling has and the warrants necessary for the inspection,
been basically ignored by the distilling industry, surprised the staff by asking only for a sample
except for those engaged in beverage alcohol of brewer’s spent grains. The staff relaxed and
production. Now that we have begun to discuss gladly obtained a sample of spent grains under
and accept the ‘Biorefinery’ concept, we are the inspectors’ directions.
forced to realize that a good sanitation program Two weeks later the plant was upset with a
and procedures are essential to the bottom line. notice of citation for improper sanitation
Outbreaks of food animal diseases such as BSE procedures since insect fragments, mold,
and food-borne diseases such salmonella, as well mycotoxins and insecticide residuals were found
as prohibitions against mycotoxins and in those spent grains. Since animal feed laws
insecticides in grains going into food animals were less stringent in the 1970s than today, the
and ultimately the human food chain, have plant was allowed to continue selling the spent
grabbed our attention. We suddenly realize that grains. The agency just used the contamination
distillers dried grains with solubles (DDGS) is a detected as a means of bringing attention to an
valuable co-product and not just an unavoidable inadequate sanitation program as established for
nuisance. Also, if we cannot sell it because of breweries under the Federal Food, Drug and
contamination, we are out of business. Cosmetic Act.
We first became aware of the importance of a
sanitation program in grain handling systems
some years ago in the brewing industry. Good manufacturing practices
Breweries by definition prepare a food product,
beer, and are subject to all possible scrutiny by Since 1969, the Act has contained guidelines
the Food and Drug Administration (FDA) and called the Good Manufacturing Practices or the
related state and local food safety bureaus. GMPs (FDA, 2000). This Amendment to the Act
Breweries routinely underwent very involved is titled ‘Human Foods, Current Good
inspections for food contamination and looked Manufacturing Practice (Sanitation) in
forward to them with much dread. In the 1970s, Manufacture, Processing, Packing or Holding’.
a group of federal inspectors appeared at one Sections 3 through 8 of the amendment apply
plant and after providing the proper credentials in determining whether the facilities, methods,
34 D.J. Radzanowski
practices and controls used in the manufacture, (BATF), a regulatory agency that imposes and
processing, packing or holding of food are in collects tax monies on the ethanol generated by
conformance with and are operated or a distiller. Recent changes in the US with the
administered in conformity with good creation of the Homeland Security Act have split
manufacturing practices to produce under up the BATF and reduced the strength of the
sanitary conditions, food for human organization pertaining to alcohol producer
consumption. enforcement. This leaves a gap that the FDA
Knowledge of the GMPs, with an eye toward seems to be filling, slowly but forcefully. What
meeting the outlined standards, is a valuable does this mean to the distiller? In all probability,
guide toward setting up an operation that stricter requirements demanding reduced
facilitates the proper handling of grain. Section contamination in co-products in animal feeds
3 of the amendment provides guidelines for the such as DDGS will be established. We have
correct positioning of the plant and the care of already seen some activity from other countries
the grounds. It goes further to detail the rejecting DDGS generated in the US due to the
appropriate construction of the facility utilizing content of mycotoxins and other unwanted
cleanable surfaces, adequate lighting, ventilation, contamination. How can we help prevent
employee facilities and screening and protection rejection or seizure of this valuable by-product?
from pests. Section 4 contains the requirements Adopting strict sanitation programs is an aid to
for the equipment and utensils. All must be designing or remodeling plants to conform to
suitable for the intended use, durable, cleanable the guidelines needed.
and kept in good repair.
Sections 5 and 6 are closely related. They deal
with the needs for adequate water, plumbing and Where do we start?
sewage disposal, toilet and hand washing
facilities for employees, and the proper disposal William Schoenherr, formerly of the Lauhoff
of offal and rubbish. The requirements for Grain Company, which is now part of Bunge
general maintenance, pest control and sanitizing Milling is considered the father of sanitation in
of equipment and utensils are described in detail. grain handling and storage facilities as well as
To ensure that food will not be contaminated, in breweries. Schoenherr started as an FDA
Section 7 requires the inspection of all raw inspector and brought his enormous expertise
materials and ingredients as well as the inspection to the grain industry. Lauhoff recognized that it
of the containers and the common carriers had obtained a great asset and allowed Bill to
delivering the goods. This section also requires share his expertise with the grain handling world
frequent cleaning and sanitization, processing and particularly their customers. Schoenherr
under conditions to minimize bacterial and always talked about a number of key factors that
micro-organic growth, chemical and biological must be considered in a good sanitation program.
testing and the maintenance and retention of We will use these as guidelines for the following
records. discussion.
Under Section 8, plant management is required
to assure that employees are free from disease,
observe good cleanliness practices in clothing FORMS OF CONTAMINANTS, DETECTION AND
as well as personal habits and that they are REMOVAL
properly trained and educated to perform their
duties in a safe and sanitary manner. First, it is important to be aware of the different
forms of contamination that may be present in
the raw materials brought in for processing.
STRICTER REQUIREMENTS LIKELY Second, everything possible must be done to
ensure detection of such contamination before
Distillers in the United States, as well as in other it enters the facility. And, third, everything
countries, do not normally come under the possible must be done to avoid, remove or
jurisdiction of the FDA or related organizations. destroy the contaminants before using these raw
They are normally governed by bureaus like the materials. These three are very closely related.
US Bureau of Alcohol, Tobacco and Firearms What kind of contaminants can we expect?
Grain handling: a critical aspect of distillery operation 35
Among others, insects, rodents and birds and farmers supplying corn to some of the newest
their droppings, stone, cobs, weed seeds, glass, fuel ethanol plants coming on stream was very
wood, rags, tramp metal, residual insecticides revealing concerning the state of the corn market
and fumigants, water, bacteria, mold and in some areas of the country. The farmers were
mycotoxins. Some of these are from unclean quite happy that the new distillers were paying
grain handling and transport. Others have been the same price for No. 2, 3, 4 and 5 grain as
added as fillers, adding weight without they are for No. 1. This seems not only wasteful
producing usable substrate. Some, like stone and on the part of the distiller, but potentially
tramp metal, are dangerous and may cause fires dangerous. The more allowable extraneous
and explosions in the grain handling system. The content in the lower grade incoming grain, the
more difficult contaminants to detect such as greater the contribution to lower yields and the
residual insecticides and fumigants as well as higher likelihood of gross contamination and
mycotoxins, can be present at levels that can possible ignition of fires and explosions.
result in unmarketable DDGS. High moisture is the most common reason for
Most operations are located close to the source rejection of a shipment. This is a good start. High
of the substrate used for fermentation, so most moisture contributes to difficulties in milling and
of the material is received in trucks with only a handling and to lower yields. High moisture also
limited amount of rail transit. Samples are contributes to elevated mold growth, which
generally obtained by probing each truck in two typically leads to contamination with mycotoxins
places, front and back, and then composited. that become concentrated in the DDGS.
Statistically, this is not sufficient. Two pounds First, decide on specifications for acceptance
of sample from a 50,000-pound load is not very of the incoming substrate. Next, provide the
indicative of the condition of the contents of the people and the testing equipment to determine
truck. if the substrate meets those specifications. Stick
Bushel weight, moisture, damaged kernel and to the established specifications. Only when
broken kernel/foreign material tests are run satisfied should you begin to unload the material
immediately. High insect infestation may be into the plant’s infrastructure.
determined by visual inspection, but low
infestations may be more difficult to find. Also,
the true grain weevils, because they bore into RECEIPT AND STORAGE
kernels, may not be visible on the surface of the
grain. Only a few plants use a black light to Specifications should call for all shipments to
detect aflatoxin presence. Black light is also be received in covered hopper railroad cars and
usable for the detection of rodent urine hopper trucks fully covered by tarpaulin or some
contamination. This should be a routine test for other means. This will help prevent appearance
all shipments in all plants. NIR technology offers of rodents and other pests while the shipment is
a solution to many of the grain testing needs for in transit.
quick decision making by distillers. That subject Once the shipment is accepted, it can then
is covered in a separate chapter in this volume. proceed to the unloading area. Most people drop
Do not neglect the senses. Examine the grain, the contents of the railcar or truck into a pit from
checking for insect presence, rodent droppings, which the material is conveyed in some manner
bird feathers as well as foreign seeds and to the storage bin. It is important that this station
extraneous matter such as cobs, straw, wood and be constructed in a manner that allows it to be
metal. Smell the material for any hint of mold kept in good sanitary condition without much
presence or chemical contamination. Feel the effort. Pay attention to the grid covering the pit.
grain for moisture and slime presence. Some grids have wide-spaced bars that could
Most plants use the specifications for No. 2 allow whole rats to pass through to the pit and
corn as a guideline, which call for a bushel onward to the bin. This is unacceptable. The grid
weight of 56 pounds, a limit of 0.2% heat system should be designed for the rapid flow of
damaged kernels, 5% total damaged kernels and the grain and nothing else.
3% total broken kernel/foreign material There are many different styles of grain
presence. A recent discussion with a group of conveying systems, including bucket elevators,
36 D.J. Radzanowski
positive and negative pneumatic systems, belt breweries since the brewers are not really
conveyors and the newer pulse pneumatic receiving raw materials, but barley that has been
systems. The most user unfriendly and cleaned and processed by malting. For brewers
dangerous system seems to be the bucket such a system works well. The earlier fuel
elevator. This is due to the high number of parts ethanol plants had systems like this installed
with a greater maintenance requirement and the because the engineers had been designing for
higher number of possible ignition points. breweries for a number of years and did not
Manufacturers have attempted to alleviate some recognize that distillers are involved in handling
of this by the introduction of plastic buckets, a different type of material. Distillers have
but the hangers and chains are still, for the most become aware of this and are requesting that
part, made of metal. Bucket elevators also cleaning systems be placed ahead of the bins.
provide more harborages for insects and are Keep records of the amount of debris removed,
harder to clean and fumigate. Pneumatic systems as you do not want to pay for trash. Keep a record
tend to be abraded by the impingement of the of the net amount of clean substrate entering the
grain on the tubing system, but a well-designed storage bin. This figure will be helpful later in
and balanced system can extend the life determining true yield.
expectancy of the tubes indefinitely. Bins should be constructed of concrete or metal
Grain should ideally be cleaned on the way to and be well sealed. We have seen bins
the bin. Why store dust, straw, cobs, wood, metal, constructed of corrugated metal sheets that
dead rats and other extraneous material? leaked so badly that the grain was beginning to
Unfortunately, most of the systems we see are ferment in the storage bin after a very light
similar to Figure 1. rainfall. One of the most important design criteria
The facility illustrated in Figure 1 is typical in for grain bins is their ability to be fully emptied.
Receiver Receiver
Exhaust Exhaust
Airlock Airlock
Vent
hopper
Cleaner
Rotary feeder
Surge
Blower hopper
We have seen too many storage bins at every two weeks and inspections of the more
distilleries that hold 3-5 days of material that critical zones should take place on a weekly
cannot be totally emptied. This can be very basis. Any evidence of an unwanted invasion
dangerous and conducive to infestation. Mold should be dealt with immediately. The cleaning
growing in stationary grain simply contaminates schedule should also be devised with the insect
successive loads. life cycle in mind. Certain areas of the plant, like
One of the great tools for protection against the receiving station and subsequent conveyors,
insect infestation is the understanding of the life dust collectors and air conveyor systems require
cycles of the various insect pests. Interrupting more frequent inspection, cleaning and possibly
that life cycle prevents an infestation from fumigation on a very regular basis. The program
becoming established. A 3-5 day storage limit should determine the risk and set the appropriate
of grain in a bin will interrupt the life cycle of schedule.
the common grain insects, but the bin must be A pest control program can be dangerous in
totally emptied each time it is filled. If the design the hands of unqualified people. Bait stations,
does not allow total emptying, there will be a insecticides, and fumigants are poisons and
continuous infestation. harmful to humans, animals and pets as well as
the pests they are intended to control. These
compounds are handled and applied by licensed
STORAGE CAPACITY individuals, however it is necessary for the
person in the company in charge of the sanitation
How large should storage capacity be? Financial program to have training and knowledge of pest
and logistic experts must contribute to this control operations in order to evaluate the work
decision. Storage capacity must accommodate of the hired contractors.
both the processing schedule and cash flow With the direction and agreement of
consideration. If running 24-hrs a day, seven management, establish the program. Determine
days a week and 50 weeks of the year, then the inspection schedule. Hire the pest control
either storage capacity or local sources of firm. Determine the schedule for application of
substrate must keep the plant supplied. Local the tools and chemicals of the pest control
supplies reduce storage needs, so that capacity operation. Determine the cleaning schedule.
may be needed only to maintain supplies through Keep complete records. Record keeping cannot
weekends and holidays. If shipments must come be stressed too strongly. Obtain all reports from
from afar, more capacity is needed. This is where the contract pest control operation including the
the logistics advice is needed. dates of application, the places of application,
the chemicals used in the application, the strength
of the materials used, the technician performing
The sanitation program the service and the records and results of their
inspections as well as the results of the service.
Sanitation programs must be adjusted to the type Review the work of the hired contractor for
of facility. Very few large operations today are effectiveness. Review the program and make
contained within a structure. Most are of an open appropriate changes as they are necessary.
air design. The enclosed plant has the sanitation
advantage of controlled entry and egress,
making it easier to exclude rodents and birds. MILLING
Regardless of design, full inspection and
cleaning techniques must not be relaxed. Spills When grain leaves the bin on the way to the mill,
must be cleaned up quickly to prevent attracting it should pass through de-stoners to protect the
pests. Scrape, shovel and vacuum the debris. mills from damage and magnets to prevent
Never use an air hose since the subsequent sparks and explosions. Dry milling operations
dispersal just spreads infestation and infection. should also incorporate explosion dampers in the
Any inspection and cleaning program should system. This brings us to the question of what
be familiar with storage insect life cycles and type of a mill to use.
take advantage of this knowledge. Full Brewers and Scotch or Irish whisky distillers
inspections of the total operation should be made can work very well with a roller mill, preferably
38 D.J. Radzanowski
Table 1. Recommended malt sieve analysis for lauter tun and mash filter wort separation systems.
Fraction US sieve number (inches) For lauter tun (% retained) For mash filter (% retained)
Husks I 10 (0.0787) 14 6
Husks II 14 (0.0555) 20 10
Coarse grits I 18 (0.0394) 29 6
Coarse grits II 30 (0.0234) 18 18
Fine grits 60 (0.0098) 9 43
Coarse flour 100 (0.0059) 4 10
Powder Bottom 6 7
with five or six rolls. The roller mills work A fine grind exposes more surface area,
extremely well with the soft, already modified facilitating better slurry, cook and liquefying
and malted barley, which is the principal operations. The more surface area exposed, the
substrate of these operations. Simple adjustments greater the effectiveness of the enzymes
can be made in the grind according to the needs allowing for greater activity and more complete
of the wort separation systems in use, whether reactions. An idea of the importance of particle
they incorporate mash filters or lauter tuns. size in determining surface area exposure can
Table 1 contains the recommended malt sieve be obtained by reviewing Table 3, which uses
analysis for the different wort separation the explanation of the need for small bubbles
systems. This is based on the American Society for better aeration as an example.
of Brewing Chemists (ASBC) testing sieve
system with a 100 g (0.22 lb) sample. Table 2. Typical sieve analysis of corn meal.
For most grains such as corn and other Screen size Hole size (inches) Corn on screen (%)
substrates such as cassava chips, hammer mills
12 0.0661 3.0
will probably perform the most efficiently. The
16 0.0469 8.0
proper screen size to use in the US would be 20 0.0331 36.0
number 6 through 8. This is the equivalent of 30 0.0234 20.0
openings in the range of 1/8 to 3/16 inches. In 40 0.0165 14.0
Europe, the screens would be 2 to 4 mm slot 60 0.0098 12.0
size. Through 60 7.0
A sieve analysis should be accomplished daily
(Table 2). The importance of a fine grind cannot Most distillers weigh the grist on the way to the
be overemphasized, however a very fine grind cooking operation by the use of weigh hoppers
must be accompanied by good mixing agitation and weigh belts. They use these figures for
to avoid formation of pills and lumps, creating proportioning solids and liquid in the slurrying
insufficient exposure of the solids to the liquid. operation.
Table 3. The free surface area of one inch of air when split up into bubbles of varying size1.
Bubble diameter Volume of bubble Surface area of bubble Surface area of 1 cubic inch of air = A/V
(In.) V=0.5236 d3 (cubic inches) A=3.1416 d2 (square inches) (square inches)
1.0 0.5236 3.1416 6
0.5 0.0645 0.7854 12
0.1 0.0005236 0.031416 60
0.01 0.0000005236 0.00031416 600
0.001 - - 6,000
1
Note: A = 3.1416 d2 = 6 = surface area
V 0.5236 d3 d
Grain handling: a critical aspect of distillery operation 39
Chapter 5
CHARLES A. ABBAS
Director of Yeast and Renewables Research, Archer Daniels Midland, Decatur, Illinois, USA
Introduction
The US alcohol industry has the capacity to 1996). According to a recent US Department of
process about 1 billion bushels of corn per Energy (DOE) draft report, the total
annum; which accounts for most of the fuel lignocellulosic feedstock potentially available in
ethanol produced domestically. This represents the US is estimated at 623 million dry tons
approximately 10% of the expected 2003 US (Figure 1). Excluding the biomass contribution
harvest. As production increases to the expected from energy crops and from sludge and biogas,
five billion gallons, the amount of corn used for the remaining lignocellulosics that are primarily
ethanol will also double. Can we continue to use derived from agricultural crop residues, forest
increasing amounts of grain? What would residues and municipal-source wood/paper/
happen if all of US gasoline demand (~150 organics can be estimated at a total of about 400
billion gallons) was replaced with ethanol? (This million dry tons. Assuming that these available
happened in Brazil in the 1970s and 1980s.) lignocellulosics are converted to ethanol with
Assuming a harvest of 10 billion bushels (250 yields approaching 80 gallons per ton, domestic
million tonnes) and excellent conversion ethanol production can be expanded further by
efficiency, the amount of ethanol produced an additional estimated 32 billion gallons. This
would still fall far short of the target. figure represents up to 20% of the current US
Demand for ethanol is increasing at a rate that gasoline consumption, which is estimated in
will require serious consideration of alternatives 2003 at 150 billion gallons. This chapter serves
to the primarily starch-based feedstocks over the to provide an overview of progress of research
next decade. Various lignocellulosic biomass on lignocellulosics and the opportunities and
sources such as agricultural residues, wood and technical challenges to the production of ethanol
forest wastes, municipal solid wastes and wastes from these renewable biomass feedstocks.
from the pulp and paper industry have the
potential to serve as low cost and abundant
feedstocks for ethanol production (Bothast and Major components of lignocellulosics
Saha, 1997; Galbe and Zacchi, 2002; Katzen et
al., 1999; Monceaux and Madson, 1999; Rogers, The composition of lignocellulosics varies
1997; Rogers et al., 1997; Singh and Mishra, greatly among the major categories of plant
1995; Tolan, 1999; von Sivers and Zacchi, sources (i.e. softwoods, hardwoods, straws) and
42 C.A. Abbas
Sludge
50 MdT
(manure and biosolids)
Biogas
11 MdT Agricultural crop residues
(landfill, digester 156 MdT
and sewage gas) (corn stover, wheat and
rice straw, cotton stalks)
Other wastes
161 MdT
(unused organic fraction of
municipal solid waste,
construction and demolition Forest residues
waste wood, urban tree residues 84 MdT
Total feedstocks available: 623 million dry tonnes (MdT) per year
Energy equivalent: 16 quadrillion BTU
also varies within each category (Tables 1 and more ordered or crystalline the cellulose, the less
2). In spite of these differences, the major soluble and less degradable it is. Unlike
components of all lignocellulosics are cellulose, cellulose, hemicellulose consists of shorter
hemicellulose, lignin and extraneous materials heteropoly saccharide chains composed of
(Singh and Mishra, 1995). Of these, cellulose is mixed pentosans and hexosans, which are more
by far the most abundant organic substrate. readily soluble and consequently amenable to
Cellulose is a linear homopolymer of enzymatic degradation. D-xylose and L-
anhydroglucose residues that are linked via ß- arabinose are the major constituents of the xylan
O-1, 4-glycosidic bonds. Different forms of or arabinoxylan backbone of the hemicellulose
cellulose exist with varying degrees of of straws with D-glucose, D-glucuronic, D-
polymerization and molecular weight (Singh and mannose and D-galactose as the other primary
Mishra, 1995). These forms include a highly constituents of the side chains. The principal
crystalline form, which results from the packing component of hardwood hemicellulose is
of cellulose microfibrils into highly ordered glucuroxylan; that of softwoods, glucomannan
fibers, as well as less ordered and amorphous (Jeffries et al., 1994). Minor constituents, which
forms (Lynd et al., 2002). These variations in are ether or ester-linked methyl, acetyl, or acyl
the crystal structure of cellulose affect its groups are also commonly encountered. The D-
hydrolysis by the action of enzymes and/or xylose in the xylan backbone is linked via 1,4-
chemicals. It is generally recognized that the ß-O-glycosidic linkages. The hexosan, araban
Lignocellulosics to ethanol 43
and other minor constituents of hemicellulose The primary function of lignin in lignocellulosics
are readily hydrolyzed from the backbone by is to provide plants with structural rigidity and it
chemical means, enzymatic means, or a also serves as a waterproofing agent in woody
combination of the two. By comparison, the tissue (Singh and Mishra, 1995). Lignin is an
enzymatic hydrolysis of the xylan backbone amorphous noncrystalline polymer composed of
proceeds at a much slower rate in the presence highly branched polymeric molecules of
of these constituents. The considerable variation phenylpropane-based monomeric units that are
that is present in the hemicellulose among plant linked by different types of bonds, which include
sources is responsible for the great differences alkyl-aryl, alkyl-alkyl and aryl-aryl ether bonds
in the activity of commercial xylanase or some of which are not readily hydrolyzable
hemicellulases on lignocellulosics. (Singh and Mishra, 1995). Thus, lignin is
considerably resistant to microbial degradation
Table 1. Distribution of cellulose and hemicellulose in in comparison to polysaccharides and other
wood and straw. natural polymers (Ericksson, 1981; Jeffries et
al., 1994; Odier and Artaud, 1992). Lignin is
Substrate Xylan Araban Galactan Mannan Glucan acid stable and insoluble in water but can be
%
solubilized under alkaline conditions. In
Hardwood 17.4 0.5 0.8 2.5 50.1 lignocellulosics the lignin is found closely bound
Softwood 5.7 1.0 1.4 11.2 46.3 to cellulose and hemicellulose and as a result
Straw 16.2 2.5 1.2 1.1 36.5 cannot be readily separated from these
carbohydrate polymers using conventional
Table 2. Composition (%) of common lignocellulosic methods (Jeffries et al., 1994; Singh and Mishra,
substrates. 1995). In lignocellulosics the association of
lignin with cellulose and hemicellulose impedes
Substrate Hexosans Pentosans Lignin enzymatic degradation of these carbohydrate
polymers. For this reason, physical or chemical
Bagasse 33 30 29 pretreatments have been designed to disrupt the
Birch 40 33 21 lignin-carbohydrate matrix to obtain good
Corn stover 42 39 14 enzymatic accessibility.
Groundnut shells 38 36 16
Oat straw 41 16 11
Extraneous materials in lignocellulosics consist
Pine 41 10 27 of extractable and nonextractable organics such
Rice straw 32 24 13 as terpenes, phenols and resins, which include
Sawdust 55 14 21 oils, alcohols, tannins, alkaloids, gums, silicate,
Wheat straw 30 24 18 oxalates and proteins. In some straws and hulls
nonextractables such as silica can account for
over 10% of the dry weight (Singh and Mishra,
The composition of lignin and extraneous 1995).
materials in plant sources varies considerably
(Singh and Mishra, 1995). Lignins are generally
distributed with hemicelluloses in the spaces of Processing of lignocellulosics
intracellular microfibrils in primary and
secondary cell walls and in the middle lamellae Processing of lignocellulosics is an essential step
as cementing components to connect cells and in releasing the carbohydrate components in
harden the cell walls of the xylem tissues order to derive fermentable feedstocks for the
(Higuchi, 1990). Lignins have been generally production of ethanol. This goal can be
classified into three major groups based on the accomplished by combining pretreatment and
chemical structure of the monomer units. The hydrolysis steps that involve physical, chemical,
groups are softwood lignin, hardwood lignin and thermal and/or enzymatic treatment (Bothast and
grass lignin (Higuchi, 1990). Because the lignin Saha, 1997; Galbe and Zacchi, 2002; Katzen et
and hemicellulose constituents differ, the cross- al., 1999; Tolan, 1999; McMillan, 1994a, 1994b;
links between these polymers vary from plant Ramos and Saddler, 1994; Singh and Mishra,
to plant and tissue to tissue (Jeffries et al., 1994). 1995; Stenberg et al., 1998; von Sivers and
44 C.A. Abbas
β 1-4 cellobiohydrolase
Cellulose
CH2OH H OH CH2OH H OH CH2OH
O O O
H H H H H
O O O
H OH H H OH H H
O OH H H O OH H H O OH H
H H H H H
O O
H OH CH2OH H OH CH2OH H OH
β−glucosidase
Cellobiose
CH2OH H OH
O
H OH
O
H OH H
β 1-4 endoglucanase
OH H H
OH H H H
O
H OH CH2OH
CH2OH H OH CH2OH
O O
H H H
O O
H OH H H
H H O OH H
HO H H H
O
H OH CH2OH H OH
CH2OH H OH
O
H H
O
H OH H
O OH H H O
H H
O
H OH CH2OH
O
H
HOH2C
OH H
α−arabinofuranosidase
H OH
β−endoxylanase
Hemicellulose
H H OH H H OH H
O O O
H H H
O O O
H OH H H OH H H
O OH H H O O H H O OH H
H H H H H
O O
H O H H OAc H H OH
COOH
O
H
H
Acetylxylan esterase
CH3O OH H
α−glucuronidase
H OH
β−xylosidase
H OH H
O
OH OH
H
Xylobiose OH
H
H
OH H
O
H H
O
H H OH
Zacchi, 1996). A list of pretreatment methods is proper use of enzyme loading with the use
provided in Table 3 and the reader is urged to of recycling and/or the use of recombinant
consult the references cited for greater detail. or naturally-occurring microorganisms that
The enzymes involved in the hydrolysis of the produce the required hydrolytic enzymes
polysaccharide components of lignocellulosics
are covered in the next section of this chapter 6) good sugar conversion and fermentation
and some of their substrates and catalytic action yields
sites are illustrated in Figures 2 and 3. The 7) utilization of a well designed fermentation
successful integration of the pretreatment and process that takes advantage of the currently
hydrolysis steps when combined with multistep available genetically engineered micro-
separation and fractionation, along with organisms
simultaneous saccharification and fermentation
8) a high ethanol concentration in the final
(SSF), can result in an environmentally friendly
fermentation broth which would be
process that yields a fermentable sugar slurry
economical to recover by conventional
that is nontoxic or minimally toxic to ethanol
distillation or through the use of new
producing microorganisms. The requirements
technologies such as solvent extraction or
for an economically viable process to meet the
membrane pervaporation.
above criteria are as follows:
1) a well characterized lignocellulosic feedstock Based on these criteria it becomes apparent that
the use of lignocellulosics for the production of
2) high solids loading by employment of
ethanol requires significant process integration.
physical means to increase surface area
This is best described by the employment of a
3) a treatment method to hydrolyze the holistic approach to processing referred to as
carbohydrate polymers to monosaccharides integrative bioprocessing (Abbas, 1996; Abbas,
while minimizing the formation of sugar 2000). Integrative bioprocessing has numerous
degradation products (HMF, furfural, advantages over conventional processing in that
levulinic acid) it combines advances in chemical engineering,
4) little or no by-product waste stream that bioprocessing and molecular biology to develop
requires waste treatment and/or disposal an optimized integrated system as compared to
optimizing the individual components of a
5) reduced enzyme cost and usage through process separately (Figure 4).
46 C.A. Abbas
Conventional Processing
Integrative Processing
One of the most promising feedstocks that can Enzymes for the processing of
meet many of the criteria listed is corn fiber hulls, lignocellulosics
a lignocellulosic by-product of the corn wet
milling process (Abbas, 1996; Abbas et al., The enzymatic depolymerization of
2002; Beery et al., 2003; Gulati et al., 1996; lignocellulosics requires an extensive battery of
Moniruzzaman et al., 1996; Saha et al., 1998; microbial hydrolytic enzymes from bacterial and
Saha and Bothast, 1999; Werpy et al., 2001). fungal sources that are capable of hydrolyzing
Other promising feedstocks that are currently cellulose, hemicellulose and lignin (Baker et al.,
available and meet many of the criteria listed 1995; Ericksson, 1981; Galbe and Zacchi, 2002;
above are agricultural residues and bioprocessing Ghosh and Ghosh, 1992; Jeffries, 1994;
plant by-products, which include soybean hulls, Joseleau et al., 1994; Lynd et al., 2002; Malburg
wheat straw, pentosan-containing fibrous starch et al., 1992; Nieves et al., 1997; Pelaez et al.,
streams from wheat processing, corn mesa 1995). These enzymes, which include
processing, fibrous by-product from dry grind glycohydrolyases, esterases and ligninases, must
tortilla flour operations, sugar beet pulp fibers, act on the polysaccharide backbones as well as
rice straw, sugar cane bagasse, spent brewers the branched chains and substituents attached
grains and dried distillers solubles from dry grind to lignin. Enzymatic digestion can be facilitated
ethanol plants. Beyond the above readily by pretreatment and hydrolysis, which enhances
available feedstocks, municipal wastes, paper enzyme reaction rate by improving binding to
and pulp wastes, corn stover and corncobs the solubilized polymers and consequently
represent additional attractive feedstocks. The improves process economics by reducing
conversion of these feedstocks can be enzyme loading. Enzyme preparations that are
economically feasible provided that a cost- known to be effective on lignocellulosics contain
effective process is delineated for their several exo- and endo-glycohydrolytic activities
collection, delivery, analysis, pretreatment, and include cellulase enzyme complexes that
hydrolysis of the carbohydrate fractions and/or degrade cellulose and hemicellulases such as D-
by-product disposal. The development of both xylanases, L-arabinases, D-galactanases,
cost effective hydrolytic enzymes and high D-mannases and esterases that release acetyl
solids pretreatment with good yields will be the groups from branched sugar polymers or that
major drivers for greater utilization of this last release ferulic acid or/other lignin precursors
group of lignocellulosics. (Figures 2 and 3). While an understanding of
the enzymatic arsenal involved in the primary
attack on lignin is still incomplete, several
Lignocellulosics to ethanol 47
important enzymes are known to be involved lignocellulosic (Adney et al., 1994). Additional
(Ericksson, 1981; Ericksson, 1993; Hatakka, cost reduction in processing lignocellulosics can
1994; Singh and Mishra, 1995). These enzymes be realized from reduced enzyme loading by
include the extracellular lignolytic enzymes timing enzyme addition with pretreatment and
derived from white rot fungi, which consist of hydrolysis and the use of simultaneous
lignin peroxidase (LiP), manganese peroxidase saccharification and fermentation (SSF). Use of
(MnP) and laccases (Ericksson, 1981; Ericksson, SSF aids in improved enzyme catalysis by
1993; Fiechter, 1993; Hatakka, 1994; Odier and reducing feedstock inhibition from the end
Artaud, 1992; Pelaez et al., 1995; Youn et al., products, as is the case with glucose and
1995). Unlike the polysaccharide degrading cellobiose inhibition of cellulases.
enzymes, which are substrate specific, the
enzymes that initiate degradation of lignin
appear to have a broad range of substrates Microbial fermentation of lignocellulosics
(Joseleau et al., 1994).
Most of the current commercially available For several decades microbial utilization of
carbohydrate and lignin degrading enzymes are sugars obtained from the hydrolysis of
derived primarily from fungal sources using lignocellulosics for the production of fuel ethanol
submerged liquid or in some cases solid state has been an active area of research (Corrington
surface fermentation (Hatakka, 1994; Duarte and and Abbas, 2003; DeCarvalho et al., 2002;
Costa-Ferreira, 1994; Nieves et al., 1997). When Dennison and Abbas, 2000; Dennison and
tested on a lignocellulosic feedstock of interest, Abbas, 2003; Dien et al., 1997; Green et al.,
considerable variation has been observed in 2001; Ho et al., 1999; Ho et al., 2000a; Ingram
commercial enzyme preparations in the and Doran, 1995; Ingram, 2000; Jeffries, 1985;
distribution of key enzyme activities and in the Lai et al., 1996; Wood and Ingram, 1996;
relative activities of cellulases and hemicellulases Sreenath and Jeffries, 1996; Sreenath and
(Abbas, unpublished observations). For this Jeffries, 2000; Jeffries et al., 1994; Jeffries et
reason careful attention must be paid to the al., 1995 Martin et al., 2002; Palmqvist and
method, substrate and conditions of the assay Hahn-Hagerdal, 2000; Rogers et al., 1997;
used by suppliers to measure enzyme activity. Supple et al., 2000; Joachimsthal and Rogers,
Whenever possible, additional information on 2000; Lawford and Rousseau, 1993a, 1993b;
other enzyme activities assayed should be Lawford et al., 2001; Lawford and Rousseau,
requested. Knowledge of the microbial source 2002; Lawford and Rousseau, 2003; Yamada et
and feedstock used by the manufacturer can be al., 2002; Tengborg et al., 2001; Stengberg et
useful in selecting enzyme preparations for al., 2000 ; van Zyl et al., 1999 ). This has been
further evaluation of a targeted feedstock. largely due to the absence of suitable
In addition to selection and use of the proper ethanolgens that can utilize the mixture of the
enzyme activities, care should be taken to design various pentose, hexose and higher sugars
a process that integrates enzyme addition present in hydrolysates (Singh and Mishra,
whenever possible into pretreatment, hydrolysis 1995). Research has confirmed considerable
and fermentation. In view of the current differences in the uptake and utilization of the
limitations of commercial enzymes and their various sugars by bacteria, yeast and molds (Ho
prohibitive cost, there exists the need to integrate et al., 2000b; Ingram, 2000; Jeffries et al., 1994;
and locate enzyme production facilities in the Jeffries et al., 1995; Green et al., 2001; Zhang
vicinity of, or as a part of, lignocellulosic et al., 1998; Zhang, 2002; Picataggio et al.,
processing plants (Tolan, 1999). The enzyme 1994). Several strategies have been employed
production facility should utilize a pretreated to remedy these limitations ranging from host
feedstock derived from the lignocellulosic of selection, host modification by classical strain
interest and the end user should employ a development approaches and genetic
minimally treated or a crude enzyme engineering of new strains (Ho et al., 1999; Ho,
preparation. This ensures a cost-effective 2000b; Jeffries et al., 1995; Ingram, 2000;
enzyme with the necessary mixture and ratios Picataggio et al., 1994). While an extensive list
of activities needed to process a given of microbial systems that have resulted from
48 C.A. Abbas
research in this area is outside the scope of this D-xylulose kinase (Amore et al., 1989; Gulati
chapter, a short list of ethanolgens of interest is et al., 1996; Ho et al., 1999; Singh and Mishra,
summarized in Table 4. A number of excellent 1995). By comparison the utilization of D-xylose
references on the growth of these organisms on in fungi proceeds with the action of D-xylose
lignocellulosic hydrolysates and/or sugars is reductase with the formation of xylitol as the
provided at the end of this chapter. In some product (Bruinenberg et al., 1984; Jeffries,
instances, modifications in the fermentation 1985; Jeffries et al., 1994; Jeffries et al., 1995;
process have been employed to enhance sugar Singh and Mishra, 1995; Jeppsson et al., 1999).
uptake and utilization as well as to circumvent This is followed by dehydrogenation of xylitol
hydrolysate toxicity and end product toxicity. by the action of xylitol dehydrogenase to D-
The list provided in Table 5 illustrates some of xylulose, which in turn is acted on by D-xylulose
the available options that can be used in the kinase (Singh and Mishra, 1995; Amore et al.,
fermentation of sugars in lignocellulosic 1989). The product of the phosphorylation, D-
hydrolysates by combining an appropriate xylulose-5-phosphate, is assimilated into the
hydrolysate and process. pentose phosphate pathway, which feeds into
No discussion of lignocellulosic hydrolysate the glycolytic pathway leading to the production
fermentation is complete without addressing the of ethanol.
major differences in the metabolism of the A significant difference in cofactor use exists
pentoses D-xylose and L-arabinose by fungi and between the naturally pentose-fermenting yeast
bacteria. These differences are illustrated in such as Pichia stipitis and the brewing yeast,
Figures 5 and 6. In bacteria, D-xylose utilization Saccharomyces cerevisiae, as illustrated in
involves the action of D-xylose isomerase Figure 7. In Pichia stipitis and in other genera
followed by phosphorylation of D-xylulose by of pentose fermenting yeast, the reductases that
Lignocellulosics to ethanol 49
In Bacteria In Fungi
D-XYLOSE
Xylose reductase (XR)
NADPH
Xylose NADP+
isomerase (XI) Xylitol
NAD+
NADH
Xylitol dehydrogenase (XDH)
Xylulose
ATP
Xylulokinase (XK)
ADP
Xylulose-5-P
Pentose phosphate
pathway
Embden-Meyerhoff pathway
(Glycolysis)
In Bacteria In Fungi
L-ARABINOSE
NADPH-dependent
NAD(P)H aldopentose reductase
L-arabinose
isomerase NAD(P)+
L-arabitol
NAD+
L-ribulose Penitol-NAD-
NADH dehydrogenase
ATP
Kinase
ADP
L-xylulose
L-ribulose-5-phosphate L-xylulose-NADPH
Epimerase reductase
D-xylulose-5-phosphate Xylitol
Penitol-NAD-
D-xylulose dehydrogenase
kinase
Pentose D-xylulose
phosphate
pathway
reduce D-xylose to xylitol can utilize either yeast is unable to produce significant levels of
NADH or NADPH. In contrast, in S. cerevisiae ethanol from D-xylose under anaerobic
the host reductase that acts on D-xylose is limited conditions (Amore et al., 1989; Bruinenberg et
to NADPH. This cofactor use in S. cerevisiae is al., 1984; Jeffries, 1985). Cloning of the Pichia
responsible for the cofactor regeneration stipitis D-xylose reductase that utilizes NADH
imbalance and is one of the reasons why this into S. cerevisiae improved the ability of this
50 C.A. Abbas
yeast to utilize D-xylose and resulted in arabinose isomerase, which yields L-ribulose as
improved ethanol production by the genetically the product (Deanda et al., 1996; Saha and
engineered Saccharomyces strain (Table 6). Bothast, 1995). Further metabolism of L-ribulose
Further genetic engineering of S. cerevisiae to proceeds with its phosphorlyation by the action
increase D-xylulose kinase activity yielded new of a L-ribulose kinase to form L-ribulose-5-
recombinant yeast strains with significant phosphate which is converted to D-xylulose -5-
ethanol production from D-xylose (Ho et al., phosphate. As indicated earlier, D-xylulose-5-
1999; Ho et al., 2000a; Eliasson et al., 2000). phosphate is assimilated further via the pentose
Most recently, employment of metabolic flux phosphate pathway into glycolysis, which in
analysis to recombinant S. cerevisiae grown on ethanolgenic recombinant bacteria culminates in
a mixture of D-glucose and D-xylose has been the production of ethanol. In fungi the utilization
used to further delineate additional genes that of L-arabinose involves the initial action of an
can be targeted to improve ethanol production aldopentose reductase with the production of L-
from D-xylose (Figure 8). This latest approach arabitol. The metabolism of L-arabitol is carried
highlights the great potential offered by on further through the action of a penitol
metabolic engineering to achieve further dehydrogenase, an L-xylulose reductase, a
improvements in ethanol yield and productivity second penitol dehydrogenase and finally D-
in recombinant Saccharomyces cerevisiae xylulose kinase. The product of the last step,
(Nielsen, 2001; Ostergaard, 2000). D-xylulose-5-phosphate, is assimilated into the
pentose phosphate pathway and from there
Table 6. Effect of XK overexpression by S. cerevisiae.
further into the glycolytic pathway.
Recently two approaches that utilized
+XR +XDH Brewing yeast expressing these two knowledge gained from L-arabinose metabolism
genes are poor producers of ethanol.
in bacteria and fungi were successfully employed
+XR +XDH +XK Brewing yeast expressing these to engineer strains of S. cerevisaie to ferment
three genes show greater ethanol this sugar to the ethanol. In the first approach,
production. the bacterial genes from Escherichia coli (Ara B
Recombinants expressing a
and Ara D) and Bacillus subitilis (Ara A) were
truncated XK are good D-xylose
fermenters
cloned into a S. cerevisaie host strain
(Becker and Boles, 2003). This strain was further
engineered to over express Gal 2 (galactose
The metabolism of L-arabinose in bacteria permease) and to enhance the enzyme
(outlined in Figure 6) involves the enzyme L- transaldolase activity while reducing L-
NADPH NADP
NADH NAD NAD NADH P. stipitis
XR XDH
XR XDH
*Fructose-6-P Xylose
Phosphofructokinase
*Ribose-5-PTransketolase dehydrogenase
Xylulose-5-P NADH
Fructose 1,6
bisphosphate
Transladolase Xylose
*Erythrose-4-P Xylulose
Phosphofructoaldolase kinase
Dihydroxyacetone Glyceraldehyde
phosphate Triosephosphate 3-phosphate
isomerase
Sedoheptulose-7-P
NADH Glyceraldehyde-3-P
Glycerol-3-phosphate dehydrogenase NADH dehydrogenase
Glycerol-3-phosphatase
Glycerol
*3-phosphoglycerate Acetyl CoA
Enolase CO2
NADH
*Phosphoenolpyruvate Respiration
Pyruvate O2 H2O
dehydrogenase
Pyruvate kinase NADH
*Oxaloacetate Isocitrate
Glycolysis
*Pyruvate Malate NADPH
Pyruvate carboxylase dehydrogenase
CO2 Oxaloacetate Isocitrate
Pyruvate CO2 Malate dehydrogenase
decarboxylase
NADH NADPH CO2
Fumerase
Ethanol Acetaldehyde
Alcohol Succinate
dehydrogenase dehydrogenase
Acetaldehyde CO2
dehydrogenase * Acetyl CoA Succinate *α-ketoglutarate
NADPH
Acetate NADH
ribulokinase activity. The engineered strain research would be industry driven. Over the next
fermented L- arabinose to ethanol under reduced decade the successful integration of the research
oxygen conditions with a doubling time of 7.9 results on lignocellulosics from the DOE
hrs in a medium containing L-arabinose as the platforms and from DOE co-funded industry-
sole carbon source. In the second approach, the led pilot plant biorefineries will be key to further
five genes involved in L-arabinose metabolism US efforts in improving the economics of
in fungi were cloned into a S. cerevisiae host biomass conversion to ethanol and chemicals.
strain (Richard et al., 2002; Richard et al.,
2003). These genes were over expressed in the
host strain leading to growth on L-arabinose and Concluding remarks
the production of ethanol under anaerobic
conditions. In the US, lignocellulosics are attractive
feedstocks for further expansion of fuel ethanol
production. The recent advances in pretreatment
Current status of lignocellulosic research and hydrolysis when combined with the
in the US availability of new ethanolgens have increased
the feasibility of hemicellulose conversion to
Over the past decade several initiatives that aim ethanol. The conversion of cellulose continues
to improve the economics of the utilization of to be economically and technically challenging
lignocellulosics for fuel ethanol production have due to the current high cost of commercially
been proposed and funded by the US DOE. available enzymes as well as the energy cost for
These initiatives focused primarily on three pretreatment and hydrolysis and cost of
areas: subsequent detoxification. Processing facilities
that can utilize corn crop residues such as corn
1) sugar or feedstock development platform with stover or corn wet mill biorefineries that
most recent focus on lignocellulosics present currently produce several hundred million
in corn fiber hulls from corn wet mills and pounds of corn fiber hulls annually will represent
corn stover the first generation of lignocellulosic-based
ethanol production plants built. These facilities
2) enzyme development platform, which will be adjacent and fully integrated into existing
primarily targets improvements in cellulase plant operations and thereby benefit from the
enzyme technology with the goal of reducing available infrastructure. Based on current
the cost of these enzymes to bring their research trends it is predicted that these plants
commercial costs in line with other industrial will be in operation within the next decade and
enzymes therefore will represent in the near term the
3) ethanolgens development platform with a successful commercialization of lignocellulosic-
focus on yeast and bacterial recombinants based biorefineries.
with broad and improved feedstock utilization
hydrolysis of corn fiber. 24th Symposium on Dennison, E.K. and C.A. Abbas. 2000.
Biotechnology for Fuels and Chemicals. Evaluation of recombinant Zymomonas mobilis
Gatlinburg, TN (abstract). strains on corn fiber hydrolysate. 22 nd
Adney, W.S., C.I. Ehrman, J.O. Baker, S.R. Symposium on Biotechnology for Fuels and
Thomas and M.E. Himmel. 1994. Cellulase Chemicals. Gatlinburg, TN (abstract).
assays. Method for empirical mathematical Dennison, E.K. and C.A. Abbas. 2003.
models. ACS Symposium Series 566:218-235. Evaluation of recombinant microorganism
Amore, R., M. Wilhelm and C.P. Hollenberg. ethanol fermentation of corn fiber hydrolysate.
1989. The fermentation of xylose-an analysis 25th Symposium on Biotechnology for Fuels
of the expression of Bacillus and Actinoplanes and Chemicals. Breckenridge, CO (abstract).
xylose isomerase genes in yeast. Appl. Dien, B.S., R.B. Hespell, L.O. Ingram and R.J.
Microbiol. Biotechnol. 30:351-357. Bothast. 1997. Conversion of corn milling
Baker, J.O., W.S. Adney, S.R. Thomas, R.A. fibrous co-products into ethanol by
Nieves, Y-C. Chou, T.B. Vinzant, M.P. Tucker, recombinant Escherichia coli strains K011 and
R.A. Laymon and M.E. Himmel. 1995. SL40. World J. Microbiol. and Biotech.
Synergism between purified bacterial and 13:619-625.
fungal cellulases. ACS Symposium Series Duarte, J.C. and M. Costa-Ferreira. 1994.
618:113-141. Aspergilli and lignocellulosics: enzymology
Becker, J. and E. Boles. 2003. A modified and biotechnological applications. FEMS
Saccharomyces cerevisiae strain that consumes Microbiology Reviews 13:377-386.
L-arabinose and produces ethanol. Appl. Env. Eliasson, A., C. Christensson, C.F. Wahlbom and
Microbiol. 69(7):4144-4150. B. Hahn-Hagerdal. 2000. Anaerobic xylose
Beery, K.E., C.A. Abbas, T. A. Werpy, A.J. fermentation by recombinant Saccharomyces
Schmidt and R.J. Orth. 2003. Separation and cerevisiae carrying XYL1, XYL2 and XKS1 in
conversion of corn fiber. 225 th American mineral medium chemostat cultures; plasmid
Chemical Society National Meeting. New YIpXR/XDH/XK-mediated recombinant yeast
Orleans, LA (abstract). construction for improved ethanol production.
Bothast, R.J. and B.C. Saha. 1997. Ethanol from Appl. Environ. Microbiol. 66(8):3381-3386.
agricultural biomass substrates: a review. Adv. Eriksson, K-E. 1981. Microbial degradation of
Appl. Microbiol. 44:261-286. cellulose and lignin. Ekman-Days 1981, Int.
Bruinenberg, P.M., P. H.M. de Bot, P. van Dijken Symp. Wood Pulping Chem.
and W. A. Scheffers. 1984. NADH-linked Eriksson, K-E.L. 1993. Lignin biodegradation
aldose reductase: the key to anaerobic and practical utilization: concluding remarks-
alcoholic fermentation of xylose by yeasts. where do we stand and where are we going?
Appl. Microbiol. Biotechnol. 19:256-260. J. Biotech. 30:149-158.
Corrington, P. and C. Abbas. 2003. Screening Fiechter, A. 1993. Function and synthesis of
of ethanolgens on corn fiber hydrolysates. 25th enzymes involved in lignin degradation. J.
Symposium on Biotechnology for Fuels and Biotech. 30:49-55.
Chemicals. Breckenridge, CO (abstract). Galbe, M. and G. Zacchi. 2002. A review of the
De Carvalho Lima, K.G., C.M. Takahashi and production of ethanol from softwood; alcohol
F. Alterthum. 2002. Ethanol production from production via hemicellulose, cellulose and
corn cob hydrolysates by Escherichia coli lignin using cellulase hydrolysis process,
KO11. J. Indust. Microbiol. and Biotech. fungus fermentation and distillation: a review.
29:124-128. Applied Microbiol. and Biotech. 59(6):618-
Deanda, K., M. Zhang, C. Eddy and S. 628.
Picataggio. 1996. Development of an Ghosh, B.K. and A. Ghosh. 1992. Degradation
arabinose-fermenting Zymomonas mobilis of cellulose by fungal cellulase. In: Microbial
strain by metabolic pathway engineering; Degradation of Natural Products (G.
metabolic engineering for ethanol production Winkelmann, ed). VCH Publishers, New York,
from agriculture residue waste. Appl. Environ. NY, U.S.A.
Microbiol. 62(12): 4465-4470.
54 C.A. Abbas
Green, E., N. Baghael-Yazdi and M. Javed. Degradation (C. Ratledge, ed). Kluwer
2001. Fermentative production of ethanol by Academic Publishers, The Netherlands, pp.
heterologous expression of a pyruvate 233-277.
decarboxylase gene in Bacillus sp. Jeffries, T.W., K. Dahn, W.R. Kenealy, P. Pittman,
WO0149865A1 International patent H.K. Sreenath and B.P. Davis. 1994. Genetic
application, 12 July, 2001 engineering of the xylose-fermenting yeast
Gulati, M., K. Kohlmann, M.R. Ladisch, R. Pichia stipitis for improved ethanol production.
Hespell and R. Bothast. 1996. Assessment of Am. Chem.Soc. 208th ACS National Meeting
ethanol production options for corn products. (abstract).
Bioresource Tech. 58:253-264. Jeffries, T.W., B.P. Davis, H.K. Sreenath, J.Y.
Hatakka, A. 1994. Lignin-modifying enzymes Cho and K. Dahn. 1995. Genetic development
from selected white-rot fungi: production and of the xylose fermenting yeast Pichia stipitis.
role in lignin degradation. FEMS Am. Chem. Soc. 209th ACS National Meeting
Microbiology Reviews 13:125-135. (abstract).
Higuchi, K. 1990. Lignin biochemistry: Jeppsson, H., K. Holmgren and B. Hahn-
biosynthesis and biodegradation. Wood Sci. Hagerdal. 1999. Oxygen-dependent xylitol
Technology. 24:23-63. metabolism in Pichia stipitis; influence of
Ho, N.W., Z. Chen, A.P. Brainard and M. Sedlak. oxygen on utilization of xylitol alone, or as a
1999. Successful design and development of mixture with xylose or glucose, for ethanol
genetically engineered Saccharomyces yeasts production. Appl. Microbiol. Biotechnol.
for effective cofermentation of glucose and 53(1):92-97.
xylose from cellulosic biomass to fuel ethanol. Joachimsthal, E.L. and P.L. Rogers. 2000.
In: Advances in Biochemical Engineering/ Characterization of a high-productivity
Biotechnology Vol. 65:163-192 (T. Scheper, recombinant strain of Zymomonas mobilis for
ed). Springer-Verlag and Berlin Heidelberg, ethanol production from glucose/xylose
Germany. mixtures. Applied Biochemistry and
Ho, N.W., Z. Chen, A.P. Brainard and M. Sedlak. Biotechnology - Part A Enzyme Eng. and
2000a. Genetically engineered Saccharomyces Biotech. 84-86:343-356.
yeasts for conversion of cellulosic biomass to Joseleau, J-P. , S. Gharibian, J. Comtat, A.
environmentally friendly transportation fuel Lefebvre and K. Ruel. 1994. Indirect
ethanol. ACS Symposium Series, pp. 143-159. involvement of ligninolytic enzyme systems
Ho, N.W., Z. Chen and M. Sedlak. 2000b. in cell wall degradation. FEMS Microbiology
Strategies for successful metabolic engineering Reviews 13:255-264.
of Saccharomyces yeasts to effectively co- Katzen, R., P.W. Madson and D.A. Monceaux.
utilize glucose and xylose from renewable 1999. Lignocellulosic feedstocks for ethanol
cellulosic biomass for the production of production: the ultimate renewable energy
ethanol and other industrial products. Am. source. In: The Alcohol Textbook. 3rd Edition
Chem. Soc. 219th ACS National Meeting (K.A. Jacques, T.P. Lyons and D.R. Kelsall,
(abstract). eds). Redwood Books, Trowbridge, Wiltshire,
Ingram, L.O. 2000. Engineering bacteria for UK, pp. 117-127.
ethanol from lignocellulosics. Am. Chem. Soc. Lai, X., L.O. Ingram and R.B. Hespell. 1996.
219th ACS National Meeting (abstract). Cloning of celliobiose utilization genes from
Ingram, L.O. and J.B. Doran. 1995. Conversion G- and G+ bacteria: functional expression in
of cellulosic materials to ethanol. FEMS Escherichia coli. Gen. Meet. Am. Soc.
Microbiology Reviews 16: 235-241. Microbiol. 96th Meeting (abstract).
Jeffries, T.W. 1985. Emerging technology for Lawford, H.G. and J.D. Rousseau. 1993a.
fermenting D-xylose. Trends in Biotechnology Production of ethanol from pulp mill hardwood
Vol.3(8):208-212. and softwood spent sulfite liquors by
Jeffries, T.W. 1994. Biodegradation of lignin and genetically engineered E. coli; Escherichia
hemicelluloses. In: Biochemistry of Microbial coli application in wood pulp waste-disposal
Lignocellulosics to ethanol 55
from spent sulfite liquor. Appl. Biochem. hemicellulose hydrolyzates to ethanol. ACS
Biotechnol. 39-40:667-685. Symposium 566:411-437.
Lawford, H.G. and J.D. Rousseau.1993b. McMillan, J.D. 1994b. Pretreatment of
Developments in hemicellulose bioconversion lignocellulosic biomass. ACS Symposium
technology; ethanol production by xylose, Series 566:292-324.
newsprint or spent sulfite liquor fermentation Monceaux, D.A and P.W. Madson. 1999.
by recombinant Escherichia coli carrying Alternative feedstocks: a case study of waste
plasmid pLOI297: a review. Energy Biomass conversion to ethanol. In: The Alcohol
Wastes XVI: 559-97. Textbook. 3rd Edition (K.A. Jacques, T.P. Lyons
Lawford, H.G. and J.D. Rousseau. 2002. and D.R. Kelsall, eds). Redwood Books,
Performance testing of Zymomonas mobilis Trowbridge, Wiltshire, UK, pp.129-135.
metabolically engineered for cofermentation Moniruzzaman, M., B.S. Dien, B. Ferrer, R.B.
of glucose, xylose and arabinose. Applied Hespell, B.E. Dale, L.O. Ingram and R.J.
Biochemistry and Biotechnology - Part A Bothast. 1996. Ethanol production from AFEX
Enzyme Engineering and Biotechnology 98- pretreated corn fiber by recombinant bacteria.
100:429-448. Biotech. Letters 18(8):985-990.
Lawford, H.G. and J.D. Rousseau. 2003. Nielson, J. 2001. Metabolic engineering. Appl.
Cellulosic fuel ethanol: Alternative process Microbiol. Biotechnol. 55:263-283.
designs with wild-type and recombinant Nieves, R.A., C.I. Ehrman,W.S. Adney, R.T.
Zymomonas mobilis. Applied Biochemistry Elander and M.E. Himmel. 1997. Technical
and Biotechnology – Part A Enzyme communication: survey and analysis of
Engineering and Biotechnology 105-108:457- commercial cellulase preparations suitable for
469. biomass conversion to ethanol; cellulase
Lawford, H.G., J.D. Rousseau and J.S. Tolan. complex characterization a view to its use in
2001. Comparative ethanol productivities of ethanol production from lignocellulose. World
different Zymomonas recombinants fermenting J. Microbiol. Biotechnol. 14(2):301-304.
oat hull hydrolysate. Applied Biochemistry and Odier, E. and I. Artaud. 1992. Degradation of
Biotechnology- Part A Enzyme Engineering lignin. In: Microbial Degradation of Natural
and Biotechnology 91-93:133-146. Products (G. Winkelmann, ed). VCH
Lynd, L.R., P.J. Weimar, W.H. van Zyl and I.S. Publishers, New York, NY, U.S.A.
Pretorius. 2002. Microbial cellulose utilization: Ostergaard, S., L. Olsson and J. Nielsen. 2000.
fundamentals and biotechnology. Microbiol. Metabolic engineering of Saccharomyces
and Molecular Biol. Rev. 66(3):506-577. cerevisiae. Microbiol. Mol. Biol. Rev. 64:34-
Malburg, L.M. Jr., J.M. Tamblyn Lee and C.W. 50.
Forsberg. 1992. Degradation of cellulose and Palmqvist, E. and B. Hahn-Hagerdal. 2000.
hemicelluloses by rumen microorganisms. In: Fermentation of lignocellulosic hydrolysates.
Microbial Degradation of Natural Products (G. II: Inhibitors and mechanisms of inhibition; the
Winkelmann, ed). VCH Publishers, New York, inhibitory effect of weak acid, furan derivative
NY, U.S.A. and phenolic compound on ethanol production
Martin, C., M. Galbe, C.F. Wahlbom, B. Hahn- by Saccharomyces cerevisiae from a
Hagerdal and L.J. Jonsson. 2002. Ethanol hydrolysate substrate; a review. Bioresource
production from enzymatic hydrolysates of Technol. 74(1):25-33.
sugarcane bagasse using recombinant xylose- Pelaez, F., M.J. Martinez and A.T. Martinez.
utilizing Saccharomyces cerevisiae; involving 1995. Screening of 68 species of
sugarcane bagasse hydrolysate, detoxification, basidiomycetes for enzymes involved in lignin
phenoloxidase laccase, recombinant xylose- degradation. Mycol. Res. 99(1):37-42.
utilising Saccharomyces cerevisiae and Picataggio, S.K., M. Zhang; and Mark
fermentation. Enzyme and Microbial. Tech. Finkelstein.1994. Development of genetically
31(3):274-282. engineered microorganisms for ethanol
McMillan, J.D. 1994a. Conversion of production. ACS Symposium Series 566:342-
362.
56 C.A. Abbas
cerevisiae on different carbon sources; xylose Yamada, T., M.A. Fatigati and M. Zhang. 2002.
conversion into ethanol by fermentation of Performance of immobilized Zymomonas
recombinant Saccharomyces expressing mobilis 31821 (pZB5) on actual hydrolysates
Pichia stipitis xylose-reductase and xylitol produced by Arkenol technology. Applied
dehydrogenase. Appl. Microbiol. Biotechnol. Biochemistry and Biotechnology 98-100:899-
52(6):829-833. 907.
Von Sivers, M. and G. Zacchi. 1996. Ethanol Youn, K-D, Y.C. Hah and S-O. Kang. 1995. Role
from lignocellulosics: a review of the of laccase in lignin degradation by white-rot
economy; alcohol production and fungi. FEMS Microbiol. Letters 132:183-188.
lignocellulose degradation and saccharifi- Zhang, M. 2002. Novel recombinant
cation. Bioresource Technol. 56(2-3):131-140. Zymomonas mobilis 31821/pZB5 strain
Werpy, T.A., J. Franz, M. Alnajjar, A. Schmidt, derived from Z. mobilis ATCC 31821 useful
R.J. Orth, J. Magnuson, T. P. Binder, C. A. for producing ethanol by fermenting xylose;
Abbas, T. A. Stoa and B. Sedlacek. 2001. Corn xylose utilization in ethanol production by
fiber separation and subsequent conversion to vector-mediated gene transfer, expression in
fuels and chemicals. 23 rd Symposium on host cell. WO 2002038740.
Biotechnology for Fuels and Chemicals. Zhang, M., C.Y. Chou, S.K. Picataggio and M.
Breckenridge, CO. Finkelstein. 1998. New Zymomonas mobilis
Wood, B.E. and L.O. Ingram.1996. Production strain containing exogenous genes; for growth
of recombinant bacterial cellulases by on xylose and arabinose for ethanol
ethanolgenic bacteria: evaluation for cellulose production. WO 9850524.
fermentation. Gen. Meet. Am. Soc. Microbiol.
96th Meet (abstract).
Ethanol production from cassava 59
Chapter 6
NGUYEN T. T. VINH
The Research Institute of Brewing, Hanoi, Vietnam
Introduction to cassava
Cassava (Manihot esculenta), which is believed crude protein, 0.7-2.6%; fat, 0.2-0.5%; fiber, 0.8-
to have originated in Latin America, is an 1.3%; ash, 0.3-1.3%; cellulose, 1-3.1%;
important food crop in many tropical countries polyphenol, 0.1-0.3%. Cassava chips typically
of Africa, South America and Asia (Dougrell, contain 65-70% of starch, which makes it one
1982). A number of attributes have made it an of the richest sources of fermentable substrate
attractive crop for small farmers with limited for the production of alcohol. In nutritional terms,
resources in marginal agricultural areas: cassava is primarily a source of carbohydrate-
derived energy, most of which is derived from
• it is one of the most efficient carbohydrate- starch. In addition, cassava contains a small
producing crops; amount of B vitamins and toxic substances.
Vitamins include B 1 , B 2 and B 6 . The toxic
• it is tolerant of poor soil fertility and drought;
compound in cassava is phazeounatine,
• it has the ability to recover from the damage consisting of two cyanogenic glucosides,
caused by most pests and diseases; linamarin and lotaustralin. Linamarin accounts
for more than 80% of the cassava cyanogenic
• the roots can be left in the ground for long
glucosides. Phazeounatine content ranges from
periods as a food reserve and thus provide
0.001-0.04 mg/100g; and is present mainly in
insurance against famine;
the outer layer of bitter cultivars. Phazeounatine
• the crop is well adapted to traditional mixed itself is not toxic, but it is easily degraded to
cropping agricultural systems and release hydrocyanic acid (HCN). Typical
subsistence cultivation in which farmers seek composition of cassava chips is shown in Table
to minimize the risk of total crop failure. 1.
Yields of cassava range from 15-150 tonnes per Table 1. Composition of cassava chips.
ha. In Brazil it can reach 80-90 tonnes; and in Composition %
Colombia yields of >50 tonnes/ha are
consistently obtained. Moisture 14
The roots are the principle edible portion of Carbohydrate 67.6
Protein 1.75
the plant and typical ranges of composition are Lipid 0.87
water, 62-65%; total carbohydrate, 32-35%; Cellulose 3.38
Minerals 1.79
60 N.T.T. Vinh
temperature, is easily liquified and saccharified. Processing cassava for alcohol production
The main advantage of cassava over any other
crop for this purpose is the presence of highly MILLING
fermentable sugars after saccharification. Large
volumes of the saccharified starch are fed into The purpose of milling is to break up the cassava
fermentation vessels and inoculated with actively roots to as small a particle size as possible in
growing yeast (Saccharomyces cerevisiae). A order to facilitate subsequent penetration of water
continuous saccharification system is shown in in the cooking process. Most distilleries use
Figure 1. The pH of the mash for fermentation is hammer mills for the cassava dried chips. In
optimally 4–4.5 and the temperature range is 28- contrast to corn and rice, cassava chips are soft
32°C. Alcohol is recovered from fermented mash and easy to mill. With a hammer mill, the raw
after 48-72 hrs by distillation. material is introduced into a chamber in which a
Ethanol yields are 70-110 liters of absolute number of hammers rotate at high speed from
alcohol per ton of cassava roots depending on 75-80 m/s. The collision of the hammers with
the variety and method of manufacture, which the raw material causes breakdown to a meal.
compares to an expected yield of 350-400 liters/ The mill outlet contains a retention screen that
tonne from maize. The crude alcohol of cassava holds back larger particles until they are broken
is described as average in quality. It has a down further so that there will be a known
disagreeable odor, but can be improved if the maximum particle size in the meal. The screens
first and last fractions in the distillation process are normally in the size range of 2-3 mm for a
are discarded. It is utilized for industrial fine grind and 6-8 mm for a coarse grind.
purposes, including cosmetics, solvents and Depending on the cooking method, the target
pharmaceutical products. If production is size of particles may differ. Liquefaction in
required for human consumption, special care response to cooking and amylase will improve
should be taken in handling the roots to rid them with fineness of the grind, which is a significant
of hydrocyanic acid. factor in the final alcohol yield. It is possible to
obtain a 5-10% difference in yield when moving
from coarse to finely ground meal, as is the case
with cereal grains.
Enzyme
tank
30%
Mash
Saccharification
tank 1
70%
55 - 60°C
15 - 20 min
Saccharification Fermentors
system 2 Cooler
10 - 15 min
Pump
Figure 1. Typical continous saccharification system.
62 N.T.T. Vinh
Cooking is the entire process beginning with There are three methods for cooking cassava:
mixing the cassava meal with water through to batch, semi-continuous and continuous. The
delivery of a mash ready for fermentation. The ratio of water to material is normally 3.5:1-3:1
key to cooking is to liquefy the starch so it can with amylase. The batch method is still used in
be pumped. There is always a protective many countries, but has the disadvantages of
membrane around starch granules, whether the long cooking times at high temperature with
feedstock is corn, rice, wheat or tubers such as accompanying sugar loss. The continuous
manioc, potato, and sweet potato. Therefore the method requires strict control. The heat-stable
essential purpose of cooking is to break the amylase is popular for alcohol plants, and is used
membrane, releasing the starch. at a rate of 0.02–0.03%. Around 80% of the
Gelatinization is the phenomena by which a amylase is used for the first liquefaction and 20%
granule of starch swells in presence of water and is added for additional liquefaction after boiling
heat so that surface area exposed to enzymes is and cooling to 90°C.
increased. During gelatinization water penetrates
the granule making the breakdown of the starch Liquefaction and saccharification
easier for the α-amylase. Starch is composed of
amylose and amylopectin. While starch in corn Hydrolysis by the endoenzyme α-amylase is
contains 85-90% amylopectin, cassava starch random and rapid, reducing starch to detrins of
contains 70% amylopectin. For each type of raw varying chain length. The shorter the chain
material there is a typical gelatinization length, the less work remains for the exoenzyme
temperature range, which depends on starch glucoamylase, which releases single glucose
granule size. When granule size is small molecules by hydrolyzing successives α-1,4
gelatinization temperatures are lower. linkages beginning at the non-reducing end of
Gelatinization temperature range of corn starch is the dextrin chain. Glucoamylase also hydrolyzes
67-80°C, rice starch 65-85°C, and 61-70°C for α-1,6 branch linkages, but at a much slower rate.
cassava starch. In addition, gelatinization Recently simultaneous saccharification and
temperature depends on the electrolytes in solution. fermentation (SSF) of cassava has been of interest
Alkaline substances decrease gelatinization as a means of reducing the cost and time
temperature, while sugars increase gelatinization involved in ethanol production. In an experiment
temperature. by Keawsompong and co-workers (2003) in
When cooking in weak acid conditions, Thailand, conventional liquefaction/sacchari-
cellulose normally cannot be hydrolyzed. fication was compared with an SSF process that
Hemicellulose is hydrolyzed by xylanase in the employed Rhizozyme TM , a suface culture
presence of H+ and high temperatures forming glucoamylase targeted for the temperature and pH
dextrins and lower molecular weight substances conditions of the fermentor. After 48 hrs of
including arabinose and xylose. fermentation, similar amounts and conversion
During cooking, a small amount of starch will efficiencies were obtained, however by leaving
be converted to sugars by endogenous amylase. out the saccharification step, the SSF process was
Sucrose, glucose, fructose and a little maltose completed 25% faster than the conventional
are formed in cooking. At high temperatures method of sequential liquefaction and
these sugars may be converted into caramel, saccharification (Figure 2). Yield increases were
furfural, oxymethyl furfural and other also observed, presumably because other
compounds. The rate of conversion depends on materials became available for fermentation. At
temperature and pH. Experience has shown that 36 hrs the conventional system had 8-9% ethanol
adjusting mash pH to 3.5 may reduce the rate of compared to 10%+ in the Rhizozyme TM
sugar conversion (5% for glucose and 26% for fermentor. Cell counts were also significantly
fructose). If pH is raised to 6.5, then conversion higher.
rate will be 80% for glucose and 90% for
fructose. FERMENTATION
2.0 26 12
Conventional
Ethanol
24
10
1.5 22
1.0 18 6
16
TSS 4
.5 14
2
12
0.0 10 0
0 12 24 36 48
Total
soluble solids Time (hr)
(Brix)
2.0 26 12
24 RhizozymeTM
10
Ethanol
1.5 22
Cell (108 cells/ml)
1.0 18 6
Cell
16
4
.5 14 TSS
2
12
0.0 10 0
0 12 24 36 48
Total
soluble solids Time (hr)
(Brix)
Figure 2. Changes in total soluble solids (TSS), brix, cell count and ethanol during cassava fermentation by yeast in a conven-
tional (top) liquefaction and saccharification process and a simultaneous saccharification and fermentation (bottom) process using
RhizozymeTM. The SSF process required 25% less time (from Keawsompong et al., 2003).
Dougrell, L.M. 1982. Cassava - modern 2003. Ethanol production from cassava chips:
agricultural system development towards simultaneous saccharification and fermentation
ethanol production in Australia. International process. BioThailand Conference, Pattaya,
Alcohol Fuel Technology Symp. NZL, p.63- Thailand.
72. Sriroth, K., K. Piyachomkwan, K. Sangseethong
FAO/GIEWS - Food Outlook No 3, June 2001, and C. Oates. 2002. Modification of cassava
p.9. starch. X International Starch Convention, 11-
Keawsompong, S., K. Piyachomkwan, S. 14 June 2002, Kracow, Poland.
Walapatit, C. Rodjanaridpiched and K. Sriroth.
Whey alcohol - a viable outlet for whey? 65
Chapter 7
JACK O’SHEA
Alltech Inc., Dunboyne, County Meath, Ireland
Introduction
Whey is the main by-product from casein and Table 1. Approximate composition of whey.*
cheese manufacture. In the past, cheese making Component %
was a traditional cottage industry using the milk
Water 94.50
from various animals such as goats, camels and Dry matter 5.75
reindeer. Today cheese is usually made in large Protein 0.80
production facilities from cow’s milk by the Lactose 4.40
application of microbiology and chemistry. Ash 0.50
Casein, the chief milk protein, is coagulated by Fat 0.05
the enzymatic action of rennet or pepsin, by lactic *This is an average analysis, the type of cheese produced
acid producing bacteria, or by a combination of will influence composition of the whey.
the two. The next step involves separating the
curds from the whey, a process facilitated by Approximately 100 kg of milk are required to
heating, cutting, and pressing. With the produce 10 kg of cheese. Quantities of milk
separation of the curds most of the butterfat and produced in the top 40 milk producing countries
a number of the other constituents are removed. in the world are outlined in Table 2. The top 40
The butterfat usually comprises only 3-4% by cheese producing countries are listed in Table
volume of whole milk. Whey makes up to 80- 3. The whey production from all this milk is
90% of the total volume of milk that enters the considerable and can present a disposal problem.
cheese making process and contains about 50% The challenge to dairy production plants is to
of the nutrients. These nutrients are soluble dispose of whey in an environmentally friendly
whey protein, lactose, vitamins and minerals. and economic manner. The problems facing
Whey from cheese and rennet casein is known whey producers and the uses for whey are
as sweet whey; and the manufacture of acid explored below.
casein yields acid whey. Acid whey is produced
by removal of the fat by centrifugation to make
butter. The milk is then acidified to coagulate Problems with whey
the protein. This is then centrifuged off as casein,
leaving acid whey. Whey produced by dairies has historically been
and remains today a potential pollution problem.
66 J. O’Shea
1
National statistics, Eurostat, 1999.
Whey alcohol - a viable outlet for whey? 67
1
National statistics, Eurostat, 1999.
68 J. O’Shea
Pre-treatment separation
Brown cheese
Whey powder Hydrolysis Fermentation
Concentrated whey
for animal feed
Glucose/galactose Alcohol
syrup Lactic acid
Whey drinks Gas
Whey syrup Penicillin
high humidity in barns housing whey-fed pigs, high quality powder. The free-flowing whey
and respiratory problems are a very real deterrent powder resulting from drying has a number of
to maximizing profits from intensive piggeries. applications in the food industry, including
In addition, the whey supply in many regions is incorporation into dry mixes, baked foods,
seasonal; it dries up between October and April. confections, ice cream and processed cheese.
The pig farmer is therefore vulnerable to these Whey may also be made into cheese such as the
fluctuations in feed supply, and is more likely Scandinavian primost and mysost (Edelman and
to opt for feed sources available all year such as Grodnick, 1986).
barley plus soya. At the same time, the summer Reverse osmosis may also be used to
surplus of whey cannot be utilized by the pig concentrate whey. It works on the same principle
farmer with a fixed number of pig spaces. In as ultrafiltration except that membranes are such
addition, whey is deficient in protein and fat- that only water is allowed to pass through them.
soluble vitamins and requires balancing with Reverse osmosis is cheaper than thermal
appropriate additives. The difficulties recently evaporation for removing up to 70% of the water
experienced by the European pig industry from whey. Savings are therefore possible by
indicate how volatile this industry is and that it using reverse osmosis for pre-concentration
is not a guaranteed ‘sink’ for whey. Many whey before evaporation or membrane processing
producers would prefer to have several relatively (Burgess, 1980).
secure outlets for their whey.
LACTOSE MANUFACTURE
WHEY POWDER
One of the oldest methods of whey utilization is
Whey is processed by several methods. One of the manufacture of lactose. The first stage is
the most common methods is the production of concentration of the whey by low temperature
whey powder by drying. Table 4 outlines whey evaporation, which ensures that the whey protein
powder production in Europe from 1995 to is not damaged. The concentrate, containing 60-
1998. Drying sounds like a simple process, but 65% solids, is fed to a crystallization tank and
a number of stages are necessary to produce a seeded with small crystals of lactose. Edible
1
Eurostat, 1999.
1
Eurostat, 1999.
70 J. O’Shea
lactose is able to absorb flavor; and it is therefore molecular weight of 12,000-13,000 are
used in certain applications in the food industry. extracted, leaving the permeate suitable for
Lactose may also be used as a wine sweetener. fermentation. The main feature of ultrafiltration
For pharmaceutical applications, e.g., tablet is the semi-permeable membrane, usually made
formulation, the lactose crystals must be further of cellulose acetate, polyamide or other
refined to remove traces of protein and minerals. copolymers. The membrane acts as a filter that
Table 5 outlines lactose manufacture in Europe is permeable to the low molecular weight salts
from 1995 to 1998. and lactose, but impermeable to the higher
molecular weight proteins. Pressure is applied
to the whey side of the membrane and water,
LACTOSE HYDROLYSIS salts and lactose are forced through it while
protein is retained. The protein concentration
Lactose is not very sweet compared to cane in the whey therefore increases in proportion to
sugar or even its constituent monosaccharides its reduction in volume and a whey protein
glucose and galactose. The sweetness of lactose concentrate containing up to 60% protein on a
can be considerably increased by hydrolysis. A dry weight basis can be produced. After
number of technologies are available for lactose ultrafiltration, the whey protein concentrate is
hydrolysis. These can be classified as either evaporated to 40-50% total solids by low
enzymatic methods using ß-galactosidase or ion temperature vacuum evaporation and spray-
exchange methods (Burgess, 1980). dried. The whey proteins have a high nutritional
value.
DEMINERALISED WHEY
be viewed as a potential substrate for alcohol process the fermentation time will be between
production. The technology was readily 18 and 24 hrs. At the end of fermentation the
available; and today a number of industrial yeast cells may be separated for re-use. If a yield
plants successfully produce alcohol from whey of 80% is attained from a lactose concentration
permeate. of 4.4%, then production of 1 liter of 100%
Lactose is the major constituent of whey ethanol requires 42 liters of whey. The quality
permeate and accounts for approximately 70% of the alcohol produced is very good with only
of the total solids. It must be stressed that this is very low levels of acetaldehyde and fusel oils.
an average analysis for sweet whey. The type The permeate can be further concentrated; and
of cheese produced will obviously influence the the higher lactose content will increase alcohol
whey composition. Small amounts of other content and thus lower distillation costs. This is
nutritionally desirable materials including very useful where permeate must be transported
riboflavin, amino acids and relatively high levels to alcohol plants.
of calcium and phosphorus are also present in One of the greatest difficulties facing a whey
whey. To obtain the maximum reduction in alcohol plant manager is infection. In certain
BOD, it is necessary to develop a process instances there will be a carryover of lactic acid
whereby practically all the lactose is converted bacteria from the cheese plant. These
to alcohol. microorganisms grow very easily on whey
In practice, a lactose content of 4.3-5.0% is permeate thus reducing the amount of substrate
obtained in the permeate. Using the yeast strain available for alcohol production. Infection in
Kluyveromyces marxianus a yield of 86% is such fermentations is usually controlled by using
attainable from an initial lactose concentration antibiotics.
of 4.3%. This yeast has the ability to ferment Not withstanding such difficulties, utilization
lactose, a disaccharide composed of glucose and of whey as a substrate for alcohol production
galactose (Figure 2). on a worldwide basis has become increasingly
The alcohol concentration following popular. This is evidenced by the diverse
fermentation of permeate is low thus requiring geographical locations of whey alcohol plants,
a very high energy input in the subsequent examples of which are given in Table 7.
distillation. On the other hand, the pollution
problem is greatly decreased with the total solids
(lactose and protein) being reduced to 10% of Considerations relevant to construction
that initially present. of a whey alcohol plant
Biomass produced during the fermentation will
add to the effluent load. Whey pH ranges from EFFLUENT TREATMENT
4.6 to 5.6. The temperature range for
fermentation can be 25-35°C. Using a batch Although ultrafiltration removes the protein from
HOCH2 HOCH2
O O
HO H H
H H
O
OH H OH H
H OH
H OH H OH
α-Lactose
C12H22O11 + H2O → 4C2H5OH + 4CO2
Lactose + water Alcohol + carbon dioxide
Figure 2. Structure of lactose and its conversion to alcohol by Kluyveromyces marxianus.
72 J. O’Shea
whey and the fermentation process converts the was made apparent by two oil crises of the 1970s,
main ingredient (lactose) into alcohol, biomass and the low gasoline octane ratings caused by
is produced. The other ingredients of whey will reduced use of lead after the approval of the
contribute to the BOD/COD of the final effluent Clean Air Act in 1977. Ethanol production from
from the alcohol plant; and adequate effluent corn was seen as a way to boost farm incomes
treatment facilities must be provided. One option caused by the grain surplus in the wake of the
is to use an anaerobic digestion system. This Soviet embargo. Also, addition of ethanol to
will convert a portion of the BOD into gas that gasoline was seen as a means to reduce air
can be used to fuel the boilers in the distillation pollution. These provide a ready market for any
plant. alcohol produced. Within the EU there has been
much discussion about bio-ethanol, but only
recently has there been any commitment to bio-
MARKETS FOR ALCOHOL fuels. Tax concessions for pilot plants producing
bio-fuels have been allowed; and as a
The alcohol markets in the US and Europe are consequence new bio-fuel projects have been
very different. Because of government announced in the Netherlands, Sweden and
intervention in the US, ethanol production has Spain. France has made the most progress. In
grown from an insignificant amount in 1978 to 1996 the French approved a draft law that made
a record 6.4 billion liters in 1998. In Europe the the use of oxygenated components in fuel
situation is very different. There has been no mandatory by 2000.
such government intervention and an
oversupply of alcohol has existed for some time.
Until recently a large portion of this alcohol was Conclusions
sold to Russia, but this market has diminished.
The low price of oil has helped the majority of The attachment of a whey alcohol plant to a dairy
alcohol producers to keep production costs plant producing the appropriate substrate solves
down; however as oil prices rise these production a number of major problems confronting whey
costs will also increase. producers. Alcohol production provides a
In the US, ethanol was promoted as a solution guaranteed outlet for the whey that is not subject
for a variety of complex problems. Among them: to the fortunes of pig markets or other elements
US dependence on foreign oil supplies, which outside their control. Whey production (or
Whey alcohol - a viable outlet for whey? 73
disposal) does not form a limiting step in cheese Eurostat. 1999. Institut National de Statistique,
or casein production. The whey protein Brussells, Belgium, http://europe.eu.int/en/
recovered by ultrafiltration is a valuable item, comm/eurostat/eurostat.html.
the sale of which contributes to process Burgess, K.J. 1980. Uses of waste dairy products.
economics. There is also a universal move Technology Ireland, June, pp. 43-44.
towards bio-fuels, therefore the markets for Murtagh, J.E. 1995. Molasses as a feedstock for
alcohol should be increasingly secure in the alcohol production. In: The Alcohol Textbook
future. (T.P. Lyons, D.R. Kelsall and J.E. Murtagh,
eds). Nottingham University Press,
Nottingham, UK. pp. 27-34.
References New Zealand Dairy.1999. http://www.nzdairy.
Berg, C. 1999. World Ethanol Production and co.nz/public/educational/whey/ethanol/
Trade to 2000 and Beyond. http:// Ethanol~t.htm.
www.distill.com/berg. Nielson, W.K. 1997. Whey processing. Technical
Edelman, E. and S. Grodnick. 1986. The Ideal Bulletin. APV Ireland Ltd, Dublin.
Cheese Book. http://cbs.infoplease.com/ce5/
CE010398.html.
Treatment and fermentation of molasses when making rum-type spirits 75
Chapter 8
ROBERT PIGGOT
Alltech Inc., Nicholasville, Kentucky, USA
crystals is then centrifuged to separate the crystals process for making high test molasses is exactly
and the syrup residue. This first residue, which the same as the production of the sugar, however,
still has a high content of sugar, is referred to as the sugar cane syrup is instead just concentrated,
‘A molasses’. The process is repeated to yield partially inverted using either invertase from
more sugar and subsequent B, C, or even D Saccharomyces cerevisiae or acid (not favored
molasses grades. Sugar mills normally evaporate as it destroys about 5% of sugars) to prevent
and centrifuge a maximum of three times, but crystallization in the final product. Apart from
the number of treatments will depend on the fact that high test molasses is more expensive,
marketplace economics. Obviously successive this product has several advantages over typical
crystallizations reduce the amount of sugar molasses as a feedstock for the production of
present and lessen its value as a fermentable ethanol. It is lower in ash, has fewer dissolved
substrate for ethanol production. Repeated impurities and has a more accurate and controlled
crystallization also tends to degrade molasses sugar profile and sugar content. Typical
sugars to unfermentable compounds (Murtagh, composition is summarized in Table 3.
1999).
Table 4. Comparative chemical composition of beet and molecule. Sucrose is the principal sugar in
cane molasses. molasses. On hydrolysis (reaction with water),
sucrose yields the two monosaccharides (Figure
Cane Beet 1). It should be noted that the molecular weight
Range or typical value (%)
of sucrose is less than the total of its component
monosaccharides because a molecule of water
Total sugars 45-55 48-52 is displaced from sucrose and transferred to the
Sucrose 25 to 35 46-52 simple sugars.
Reducing sugars 20 to 35 0.5-4.0
pH 5-5.5 ~7.0
Ash 10-16 10-12
Non-sugar organic matter 10-12 12-17 INVERT AND REDUCING SUGARS, SUGAR
Starch/polysaccharides 0.5% MEASUREMENTS IN MOLASSES
H2O
C12H22O11 C6H12O6 + C6H12O6
+ water
(MW=18)
Sucrose Glucose + Fructose
(MW=342) (MW=180 + MW=180)
6 α-D-glucose
CH2OH
5
O
H 1
O H H
C 4 1
2 OH H
H C OH OH 3 2 OH
3
HO C H H OH
4
Free anomeric carbon
H C OH (the reducing end of the sugar)
6
5 CH2OH
H C OH 5
O OH
6 H
CH2OH 4 1
OH H
D-glucose (linear) OH 3 2 H
H OH β-D-glucose
fermentation yeast produces an enzyme can occur when molasses has been stored for a
(invertase) that breaks sucrose down into invert long period of time or if it has been contaminated
sugars. with Leuconostoc mesenteroides. While dextrans
Total sugars as invert (TSAI) is a measurement appear as reducing sugars, they are not
commonly used to gauge molasses sugar fermentable. A bad contamination will result in
content. This is either a calculated value or based molasses having long ‘strings’ or a ropey
on reducing power of a sample, which is at best appearance caused by dextran formation. A
a crude measurement. Measuring reducing recent development by Alltech has been use of
power assumes that all the reducing compounds dextranase (used at a rate of 0.01% by weight
in molasses are sugars. Thus any sodium sulfite of molasses) to hydrolyze the dextran chains.
added would appear as sugar in such a test. There Both sucrose and its component mono-
is often up to 5% difference in this test method saccharides are readily fermented by yeast.
over the actual amount of reducing sugars
present. Sugar measurements for molasses are
summarized in Table 5. BRIX VALUES
Term Description/calculation
scale is the same as the Balling scale, which is a the molasses. In addition, carbon dioxide is
measure of what the sugar content of a liquid evolved during molasses degradation. Carbon
would be if all the dissolved and suspended dioxide bubbles rise very slowly in the molasses,
solids were sugar. Expressed another way, it is meaning that density at various levels in a storage
the sugar content of a sugar solution with the tank will be quite different. This makes it
same specific gravity as the sample. Thus, an impossible to calculate with precision from a
80°Brix molasses has a specific gravity of molasses sample taken from a single point in the
1.416, which is the same as a sugar solution tank and a measure of the liquid height to tell
containing 80% sugar by weight. Brix is exactly how much molasses is present in the tank.
measured using a hydrometer originally The way to more accurately measure the amount
intended only for sugar solutions. Unfortunately of molasses present is to take a reading with a
with molasses, the presence of other dissolved device such as a pressure cell that will measure
solids means that Brix readings often bear little the actual weight of the molasses in the tank.
relationship to the amount of sugar actually This will not significantly vary with the amount
present. Despite this, Brix values are quite of entrained air. It also means that production
commonly used in the sugar industry; and procedures are more consistent in that it is
molasses is often defined as around 80°Brix. possible to know exactly how much molasses
Brix values determined by refractometer and has been processed.
hydrometer rarely agree, so the method by
which the reading was obtained should be
clearly stated. VISCOSITY: DILUTING MOLASSES
Table 6. Molasses dilution: dilute 1 ton (or tonne) of 80oBrix molasses to 25oBrix.
Gallons Liters
1415
1410
1405
Density (kg/m3)
1400
1395
Temperature = 20OC
1390 Brix = 80O
Purity = 80%
1385
Density = 1414.1 kg/cm3
1380
1375
20 30 40 50 60 70 80
Temperature (OC)
40000
Viscosity centipoise
30000
20000
10000
0
10 15 20 25 30 35 40
Temperature (°C)
Figure 4. Effects of temperature on the viscosity of cane molasses (R&H Hall, 1999).
Days
3 6 9 12 15
0
-1 60°C
40°C
-2
%
-3
-4
Figure 5. Sugar losses in molasses stored at 60 or 40°C over 15 days (United Molasses).
clarify before fermentation, while others employ dilute the molasses to below 25oBrix. Yeast will
dextranase to ensure maximum extraction of not start fermenting rapidly above this point; and
whatever sugar is present. This is accomplished contamination may develop before the yeast
by diluting the molasses to approximately become established since molasses is laden with
40°Brix, then heating it. Gums and other contaminating bacteria. Some distilleries acidify
suspended solids drop out of the diluted solution. the wash to help reduce infection. Sulfate added
Following heating to pasteurization temperature at this point can come out later as calcium sulfate
for an appropriate time the molasses is cooled scale in the wash column.
and sent to fermentation. The disadvantage of Brix values of about 25 o are the highest
this process is the energy input required as well osmotic pressure that yeast will tolerate to begin
as some loss of sugar. fermentation, however this means a low sugar
Molasses at 80oBrix will not ferment without content and potential low fermentor alcohol
dilution as the sugars and salts exert a very high levels. If sugar content is initially 46%, then
osmotic pressure. It is therefore necessary to dilution to 25°Brix results in only ~14.3% sugar
82 R. Piggot
(Calculated: 25 x 46/80 = 14.3). This results in including variable amounts in the original syrup
about 8% alcohol. This is low by today’s and/or losses during during processing and
standards and results in high hydraulic loading storage. Unless FAN can be tested on a very
and large volumes of effluent. One way to regular basis it is recommended that a sufficient
increase the alcohol content of the fermentation source of FAN is added to ensure good
is by incremental feeding. After the fermentation fermentations. Nitrogen should not be in the
has been going several hours and has dropped form of ammonium sulfate as it will add to the
to 15 or 16°Brix, additional molasses (at scaling problem by forming calcium sulfate.
40°Brix) is added to bring concentration to Likewise, liquid ammonia is undesirable as it
20°Brix and this mixture is then fermented. This tends to raise the pH and encourage bacterial
allows yeast to become rapidly established. contamination unless it is counteracted with acid
additions. (Some plants using liquid ammonia
either use sulfuric acid to balance the pH, which
YEAST AND YEAST CONDITIONING introduces the undesirable sulfate anion, or use
phosphoric acid, which generally introduces
It is important that very high quality yeast are more phosphate than necessary).
added at high cell counts to the fermentation. Urea may be used to supply nitrogen in
Unlike grain fermentations where the sugar or molasses fermentations for fuel ethanol
glucose is produced throughout fermentation, production, but its use for beverage alcohol
all the fermentable substrate in molasses is production should be approached with caution.
present at the beginning of fermentation and it Urea usage may lead to the production of
is equally available to yeast and any undesirable carcinogenic ethyl carbamate, which is un-
contaminants that might be present. In the battle acceptable in alcoholic beverages. If phosphorus
determining what species is to dominate is deficient in the molasses, diammonium
fermentation, healthy yeast are critical. Many phosphate may be added with a corresponding
distilleries condition the yeast before adding it reduction in urea or other nitrogenous nutrient.
to fermentation. This is done by using ~10% of Generally, blackstrap molasses requires no other
the final fermentor volume diluted to about added nutrients for fermentation.
16°Brix. Yeast is added, as is a small amount of The three B vitamins most likely to be essential
air, yeast food (eg, AYF™ at 10-15 ppm) and for the satisfactory growth of any particular
possibly antibiotics to ensure that other strain of Saccharomyces cerevisiae yeast are
organisms are not being grown. A low pH and a biotin, pantothenic acid (pantothenate) and
temperature around 28°C also help keep inositol. Biotin is normally present in great
contamination in check. It is important that excess in cane molasses, but is normally deficient
conditioning tanks are kept clean and that yeast in beet molasses. If a yeast strain that does not
is pumped to the fermentor while still in the require biotin cannot be obtained (they are rare),
exponential growth phase. it may be advantageous to mix in 20% cane
Molasses fermentations are very rapid, often molasses when trying to ferment beet molasses,
completed in 24 hrs, and so much heat is or at least to use a 50:50 cane:beet molasses mix
generated that a good cooling system is very for initial propagation of the yeast.
important. Since many of these fermentation Pantothenate is normally borderline-to-
systems are located in tropical regions where adequate in cane molasses, but is generally in
cooling is limited, a temperature tolerant yeast excess in beet molasses. Mixing some beet
strain can be of great value. Thermosacc™, a molasses with blackstrap cane molasses may be
high temperature yeast (97-100°F) designed advantageous if it is available. Pantothenate
originally for grain distilleries, is now finding concentrations may also be low in HTM and in
application in this industry. refiners cane molasses (the by-product of
refining raw brown sugar to produce white
sugar). In these instances, it may help to add
Yeast nutrition some beet molasses to the mash although one
The free amino nitrogen (FAN) available in may obtain yeasts that do not require external
molasses varies widely for a number of reasons, sources of pantothenate for growth.
Treatment and fermentation of molasses when making rum-type spirits 83
Inositol may be deficient in HTM, but this fermentation is infected with Zymomonas
deficiency is less critical as it is relatively easy mobilis.
to find yeasts that do not require external sources Methanol is not common in molasses
of inositol. fermentation, because molasses contains very
little pectin which is broken down in some
feedstocks during fermentation to form
CONTAMINATION OF MOLASSES methanol. Hence the amount of methanol in
FERMENTATIONS molasses spirit is low.
The principal contaminant of molasses
fermentations is Leuconostoc mesenteroides, Effluent treatment
which causes sucrose molecules to polymerize
into unfermentable dextran chains. If the The liquid effluent from the rum production
contamination is very extensive, the molasses process (known as spent wash, stillage, vinasse
may appear to be ‘ropey’. Dextran may also be or dunder) is a significant environmental
found in refiners molasses produced from raw problem. Biological Oxygen Demand (BOD) is
sugar that has been stockpiled for a number of in the range of 50,000 mg/l and the effluent is
years. Apart from the loss of yield, the presence very dark in color. There is no 100% effective
of dextran increases foaming in fermentors. method of treatment available; and no single
Another important occasional bacterial treatment would suit all distilleries.
contaminant is Zymomonas mobilis. This At present there are six primary ways that
bacteria can ferment sugars to ethanol, but has molasses effluent is treated.
the side effect of reducing sulfur compounds to
produce a hydrogen sulfide smell, which is very 1. Dilution and discharge into the sea
undesirable for production of good quality rum.
2. Land application
3. Deep well injection
Distillation 4. Anaerobic digestion prior to one of the above
There are several issues that are somewhat 5. Lagoon treatment prior to one of the above
unique to the distillation of molasses. Scaling in
the wash column is a serious problem. It is largely 6. Condensation to CMS then used as a cattle
caused by lime used in the clarification process food supplement or incinerated
at the refinery. The most common ways to
combat this are to use a non-fouling tray design
and/or run the column under vacuum to keep DILUTION AND DISCHARGE INTO THE SEA
the temperature down and reduce scale
formation. Additives can be added to reduce This is a controversial method of disposal, but
scaling and special care should be taken to keep is used by several distilleries. It has been studied
the alcohol percentage on the feed tray as low in at least two cases and found not to be harmful
as possible. High strength spirit on this tray will to the environment if done properly. The BOD
increase scale accumulation. dissipates rapidly, but the color change can be
Mud, or sludge, often settles at the bottom of visible for miles from the outlet. This is both
the fermentor. If this occurs, then wash can be visibly unappealing and it also blocks light from
drawn off at a point above the fermentor base. penetrating the water. In areas of the Caribbean
Sludge that has settled in the bottom of the where tourism is a major industry, foul smells
fermentor can be discarded separately. A non- and brown seawater are not acceptable.
fouling tray should be able to handle these solids
and allow them to pass out in the dunder. If not,
then evaporator fouling can occur when LAND APPLICATION
evaporating the stillage to produce condensed
Stillage can be applied to the soil surface and
molasses solubles (CMS).
incorporated into the top 15 to 30 cm in three
Molasses fermentations have a tendency to be
different ways:
high in sulfur. This is especially true if the
84 R. Piggot
Chapter 9
INGE RUSSELL
International Centre for Brewing and Distilling, School of Life Sciences, Heriot-Watt University,
Edinburgh, UK
Introduction
The heart of the fermentation process is the yeast specifically S. cerevisiae unless otherwise noted.
cell. Yeast is the least expensive raw material This is the microorganism used globally for
input into the fermentation process whether it is potable and industrial ethanol production.
for beer, distilled spirits or fuel ethanol, and yet Ethanol is the largest commodity produced by
too often the importance of the yeast is any microorganism; and world ethanol
undervalued. The yeast is ignored, abused and production in calendar year 2003 is expected to
economized. A healthy and well-selected yeast increase to 38 billion liters/year from 34.3 billion
is needed for an efficient fermentation. When in 2002.
selecting the strain, attention must be paid to
whether a yeast is needed that can function at
low temperatures such as for lager beer What is a yeast?
production or at high temperatures for fuel
ethanol production. Whether the environment A common definition is ‘yeast is a fungus where
requires a yeast with the ability to survive high the unicellular form is the most predominant and
ethanol concentrations, or whether the flavor which reproduces by budding (or fission)’. The
compounds that the yeast produces are of interest term yeast is often used synonymously with
or concern. The yeast is an amazingly hardy Saccharomyces cerevisiae since this is the
organism that can survive under many stressful species that is produced in the largest amounts
conditions, but under stress it will not be working for both baking and fermentation purposes.
at its optimum. A careful choice as to the particular Saccharomyces yeast reproduces only by
strain and meticulous attention to growth budding. Yeast are considerably larger in size
conditions will lead to successful fermentations. than bacteria and in terms of quantity and
Traditionally yeast have been characterized economics are the most important group of
according to morphological and physiological microorganisms exploited for industrial
criteria. Taxonomists have grouped together purposes. Over 700 different species of yeast
brewing (both ale and lager), distilling, wine, have been identified by taxonomists. Many are
baking, and numerous wild yeasts, under the poorly characterized, but the ones of industrial
name Saccharomyces cerevisiae. This chapter importance have been well characterized. Indeed
will only deal with Saccharomyces yeast, Saccharomyces, of which there are thousands
86 I. Russell
of unique strains, was one of the first organisms The main parts of a yeast cell
to have its entire genome sequenced and
available on the internet. Yeast cells vary from oval to round and in size
from roughly 5 to 10 µm length to a breadth of
5 to 7 µm, but there are exceptions ranging from
Why is a yeast so effective at 2.5 to 21 µm. Yeast are roughly the size of a red
fermentation? blood cell and many times larger than a bacterial
cell (Figures 1 and 2). The mean cell size varies
Yeast is unique in that it is the only living with the growth cycle stage, the growth
microorganism that can switch from respiration conditions and the age of the cell (older cells
to fermentation. An important fact to remember becoming larger in size). The main components
about yeast is that despite the presence of of the yeast cell are illustrated in Figure 3.
oxygen, yeast will always take the fermentative Under the microscope one can see the yeast
route to utilize glucose if the glucose is there in cell wall along with multiple bud scars and a
high concentration (called the Crabtree effect). birth scar (Figure 4). The cell wall gives shape
The world of yeast is very small and we and stability as well as cell-to-cell recognition.
measure this microscopic universe in terms of It acts as a permeability barrier to large solutes
the Greek letter µ for micron (a micron being and plays a role in controlling the entry of water
10-6 meter). A single yeast cell measuring 7 µm into the cell. The cell wall also protects the yeast
x 7 µm would have a surface area of 153 µm2. cell membrane. The yeast cell wall is composed
Ten grams of pressed yeast would have a contact
surface area of ~10 m2. It is this large contact
surface area that allows yeast to so effectively 0.2 mm Minimum resolvable
take up sugars from the surrounding (200 µm) by human eye
fermentation medium. A useful rule of thumb is
that one gram dryweight of yeast equates to x10
approximately 4.87 x 1010 cells.
CELLS
x10
20 nm
MOLECULES
x10
2 nm
x10
ATOMS
Figure 1. Yeast and bacteria. Figure 2. Relative sizes of yeast and bacteria.
Understanding yeast fundamentals 87
primarily of mannan and glucan (Figure 5). The membrane. The nucleus can be seen using phase
yeast cell plasma membrane is primarily lipids contrast microscopy. The periplasmic space is
and proteins with a small amount of the area between the cell wall and the cell
carbohydrate and acts as the barrier between the membrane. It contains secreted proteins that are
cytoplasm and the outside of the cell. It is very unable to cross the cell wall. For example, some
important as it regulates what goes in and out of enzymes located here help in the catalysis of
the cell (Figure 6). sugars such as sucrose. Sucrose is broken down
The cell nucleus consists primarily of DNA in the periplasmic space by the enzyme invertase,
and protein and is surrounded by a nuclear to fructose and glucose.
Bud
Mitochondrion
Bud vacuole
Secretory
vesicles
Nucleus
Golgi complex
Pore in nuclear membrane
Glycogen granule
Vacuole membrane
Vacuole
Endoplasmic reticulum Ribosomes
Storage granule
1µm
Figure 4. Electron micrograph of a yeast cell with multiple bud scars (and birth scar).
88 I. Russell
Fibrillar
layer
Plasma membrane
Cytoplasm
Figure 5. Structure of the yeast cell wall. Figure 6. Structure of the plasma membrane.
The mitochondria can only be seen using senescence, which leads to cell death. This is
electron microscopy and consist of structures particularly important with respect to recycling
with two distinct membranes, the outer and the yeast, which cannot be done without loss of
inner, and cristae within the mitochondria are some metabolic function. A new yeast cell
formed by the folding of the inner membrane. (daughter or bud) can divide to produce 10 to
Mitochondria are of importance since it is here 33 daughters of its own depending on the strain.
that the enzymes involved in the tricarboxylic Industrial polyploid ale yeast strains have been
acid (TCA or Krebs) cycle and in electron reported to have a lifespan varying from a high
transport and oxidative phosphorylation are of 21.7 +/- 7.5 divisions to a low of 10.3 +/- 4.7
located. Mitochondria contain self-replicating divisions. Respiratory deficient (petite) mutants
DNA and protein synthesis systems. Ribosomes, of these strains were shown to have a reduced
the site of protein synthesis, are located in the lifespan. When examining yeast under the
cytoplasm. Vacuoles are part of the microscope the following are signs of aging: an
intramembranous system, which includes the increase in bud scar number, cell size, surface
endoplasmic reticulum. The size and shape of wrinkles, granularity, and daughter cells being
the vacuoles change during the cell cycle and retained. Aging is accompanied by a progressive
vacuoles are the most obvious component of the impairment of cellular mechanisms that
cell when viewed under the light microscope. eventually results in a yeast with a reduced
The vacuoles store nutrients and provide a capacity to adapt to stress and to ferment (Smart,
location for the breakdown of macromolecules 2000). Figure 7 shows the progression through
such as protein. Lipid granules are found in the the lifespan cycle of a yeast cell.
cytoplasm, as are high concentrations of
glycogen. Glycogen can be visualized under the
light microscope by staining the cells with iodine. The chemical composition of yeast
A yeast cell consists of 75% water and 25% dry
The lifespan of yeast matter. The dry matter consists of:
Specific requirements for yeast growth include: Water is a main component of all living material
and most microorganisms need at least 15% water
90 I. Russell
Table 1. Approximate composition of industrially produced aerobically grown baker’s and anaerobically grown
brewer’s yeast (expressed as g/kg dry yeast) (from Ingledew, 1999).
Cell composition (g/kg dry yeast) Baker’s yeast (aerobically grown) Brewer’s yeast (anaerobically grown)
Carbohydrate 180-440 390-600
Protein 380-590 370-420
Ash 45-75 73-81
Nucleic acid (DNA and RNA) 52-95 39-43
Lipids 40-50
P 10-19 14-20
S 3.9
K 20-21 17
Na 0.12-0.3 0.7
Ca 0.6-0.75 1.3
Fe 0.02-0.03 0.1
Mg 1.3-1.65 2.3
Co 0.008 0.0002
Cu 0.008 0.033
Mn 0.0059 0.0057
Zn 0.170-0.197 0.0387
Cr 0.0005-0.0025
Mn 0.008
Ni 0.003
Sn 0.003
Mo 0.00004
Li 0.000017
V 0.00004
Se 0.005
Pantothenate (Coenzyme A) 0.065-0.10 0.110-0.120
Choline (membranes) 2.71-4.00 3.80-4.55
Thiamin (Vit B1) 0.090 -0.165 0.092-0.150
Riboflavin (Vit B2) 0.040-0.100 0.035-0.045
Nicotinic acid/niacin (NAD) 0.30-0.585 0.450
Pyridoxine (Vit B6) 0.020-0.040 0.043-0.050
Biotin (biocytin) 0.0006-0.0013 0.001
p-aminobenzoate (folic) 0.160 0.005
Inositol (phospholipids) 3.0 3.9-5.0
Folic acid (1-C transfer) 0.013-0.015 0.010
Malony, 1998; Reed and Nagodawithana, 1991; Ingledew et al., 1977; Patel and Ingledew, 1973; Peppler, 1970.
content to grow. Yeasts prefer to grow at acidic in that form. Most brewing and distilling strains
pH. Water is discussed in greater detail in the utilize sucrose, glucose, fructose, maltose and
chapter by Ingledew in this volume. maltotriose in this approximate sequence,
although some degree of overlap does occur.
The initial step in the utilization of any sugar by
A CARBON SOURCE (SUGARS AND STARCHES) yeast is either its hydrolysis outside the
membrane (e.g. sucrose), followed by entry into
Low molecular weight sugars are the preferred the cell by some or all of the hydrolysis products
carbohydrate source of Saccharomyces. or its intact passage across the cell membrane
Fermentable carbohydrates include the (e.g. maltose). Sucrose is the first sugar to
monosaccharides glucose, fructose, mannose disappear from the wort. The hydrolysis
and galactose as well as the disaccharides products, glucose and fructose, are then taken
maltose and sucrose, and the trisaccharides, into the cell by facilitated diffusion (Figure 8).
raffinose and maltotriose (uptake is strain This is a passive uptake and both sugars are
dependent). Large polysaccharides such as directly incorporated into the glycolytic pathway
starch and cellulose cannot be taken up by yeast
Understanding yeast fundamentals 91
MALTOTRIOSE
MALTOSE
permease
permease
Maltotriose
Maltose
α-glucosidase α-glucosidase
Glucose
Fructose
permease
permease
glucoamylase invertase
immediately after entry into the cell (Figure 9). Figure 10 illustrates the typical sugar utilization
For the sugar sucrose (a disaccharide consisting pattern in a laboratory whisky mash fermentation
of a glucose molecule and a fructose molecule) where only the enzymes from the malt are
the enzyme invertase, which is located between available to break down the sugars and starches.
the cell membrane and the cell wall, hydrolyses A mash from a fuel ethanol fermentation would
the sucrose to its two constituents, glucose and have commercial enzymes added to break the
fructose, which then enter the cell by passive starch into smaller sugars that can be taken up
uptake. The hydrolysis of sucrose by the yeast’s by the yeast cell.
invertase is very rapid and so on a sugar profile
the sucrose disappears very quickly from the
substrate. NITROGEN SOURCE
Maltose (two linked glucose molecules) and
maltotriose (three linked glucose molecules) pass Yeast growth requires the uptake of nitrogen for
intact across the cell membrane by an active the synthesis of protein and other nitrogenous
process, and once inside are hydrolyzed to glucose components of the cell, but yeasts can only utilize
units by the α-glucosidase system. Maltotriose low molecular weight nitrogenous materials such
fermentation tends to begin later than that of maltose as inorganic ammonium ion, urea, amino acids
in a batch fermentation. Maltotetraose (four linked and small peptides. The yeast cannot take up
glucose molecules), higher polysaccharides proteins or break down peptides larger than
(dextrins) and pentoses are not fermented by tripeptides. The extracellular proteolytic activity
regular ethanol strains. If a very high glucose of yeast is negligible. Proteolysis does not occur
concentration relative to the other sugars is in the fermentation, unless the yeast has
present, one can encounter ‘hung’ fermentations, autolysed. Nitrogen uptake slows or ceases later
due to a condition called glucose repression. Here in the fermentation as yeast multiplication stops.
the high concentrations of glucose prevent the Nitrogen is discussed in more detail later in this
yeast from synthesizing the enzymes needed to chapter.
assimilate the other sugars present.
92 I. Russell
Hexose
Glucose Glucose
transporter
ATP
Fructose HXK
Glucose-6-phosphate
Fructose PGI
ATP Fructose-6-phosphate
HXK ATP
PFK
Fructose 1,6-diphosphate
Plasma membrane
FBA
ADH PGM
2-phosphoglycerate
Acetaldehyde
ENO
PYK
CO2 CO2
Pyruvate
Conversion of sugars to ethanol Once sugars are inside the cell, they are
converted via the glycolytic pathway (also called
Yeast is able to utilize sugars in the presence of Embden-Myerhof-Parnas or EMP pathway) into
oxygen (aerobically) and when oxygen is pyruvate. The sequence of enzyme catalyzed
excluded (anaerobically). Aerobic breakdown reactions that oxidatively convert glucose to
(catabolism) produces more energy and is called pyruvic acid in the yeast cytoplasm is known as
respiration. Anaerobic catabolism produces less glycolysis (Figure 9).
energy and is called fermentation. The very simplest expression for fermentation
is the following equation:
Understanding yeast fundamentals 93
14 Maltotriose Maltose
Glucose Glycerol
12
0
0 10 20 30 40 50
Fermentation time (h)
Mol wt 180 2 x 44 2 x 46
This equation, named after the French scientist cell yield is approximately 10% of what is
Gay-Lussac, shows that glucose yields almost possible under aeration and non-repressing
equal parts of carbon dioxide and ethanol as well glucose concentrations.
as the production of energy. Part of the energy Not all glucose is metabolized by the glycolytic
is used for cell metabolism and some of it will (EMP) pathway described above. Some of the
be lost as heat. What this equation does not sugar will be metabolized by a route called the
address is that some of the sugar will be used hexose monophosphate pathway. This pathway
for yeast growth and that there are other also yields energy, but more importantly it yields
metabolites produced (glycerol, lactic and pentose sugars, which are important in the
succinic acid, etc.). These other metabolites are synthesis of yeast nucleotides and yeast nucleic
small in quantity compared to the amount of acids.
ethanol produced.
The glycolytic (EMP) pathway operates in the
presence or the absence of oxygen to convert From glucose to pyruvate: glycolysis in
glucose to pyruvic acid, energy and reduced yeast
nicotinamide adenine dinucleotide (NADH +
H +). When oxygen is abundant and the sugar Glucose is phosphorylated in two stages. Two
level is kept low, little or no ethanol is made by ATPs are used to produce fructose 1,6
the growing yeast. The yeast uses all of the sugar disphosphate, which is then split by the enzyme
as energy for new yeast cell production (i.e. the aldolase to form two 3-carbon triose phosphates.
baker’s yeast propagation process). However Inorganic phosphate is assimilated to form two
when oxygen is reduced and/or the glucose triose diphosphates from which four H atoms
levels exceed 0.1% (w/v), then ethanol is made. are accepted by two molecules of oxidized NAD.
Under low oxygen conditions less than 5% of Lastly, four ATPs are formed by transfer of
the sugar is metabolized for growth and the total phosphate from the triose diphosphates to ADP
94 I. Russell
resulting in the formation of two molecules of that of non-growing cells. Hence for maximum
pyruvic acid. Some of the pyruvic acid and other ethanol production, best practice is to keep the
intermediates are used by the yeast cell through cells growing as long as possible at the highest
a variety of metabolic pathways as ‘building possible cell concentration. When a yeast is
blocks’ for new yeast cells and some glycerol is transferred into a suitable fermentation medium
made from the intermediate, dihydroxyacetone. under optimal growth conditions, the growth
However most of the pyruvic acid is pattern illustrated in Figure 13 is seen.
immediately converted to ethanol and carbon
dioxide (CO2) (Figures 11 and 12). Lag Exponential Stationary
200 phase growth phase phase
There are two main carriers of energy in living Figure 13. Typical batch growth curve for yeast.
cells, Adenosine TriPhosphate (ATP) and
Nicotinamide Adenine Dinucleotide (NAD +).
ATP is a carrier of chemical energy in the form CELL GROWTH PHASES
of high energy phosphate bonds. NAD + is a
carrier of hydrogen and electrons and is involved Lag phase. During this phase the yeast adapts
in many oxidation-reduction reactions in the to its new environment and activates its
cell. It can pick up and transport 2e- and 2H+. metabolism (i.e., synthesizing enzymes). It is a
The cell takes its energy source, converts it into period of zero growth but a period of intense
NADH and ATP, and then uses these carriers to biochemical activity as the yeast cell adjusts from
perform needed tasks in the cell. NAD+ and ATP the previous culture conditions to the conditions
are not the only carriers of energy, but they are in the fresh media. This phase ends with the first
the major carriers. cell division.
NADH
Triose phosphates
Pyruvate
Pyruvate CO2
Succinate Succinate
Figure 11. Aerobic and anaerobic catabolic pathways for sugar in yeast.
O2 NADH
NADH Acetaldehyde
Lactate NAD+
2 ATP
generated
Krebs (TCA)
Cycle Ethanol
2 ATP
generated
36 ATP
generated
harvested using a rotary vacuum filter. The atmosphere of inert gas, usually nitrogen. Dried
circumference of the filter is a continuous strip yeast (92–96% dry weight) has a long shelf life
of fabric coated with food-grade starch. As the and does not require cold storage (but can be
drum rotates slowly through the culture in the chilled for added security). It is very stable at
trough, the spent culture medium is drawn under room temperature and even more stable (over a
vacuum through the hollow spokes. The yeast year) if stored cool and under nitrogen gas. It
is retained on the filter, scraped off and loses ~7% viability/month at room temperature
packaged. and <10% year at 4°C under nitrogen gas. The
active dry yeast usually has a viable cell count
in the range of ~2.2 x 10 10 cells/g. It has an
YEAST SLURRY indigenous bacterial count. The process to make
active dry yeast involves extruding the
Yeast in the slurry form is used primarily by compressed yeast into spaghetti-like strands that
brewers. They can recycle their yeast at a high are then dried in air lift dryers (fluidized bed).
viability (>90%). The slurry must be kept cool Reactivation (conditioning) of dried yeast at
(4°C) and has a 3-7 day shelf life. If the yeast is pitching is carried out to prevent loss of viability.
fresh it can be acid-washed to lower the bacterial It is important that rehydration is carried out
count. The slurry usually has a cell count in the under the prescribed conditions and once
range of ~3 x 109 cells/mL at 14-18% total solids. rehydrated the yeast must be used immediately.
The availability of a shelf-stable dried yeast
inoculum where each gram contains ~2.2 x 1010
COMPRESSED YEAST viable yeasts, is a major biotechnological
Compressed yeast is more stable than slurry and advance. The yeast, which has been grown
requires cooling. The yeast is stored at 3–4 oC under extreme aeration for yield, is easier to store,
and has a 30 day shelf life. It loses viability at handle and transport. Selected strains for
~5-10 % a week. It can be rehydrated easily and individual processes are available (i.e. beverage
has an indigenous bacterial count. The versus fuel). The use of active dry yeast has
compressed yeast usually has a yeast count in reduced the need for yeast propagation expertise
the range of ~6-13 x 109 cells/mL at 33% total in the plants and increased the predictability of
solids. fermentations. Active dry yeasts are used in
Compressed yeast is prepared at 4°C by adding distilling and fuel alcohol plants to inoculate
salt to the yeast slurry and then passing the yeast fermentors to recommended values of 1.0-2.0
through a cloth press or through a rotary drum million viable cells/mL per degree Plato/Brix of
vacuum filter. The resultant yeast has the wort/mash.
consistency of butter and is then extruded into
blocks, packaged, and both stored and shipped
CONDITIONING (ACTIVATING) THE YEAST
at 4°C. The yeast is sold as 25 kg bags of
compressed moist yeast (24–33% dry weight) It is usually recommended that the active dry
or in bulk as cream yeast slurry of about 18% yeast is conditioned (according to the different
dry weight. Before using compressed yeast it manufacturer’s instructions) to get optimal
must first be slurried aseptically in a sterilized growth. (This is different from propagating a
yeast mixing vessel to provide the required starter culture with the aim of growing up large
consistency for pitching. Cream yeast, which is quantities of yeast from a small lab culture.) For
delivered, stored and pitched in bulk, is often example, one manufacturer suggests a
used for automated large-scale operations. conditioning regime for active dry yeast for fuel
ethanol plants where a yeast conditioning tank
with a 16–18° Plato/Brix mash and a 1-2%
ACTIVE DRY YEAST (ADY) glucose level is used. The recommended
Active dry yeast is used by most batch ethanol conditioning tank is 7–10% the size of the
plants. It is also used to start continuous plants fermentor and the goal is 250-300 million cells/g.
and to supplement fermentors. This yeast is dried Temperature and nutrition at this stage are critical
under partial vacuum at 45–55 o C in an to ensure optimal yeast growth.
98 I. Russell
For distilled beverages, dried yeast can be contamination is feared primarily due to the off
added directly to the fermentation vessel or pre- flavors and odors that can be produced in
mixed to a slurry with water. For direct addition to fermentation problems for the culture
inoculation, the wort cooler is adjusted to yeast. For non-beverage alcohol the concern is
produce wort at 38 o C. The warm wort is loss of alcohol due to non-utilization of the
collected to a depth of 10 cm in the washback sugars for ethanol production. The production
and the required amount of dried yeast is of excretion products such as acetic acid by wild
sprinkled onto the wort, taking care to avoid yeast such as Dekkera (also called Brettanomyces)
clumping. After 5 min. the heat exchanger is can inhibit the production yeast culture and
adjusted to the normal set temperature, (~20oC) cause serious fermentation problems in a plant.
and wort collection is completed. It is important to be familiar with the normal
For the pre-mixing method, sterile water or appearance of the production yeast under the
weak wort, 10 times the volume of dried yeast, microscope and to be constantly on the watch
is brought to 38oC in a rehydration tank and yeast for deviations from the standard. Wild yeast will
is added with stirring, to prevent clumping. After often (but not always) have a different
mixing for 5 min. the yeast slurry is pumped appearance and exhibit very differently sized
into the washback at normal set temperature. The cells from the normal population. The cells can
pre-mixing method uses the same equipment as vary from elongated cells, to lemon shaped cells
the preparation of a slurry from bags of pressed to oval cells. There may be buds with elongated
yeast, but for dried yeast the higher temperature necks or buds at odd angles. There are a number
of mixing is critical (Campbell, 2003b). of specialized media that can be utilized to
It should be noted that there are many different isolate and identify contaminating yeast cells
methods of conditioning the yeast and it will (see Campbell, 2003a). In order to protect
depend on the particular plant, the instructions against wild yeast infections it is important to
with the particular type of yeast purchased and bring in a fresh culture yeast on a regular basis
the expertise in the plant. The objectives of yeast and ensure that there is a good cleaning and
propagation/conditioning are to deliver to a sanitation program in place. Once a wild yeast
fermentor a large volume of yeast of high infection takes hold in a plant, that same
viability, high budding in the log phase of growth contaminant often comes back numerous times
and with a very low level of infection. and causes more infections. An extra-thorough
There are many methods of propagating yeast cleaning and evaluation of problem spots in the
(batch, continuous) all with pros and cons plant is often necessary to rid the plant of the
depending on the final use of the ethanol i.e. problem. Acid-washing of cropped yeast from
beverage or fuel, the unique characteristics of a fermentation is ineffective against wild yeasts
the plant, the technical skill in the plant, etc. The and many of the wild yeast are more acid-tolerant
yeast propagator is the heart of a distillery; and than the culture yeast. Washing will however
it is important to remember the key objectives reduce the number of bacteria present.
that are critical to the performance of a
fermentation. Hence cost savings in this area are
not advised since they are usually minimal when KILLER YEAST
compared to the cost of production per gallon
of ethanol. The loss of ethanol production that Some strains of Saccharomyces secrete a
is possible when an aged yeast is used can be in proteinaceous toxin called a zymocide or killer
the range of 1% alcohol, which can lead to a toxin that is lethal to certain other strains of
loss of $0.10 per gallon produced. Infection Saccharomyces. Toxin-producing strains are
resident in a plant can cause a drop of 3% alcohol termed killer yeasts and susceptible strains are
and a potential loss of $0.30 per gallon. termed sensitive yeasts. There are strains that do
not kill and are not themselves killed, and these
are called resistant. The killer character of
Wild yeast contamination Saccharomyces is determined by the presence
of two cytoplasmically located dsRNA plasmids
In the beverage alcohol industry wild yeast (termed M and L). A wild yeast containing ‘killer
Understanding yeast fundamentals 99
factor’ can have a negative impact in a beverage Yeasts purchased from ale breweries have a
or fuel ethanol plant by annihilating the maximum growth temperature of 37.5-39.8°C
production strain and reducing the yield. while lager yeasts have a maximum growth
temperature of 31.6-34.0°C. For example, if
lager yeast is used as part of the inoculum in a
Temperature whisky fermentation, they assist the distilling
yeast only in the early part of the fermentation.
Thermal damage to a yeast cell leads to a general Some Saccharomyces strains can survive 35-
denaturation of the yeast’s proteins and nucleic 43°C under ideal conditions, but not under
acids. A yeast cell cannot regulate its temperature stressed conditions. It is important to remember
and all fermentation processes produce heat. The that the growth optimum is species dependent
higher the temperature above the cell’s optimal as well as growth condition dependent. Figure
temperature for growth, the greater the loss of 14 illustrates the effect of temperature on three
cell viability. Yeast thermotolerance is at its different commercial yeast strains fermenting a
maximum when the external pH declines to 4.0. 19.3º Plato/Brix whisky mash at three controlled
There is a general correlation between growth temperatures. This figure illustrates the
rate and stress sensitivity. Cells growing quickly importance of selecting a strain that can function
in a glucose rich medium (the desired state for optimally at the temperature of the fermentation.
rapid ethanol production) are more sensitive to All fermentation processes produce heat and the
heat and other stresses than cells that are in amount of cooling available and cost of cooling
stationary phase. need to be taken into consideration when
Many yeasts are capable of growth over a wide selecting a strain. Certain strains sold
range from 5°C to 35°C . Commercial distilling commercially have been constructed for their
yeast strains can ferment well at 32-35°C, but at ability to withstand the stress of high ethanol
higher temperatures metabolic activity rapidly and high temperature and are known to function
declines. The optimal temperature for well under these difficult conditions (Figure 15).
reproduction is usually lower and around 28°C.
8
Ethanol (% v/v)
0
Strain A Strain B Strain C
Figure 14. Effect of temperature on three different commercial yeast strains fermenting a 19.3º Plato/Brix whisky mash. Bars
indicate final ethanol concentration at 48 hrs of fermentation. (Mash is 85% corn and 15% malted barley).
100 I. Russell
16 Superstart™ Thermosacc™
14
12
Alcohol (%)
10
8
6
4
2
0
90°F 100.4°F 104°F
32.2°C 38°C 40°C
Temperature
Figure 15. Ethanol production by two commercial yeast strains at three temperatures.
HEAT SHOCK AND COLD SHOCK kg /4000 L) or 5 x 106 to 2 x 107 viable cells. A
brewing rule of thumb is one million cells per
Yeast is sensitive to sudden temperature changes degree Plato/Brix. So a 12o Plato/Brix wort or
and shows signs of shock on sudden cooling mash would be pitched at 12 x 106 cells and a
with adverse effects on fermentation and cell 20 o Plato/Brix wort/mash at 2 x 10 7 cells.
multiplication. Cold shocking a yeast strain will Pitching the yeast at the correct level ensures
inhibit budding and induce vacuolar reduced infection and improves the process
rearrangements. Yeasts with elevated levels of efficiency thus reducing overall costs.
trehalose are better able to survive cold storage. Viability involves the cell’s ability to bud and
Yeast cells exposed to an elevated temperature grow, however slowly. Vitality is defined as a
will exhibit a molecular response called “the heat continuum of activity of the cell from very active
shock response” where certain proteins are to not active at all. Vitality tests are related to
synthesized which aid the yeast in responding yeast growth, yeast physiological changes and
to stress. fermentation performance.
Measurement of percentage viability alone is
insufficient to guarantee the quality of the
Pitching and viability pitching yeast and hence the concept of yeast
vitality (i.e. the ability of the yeast to ferment
To follow the progress of a fermentation it is quickly and efficiently rather than simply being
common practice to measure the quantity of alive) is of interest. Various methods have been
yeast biomass or to count the number of cells. suggested as indicators of viability and vitality.
For pitching (inoculating) on a production scale, Heggart et al. (1999) have extensively reviewed
yeast is measured by mass: either directly as the the factors that affect yeast vitality and viability.
number of 25 kg bags of yeast required, or In terms of measuring yeast health, to date no
indirectly as the volume of yeast slurry of one test exists that is accurate, simple,
standard concentration by weight. On a inexpensive and rapid. It is often not possible to
laboratory scale, counting the number of yeast rely on a single method and a combination of
cells per ml is usually employed. This requires methods is often employed. The evaluation of
a microscope and an engraved counting yeast viability includes methods based on cell
chamber slide called a haemocytometer replication (standard plate count and slide count
(originally designed to count blood cells). Yeast technique), brightfield and fluorescent staining,
is normally inoculated at a final concentration measurement of intracellular ATP and NADH,
of ~1-2 lb active dry yeast/1000 US gallons (1
Understanding yeast fundamentals 101
and other methods such as electrokinetic the colorless reduced form by the metabolic
potential and capacitance. Vitality assessment activity of the cells. However, on death, the cells
techniques include methods that measure become readily permeable to methylene blue,
metabolic activity (staining, microcalorimetry, and, since dead cells do not metabolize, the cells
vicinal diketone reduction, yeast protease remain blue. There are a number of other redox
activity, magnesium ion release, specific oxygen dyes and variations on the makeup of the
uptake rate, acidification power and intracellular methylene blue dye buffers that have been
pH), cellular components (ATP, adenylate energy suggested, but unfortunately these dyes are also
charge, NADH, glycogen and trehalose, sterols unreliable with yeasts of lower viability than
and unsaturated fatty acids) and fermentative about 90% viable (Smart, 1999). Although the
capacity (glycolytic flux rates, CO2 measurement dye method has many shortcomings it does give
and short fermentation tests). A combination of a very fast evaluation of the status of a culture
various methods is often used to obtain an as long as one remembers that if more than 10%
accurate picture of a yeast’s fermentative of the cells stain blue, the viability of the culture
capability. is in fact much lower. In other words the dye
One newer method that has been widely can be used to confidently confirm that a culture
adopted in the brewing industry to assess yeast is very healthy (>90% viable) but is very
viability is capacitance measurement. When an inaccurate when a culture has a high number of
appropriate radio frequency signal is applied to dead cells in it.
a yeast suspension, the dielectrical nature of the To see if a culture is capable of reproducing
cell membrane permits a buildup of charge. (i.e. can the cell bud and eventually form a
Depending on the number of yeast present colony) a plate count can be carried out. When
within the field, the resulting capacitance can compared to a plate count, at viabilities above
be measured and is proportional to the radius 90% the staining method consistently tends to
and volume fraction of cells present. Since yeast overestimate cell viability by not more than 10–
cells are relatively constant in size, capacitance 15%. When viability drops below 90–95% (as
can be related to cell number and biomass. Dead measured by a plate count), methylene blue
cells with disrupted membranes do not generate counts can give values lower than 50%. At a
a capacitance when a radio frequency is applied. plate viability of 0% (i.e. no growth on a plate),
Capacitance probes (Aber Instruments Ltd. methylene blue may indicate apparent viabilities
Aberystwyth, UK) capable of providing an as high as 30–40% (Heggart et al., 2000).
accurate measurement of viable yeast in real time A well propagated yeast should be close to
have been successfully adopted by a number of 100% viability, and should retain that value for
breweries worldwide (Austin et al., 1994). several weeks if stored at 4oC. Storage at ambient
Heggart et al., (2000) have extensively temperature is not recommended as it leads to a
reviewed the various techniques available for loss of viability. It also allows the rapid growth
measuring yeast viability and vitality and the of bacteria from the inevitable initial low level
reader is directed to this review paper for further of resident contamination. This low level of
information on the many techniques available. contamination can only be avoided if the yeast
culture is propagated under pharmaceutical
conditions. The extra expense for pharmaceutical
METHYLENE BLUE: THE MOST POPULAR conditions would make the product price
VIABILITY TEST prohibitive. Chilled yeast from an anaerobic
Methylene blue has been in use since the 1920s fermentation does not have the storage stability
and is recognized as a simple and standard of an aerobically grown yeast.
method in the fermentation industry for
measuring yeast viability. The buffered dye is
added to a suspension of cells and examined BUDDING INDEX
under the microscope (ASBC, 1980). Methylene
Cell counts that measure the number of budding
blue is taken up only weakly, if at all, by living
yeast cells are often employed to determine if
cells, and any taken up is rapidly converted to
the yeasts are in the active desired growth phase
102 I. Russell
since this is when ethanol is produced most No significant loss of spirit yield was observed
rapidly. The yeasts are examined under the when the lactic acid bacteria grew only at the
microscope with the use of a hemacytometer, end of the fermentation on the nutrients from
and the number of budding cells counted to the autolysed yeast (Dolan, 1976; Barbour and
obtain a budding index. A budding index of Priest, 1988). Changes in pH also affect the
~30% is desirable as this indicates that the cells activity of α- and β-amylases and limit
are in the rapid phase of growth. dextrinase, and consequently the hydrolysis of
polysaccharides during the fermentation. The
low pH associated with heavy early contamination
pH by lactic acid bacteria reduces spirit yield if the
enzymes cannot function well at that pH.
Yeast prefers an acid pH and its optimum pH is
5.0–5.2, but brewing and distilling strains are
capable of good growth over the pH range of Yeast metabolism
approximately 3.5 to 6.0.
During any fermentation, H+ ions are excreted Yeast produces a myriad of products. Table 2
by the yeast, and this results in a pH decline in summarizes some of the main products.
the media. For example, an all malt wort pitched
with a pure culture of distiller’s yeast will have
an initial pH of ~5.2-5.5, which will fall to pH GLYCEROL
4.0–4.5 (depending on the solids), and then
increase slightly during the stationary phase due Glycerol is quantitatively the next most important
to release of amino acids from autolysing yeast product of yeast alcoholic fermentation after
cells. In fermentations showing late growth of ethanol and carbon dioxide. It is an important
lactic acid bacteria, these bacteria can use the compound in that it helps to maintain the cell’s
yeast autolysate as a nutrient and further lower redox balance. When a yeast cell is growing,
the pH to ~3.8. Contamination by lactic acid removal of pyruvate for cell biosynthesis can
bacteria early in the fermentation is undesirable lead to a buildup of NADH, which can halt
and reduced spirit yield results since the sugars catabolism. To avoid this the cell reduces
used by the bacteria are then not available to dihydroxyacetone phosphate to glycerol
the yeast for the production of ethanol. The loss phosphate, which is then dephosphorylated to
of spirit yield in malt distilleries due to growth glycerol. The cell then excretes the glycerol into
of Lactobacillus from the start of fermentation the growth medium (Figure 16). Redox balance
is reported as follows: and energy metabolism are highly integrated and
the biosynthesis of new cellular material results
Up to 106 bacteria /ml <1% loss in the production of a surplus of reducing
1–10 x 106 bacteria/ml 1–3% loss equivalents of NADH. This is especially a
1–10 x 107 bacteria/ml 3–5% loss problem under anaerobic conditions and
Above 108 bacteria/ml >5% loss Saccharomyces relies on glycerol production to
(Campbell, 2003b)
Understanding yeast fundamentals 103
re-oxidize the surplus of NADH formed during Acrolein is produced from glycerol (Figure 17).
anaerobic conditions. However, during aerobic It produces an intensely bitter taste when it
conditions, the surplus of reducing equivalents combines with phenolic compounds. Certain
(as NADH) in the cytoplasm is delivered to the contaminating lactic acid bacteria in a
respiratory chain in the mitochondria. fermentation can degrade the glycerol present
The yeast must maintain a redox balance and via a dehydratase enzyme to form 3-hydroxy-
glycerol carries out a redox balancing role, but propionaldehyde, which under acidic conditions
it also plays a role as an osmoprotectant. The spontaneously degrades slowly to acrolein in
stress of high osmotic pressure and heat shock storage. The process occurs more rapidly at
both enhance glycerol production. Glycerol higher temperatures.
accumulation inside the cell is very important
for the survival of strains during osmotic stress.
An increase in the fermentation temperature SUCCINIC AND OTHER ORGANIC ACIDS
causes the yeast cell to produce larger quantities
of glycerol. In a fuel ethanol fermentation, Organic acids, depending on the carbon chain
glycerol levels range from ~1.2–1.5%, again length, can impart different flavors and aromas
depending on a plethora of environmental to beverages ranging from acidic to rancid to
factors. cheesy. Succinic acid is a main secondary end
In Germany in 1914, in order to produce large product of alcoholic fermentation. It is believed
amounts of glycerol for nitroglycerine to be synthesized and secreted by yeasts either
production, this yeast pathway was exploited. following limited operation of the citric acid
Using ‘steered fermentation’ acetaldehyde (Kreb’s or TCA) cycle or by reductive pathways
breakdown to ethanol was blocked by the involving some citric acid cycle enzymes. Low
addition of sodium bisulphite, and the glycolytic concentrations of pyruvic, malic, fumaric,
intermediates were reduced to give glycerol as oxaloacetic, citric, α-ketoglutaric, glutamic,
the major fermentation product. propionic, lactic and acetic acids are also present
as fermentation products. These are produced
as intermediates of the citric acid (TCA) cycle
ACROLEIN (Figure 18).
The production of both glycerol and organic
Acrolein has a very peppery, lacrimating and acids is greatly affected by the pH as illustrated
irritating effect on workers operating a still. in Table 3.
Pyruvate
NADH CO2
Amino
acids Acetyl CoA Lipids
Oxaloacetate
Malate Citrate
NADH
Amino
Fumarate cis-aconitate
acids
FADH2
Succinate Isocitrate
NADH
GTP NADH
Oxalosuccinate
Succinyl CoA CO2
α-ketoglutarate
CO2
Figure 18. The citric acid cycle (also called Kreb’s or TCA) for the synthesis of intermediates and energy
under aerobic conditions.
OTHER COMPOUNDS
SULFUR METABOLISM
During fermentation the carbonyl compound
The biosynthesis of S-containing amino acids found in highest concentration is acetaldehyde.
and reduction of sulfate salts in the wort/mash It is formed during fermentation and is a
are the key sulfur metabolism areas. In whisky metabolic branch point in the pathway leading
distilling there are benefits from the remedial from carbohydrate to ethanol. The acetaldehyde
effects of copper in the stills, but with the normal formed may either be reduced to ethanol, or
ratio of copper surface area to pot still volume it oxidized to acetic acid, and in the final step of
is not possible (or desirable) to remove all sulfur alcoholic fermentation, acetaldehyde is reduced
Understanding yeast fundamentals 105
biotin and many require pantothenate. A corn With the exception of zinc, grain and corn based
or grain mash should contain sufficient amounts fermentation media are not usually deficient in
of vitamins and they are not normally added these ions, but it is important to remember that
directly to address a deficiency other than differences in the fermentation substrate can
through the use of a yeast food. If deficiencies cause large differences in metal content. A proper
occur there can be resultant fermentation ratio between calcium and magnesium positively
problems. When active dry yeast is used, since influences fermentation rates. An imbalance in
the yeast only multiplies a few generations, inorganic nutrition is often reflected in complex
vitamin deficiencies are not usually encountered and often subtle alterations of metabolic patterns
unless yeast recycling is practiced. Again, it must and growth characteristics. The role played by
be stressed that each mash has unique many of these ionic species is both enzymatic
characteristics that must be considered when and structural. The macro elements potassium,
yeast nutrition is an issue. magnesium, calcium, iron, zinc, and manganese
are required at concentrations between 0.1 and
INORGANIC IONS 1.0 mM. The micro elements cobalt, boron,
cadmium, chromium, copper, iodine,
Yeast requires a number of inorganic ions for molybdenum, nickel and vanadium are required
optimum growth and fermentation and this between 0.1 and 100 mM. Minerals such as
requirement varies with i) the yeast strain silver, arsenic, barium, mercury, lithium, nickel,
employed, ii) growth media differences, and iii) osmium, lead, selenium and tellurium can be
interactions with other constituents especially inhibitory when concentrations exceed 100 µM.
interactions among the trace metals. Important Table 6 lists the effects of metal ions on yeast
cations are zinc, manganese, magnesium, viability and vitality. For additional information
calcium, copper, potassium and iron (Table 5). on the effect of metal ions such as magnesium
1
Adapted from Jones et al. (1981); Jones and Greenfield (1984); Heggart et al. (1999).
Understanding yeast fundamentals 107
on ethanol tolerance in yeast, the reader is the cell for assimilation into sulfur containing
referred to the research work of Walker (1999). amino acids such as methionine and tripeptides
such as glutathione (glutamic acid-cysteine-
Phosphorus glycine). In the presence of excess sulfate, yeast
can store sulfur intracellularly in the form of
Phosphorus is essential to yeast cells for glutathione, which can account for as much as
incorporation into structural molecules [e.g. 1% of the cellular dry weight. Under conditions
phosphomannan and phospholipids, nucleic of sulfate limitation or starvation, the glutathione
acids (DNA and RNA)] and phosphorylated in the yeast can act as a sulfur source for the
metabolites (e.g. ATP and glucose-6-phosphate). biosynthesis of amino acids. Sulfur is required
Yeast stores polyphosphates as an energy at ~0.3–0.5% of the weight of expected yeast
reserve and uses this material during times of mass per volume of medium.
starvation. Phosphorus is commonly available
in the form of inorganic orthophosphate Magnesium
(H 2 PO 4- ) and this is rapidly metabolized to
nucleoside triphosphate (ATP) on entry into the Magnesium is the most abundant intracellular
yeast cell. It can be added in the monobasic form divalent cation in yeast and it acts primarily as
or in the form of ammonium, sodium or an enzyme cofactor. It exerts a protective effect
potassium salts. A rule of thumb is that available on yeast cultures subjected to stress conditions
phosphorus should be present at ~1-2% of the such as temperature and osmotic pressure, as
cell dry weight per unit volume of the medium. well as playing a role in alcohol tolerance.
Magnesium stimulates fermentation during the
metabolism of high gravity worts/mashes. The
Sulfur requirement for magnesium is small and is
Inorganic sulfur, in the form of sulfate anions, usually amply provided by the raw materials and
is transported via an active process by yeast into water. Exogenous magnesium exerts a protective
108 I. Russell
effect on yeast cells subjected to physical and significance of Ca2+ uptake in yeast lies in the
chemical stresses (Walker, 1999). multifunctional role of this cation as a modulator
of growth and metabolic responses. Calcium also
Zinc plays an important role in flocculation.
Trace levels of Zn2+ are essential for yeast growth Copper and iron
and other metal ions cannot fill this requirement.
Zinc is a cofactor in a number of enzymatic Copper (Cu) and iron (Fe) ions are essential
reactions within the cell and the important nutrients for yeast and act as cofactors in several
fermentation enzyme, alcohol dehydrogenase, enzymes including the redox pigments of the
is a zinc metalloenzyme. Certain yeast strains respiratory chain. Copper is an essential
may require more zinc than occurs naturally in micronutrient at low concentrations, but is toxic
wort/mash, and supplementation with a yeast at higher concentrations. Strains of yeast differ
food is used. Zinc deficiency results in low yields in their sensitivity to copper and negative effects
of yeast (can inhibit budding) and slow on fermentation can be seen starting at
fermentations. A concentration of 0.3 ppm Zn concentrations of >10 ppm. Iron is usually
in the medium is considered adequate and higher abundantly provided for in brewer’s wort where
concentrations of zinc (1-5 ppm) may inhibit the average content is 0.1 ppm.
yeast activity, but this depends, at least in part,
on the level of manganese present.
HIGHER ALCOHOL PRODUCTION
Potassium
From a quantitative point of view, the most
Potassium is the most prevalent cation in the important higher alcohols (fusel oils) produced
yeast cytoplasm (~150 mM). Uptake of K + in are n-propanol, amyl alcohol, isoamyl alcohol
yeast requires an active transport mechanism and and phenethyl alcohol. In addition, over 40 other
the presence of glucose or a fermentable or alcohols have been identified. Regulation of the
respirable substrate. Transport of K+ is associated biosynthesis of higher alcohols is complex since
with the transfer of H+ ions from within the cell. they may be produced as by-products of amino
acid catabolism or via pyruvate derived from
Sodium carbohydrate metabolism (Figure 19).
The catabolic route (i.e. Ehrlich pathway)
Yeast cells do not accumulate Na + ions under (catabolic defined as a biochemical process in
normal growth conditions and continuously which organic compounds are digested, usually
excrete Na + in order to maintain very low an energy-liberating process) involves a pathway
cytoplasmic concentrations of this cation. In the in which the keto acid produced from an amino
presence of high salt concentrations, the yeast acid transamination is decarboxylated to the
cells osmoregulate by producing intracellular corresponding aldehyde, then reduced to the
compatible solutes such as glycerol and arabitol. alcohol via an NAD-linked dehydrogenase
(Figure 20). In this way, for example, isobutanol
Manganese may be produced from valine, 3-methyl-1-
Manganese (Mn) is essential for yeast growth butanol from leucine and 2-methyl-1-butanol
and metabolism and in trace levels acts as an from isoleucine. They are by-products of the
intracellular regulator of key enzymes. amino acid assimilation pathway.
Manganese ions have been reported to stimulate The anabolic route of de novo synthesis (i.e.
fermentation. biosynthesis pathway) (anabolic defined as a
biochemical process involving the synthesis of
Calcium
organic compounds, usually an energy-utilizing
process) utilizes the same pathways as those
Calcium (Ca) stimulates yeast growth but it is involved in the biosynthesis of amino acids. As
not a growth requirement. It is involved in in the catabolic route, the keto acid intermediate
membrane structure and function. Yeast cells is decarboxylated and the resultant aldehyde
maintain cytosolic Ca2+ at very low levels. The reduced to the alcohol.
Understanding yeast fundamentals 109
Isobutanol
Isobutanol
α-keto α-keto
acid acid
Glucose
Valine
Glucose
Figure 19. Higher alcohols (fusel oils) are produced by yeast metabolism via two routes both of which utilize α-keto acid
(adapted from Walker, 1999).
Figure 20. Higher alcohol production. Conversion of the amino acid leucine to isoamyl alcohol.
The relative contribution made by the two increased yields of higher alcohols. When
routes varies with individual higher alcohols. assimilable free amino nitrogen is increased in
Since there is no corresponding amino acid, the the mash, the anabolic (biosynthesis) pathway
anabolic route appears to be the sole mechanism becomes less dominant and the Ehrlich pathway
for the formation of n-propanol. takes over. If ammonium ion or urea is the main
The concentration of amino acids in the nitrogen source, higher alcohols are made via
medium is important. If there are only low levels the anabolic (synthetic) pathway.
of assimilable nitrogen then the anabolic route
is predominant. High concentrations of wort ESTERS
amino nitrogen favor the catabolic pathway due
to feedback inhibition of the amino acid Esters are a product of yeast metabolism and
synthetic pathways by their respective products. there are over 100 distinct esters identified in
Both the level and the composition of the yeast- fermentation beverages. The most abundant
assimilable nitrogen in the mash influence the ester is ethyl acetate (fruity/solvent). Other esters
formation of higher alcohols. Low levels of produced include isoamyl acetate (banana/
usable nitrogen result in less yeast growth and apple), isobutyl acetate (banana/fruity), ethyl
110 I. Russell
concentration of oxygen at 100% air saturation Active dry yeasts (ADY), which are grown
is inversely proportional to the specific gravity under conditions of high aeration, have a much
(and the temperature), thus an air saturated wort higher unsaturated fatty acid and sterol content
of 10° Plato/Brix would contain 8.5 ppm O 2, than yeasts propagated or conditioned ‘in house’
whereas a wort of 17° Plato/Brix (specific in an ethanol fermentation plant. This high
gravity 1.070) would only contain 7.9 ppm O2. ‘credit’ of sterols and unsaturated fatty acids
Oxygen solubility is also lower in mashes poised allows the active dry yeasts to complete a
at >35°C that contain increased amounts of fermentation without the same oxygen (or sterol/
solids. fatty acid) requirement in the fermentation broth.
Mash is fermented under anaerobic conditions It is important to note that these yeasts are never
and the very small amount of air (8-20 ppm O2) reused. Distillery practices do not allow for
that should be supplied around the time of initial harvesting.
inoculation can be added by ‘rousing’ the cooled Many distilleries and fuel alcohol plants use
mash using air or bottled oxygen, or by ‘splash active dry yeasts. Aeration through ‘splash
filling’ the fermentor just prior to adding the filling’ of batch fermenters or circulation of mash
yeast. This small amount of oxygen, if used through heat exchangers for cooling purposes
aerobically, would only be enough to allow is probably adequate if active dry yeast is used.
respiration of 0.008% (w/v) glucose so there is Aeration at the heat exchanger should be done
no large negative effect on sugar utilization. on the cold side of the exchanger to ensure
Since there is excessive glucose present in the maximum solubility of oxygen on injection
mash, the oxygen is probably used anaerobically. (remembering that oxygen is more soluble at
Oxygenating at low levels at the initiation of lower temperatures).
fermentation has been carried out for many years
by the brewing industry and the most effective Yeast sterols and oxygen
time to ensure oxygen availability is when the
yeast begins to actively grow. The amount of Sterols in aerobically-grown yeast may reach 1%
oxygen is important, but the timing of the supply total dry weight, but with anaerobic growth and
is also key. The oxygen must be supplied during no usable sterols in the medium, they dilute out
the growth and division phase of the yeast. In to a value of less than 0.1%. There are several
distillery practice, it may be sufficient to splash sterols found in yeast of which ergosterol is the
fill tankage just prior to fermentation, especially most prominent. Sterols play a key role in
if active dry (aerobically grown) yeasts are used. regulating membrane fluidity. When insufficient
Clear substrates without insoluble lipid oxygen is available for membrane synthesis,
materials present may require more dissolved yeast cells fail to grow and loss of membrane
oxygen. For example, with high gravity brewer’s integrity results in cell death. The regulation of
worts that have been filtered and are almost sterol synthesis is extremely complex. Yeast cells
devoid of insolubles, bottled oxygen is used to accumulate sterols into the membrane until they
oxygenate. The amount of oxygen needed by have met bulk requirement needs. After this,
yeast growing anaerobically is small, but additional sterols are esterified to long chain fatty
important, and deprivation leads to fermentation acids and stored in intracellular lipid vesicles.
difficulties. The most abundant sterol of the sterol pool is
Reproductive growth eventually ceases when zymosterol. In an aerobic culture these sterol
the limiting value of sterols is reached. When esters accumulate during the stationary phase
cell multiplication stops, the rapid uptake of of growth and when these cells are re-inoculated
nitrogen and carbohydrates is affected and cells into fresh medium these steryl esters are then
have a reduced ability to complete the hydrolyzed and incorporated into the growing
fermentation. In addition they are more cell membranes.
susceptible to external stresses such as Medium length fatty acids (C6–C10) arise via
temperature and pH. This has an impact on the the activity of fatty acid synthetase as
rate of ethanol production. The limiting level of intermediates in the formation of longer chain
unsaturated fatty acids for brewing yeast is length fatty acids, which are incorporated into
~0.5%. the various classes of yeast lipids. The fatty acid
112 I. Russell
biosynthetic pathways lead to the saturated and Toward the later stages of fermentation, the yeast
unsaturated fatty acids used in membrane lipids. restores its reserve of glycogen.
Unless oxygen is present in small quantities
when growth begins, the desaturation steps do
not proceed and unsaturated fatty acids and TREHALOSE
sterols cannot be synthesized by the yeast cell.
Trehalose is a non-reducing disaccharide
consisting of two glucose units linked together
GLYCOGEN by an α-1, 1-glycosidic bond (Figure 22).
Trehalose is present in particularly high
Glycogen is the yeast’s major reserve carbohydrate concentrations in resting and stressed yeast cells.
and it is similar in formation and structure to During sporulation trehalose is virtually the only
plant amylopectin. It is a polymer of α–D- sugar present in the yeast spores. Trehalose plays
glucose with a molecular weight of ~108. It has a protective role in osmoregulation. It protects
a branched structure containing chains of 10– cells during conditions of nutrient depletion,
14 residues of α–D-glucose joined by 1→4 against toxic chemicals (ethanol, oxygen
linkages. Individual chains are connected by radicals, heavy metals) and starvation. It also
(1→6) α–D-glucosidic linkages (Figure 21). plays a role in improving cell resistance to high
Glycogen serves as a store of biochemical and low temperatures as well as resistance to
energy during the lag phase of fermentation dessication. This protective role may be due to
when the energy demand is intense for the the stabilising effect of trehalose on cell
synthesis of such compounds as sterols and fatty membranes. It is believed that the induction of
acids. This intracellular source of glucose fuels heat shock proteins and the accumulation of
lipid synthesis at the same time as oxygen is trehalose are related. Trehalose accumulation
available to the cell. Dissimilation of glycogen occurs during late stationary phase or in response
and the synthesis of lipid are both rapid. The to environmental changes. Trehalose levels are
hydrolysis of glycogen from approximately 20- important in the preparation of active dried yeast
30% to 5% and the corresponding production since trehalose acts as a protectant during the
of lipid from 5% to 11.5% of the cell dry weight drying phase.
occurs within the first few hours after pitching.
O O O O
H H H H H H H H
OH H OH H OH H OH H
O O O O
OH
H OH H OH H OH H OH α-(1-6) linkage
CH2
CH2OH CH2OH CH2OH
O O O O
H H H H H H H H
OH H OH H OH H OH H
O O O O O
H OH H OH H OH H OH
Figure 21. Structure of glycogen: a high molecular weight polymer with branched-chain structure composed
of D-glucopyranose residues.
Understanding yeast fundamentals 113
Table 7. FAN levels of typical mashes (mg/L mash*). nitrogen nutrient would be needed in a
fermentation. If yeasts are ~6% N and we wish
Total FAN Useable FAN to grow 6 mg/mL of new yeast (~300 x 106 cells),
Wheat 82 64 then 0.36 mg/mL of usable N is needed. Thus if
Barley 84 62 the source of N is urea, 770 mg/L is needed. If
Hulless barley 124 100 the source is NH3, 440 mg/L is needed and if
Oats 193 159 the source is diammonium phosphate (DAP),
Hulless oats 184 130 1700 mg/L is needed. Tables 8 and 9 above
Rye 103 83 illustrate the differences in useable FAN
Molasses 267 141 depending on the carbohydrate source and the
Corn 70 58
use of backset and steepwater.
Starch slurry ~0 ~0
Table 8 illustrates the FAN levels in a mash with Yeast growth increases almost linearly with FAN
backset from a corn dry milling plant and Table levels up to 100 mg/L. Very high amounts of
9 is a corn wet milling example illustrating FAN FAN (up to 878 mg/L) have been shown to have
levels with backset and steep water. no additional effect on yeast growth, however
they did stimulate the fermentation rate.
Table 8. FAN in mash and backset: corn dry milling C 3.72 H 6.11 O 195 N 0.61 is the formula proposed
example1. for the major elements in yeast cell mass by
Harrison (1967) and from this one can calculate
Usage level FAN FAN contribution the nitrogen content of a cell. This is of the order
(L/1000 L) (mg/L) (mg/L) of 9% (w/w). The number is generally accurate
Backset 200 75 15*
for aerobically-grown yeast, but is an
Mash ~800 ~88.4 70.7 overestimation for yeast from anaerobic
total 85.7 fermentations.
The level of protein inside a yeast is a function
*quality questioned of the concentration of usable nitrogen
1
Ingledew, 1999 (ammonium salts, urea, amino acids, di- or
tripeptides and nitrogen derived from nucleic
acid breakdown) (Figure 23).
Table 9. FAN in backset and steepwater: corn wet
milling example1.
Cell membrane
ATP
CO2
Urea Urea ADP
Pi
CO2 Allophanate
Figure 23. Nitrogen uptake by the yeast cell (adapted from Slaughter, 2003).
For beverage alcohol the main source of Although ammonia is a preferred nitrogen
nitrogen for yeast growth is amino acids formed source for catabolic reactions, the cell will take
from the proteolysis of grain proteins. The wort/ up certain amino acids first for direct
mash contains 19 amino acids. The yeast takes incorporation into proteins. The absorption of
up the amino acids in an orderly manner with the amino acids by the yeast and their rate of
different amino acids removed at various points assimilation will vary both with the strain of
in the fermentation cycle (Table 10). Proline, the yeast and the amino acid composition of the wort/
most abundant amino acid in wort/mash, has mash. Some amino acids such as glycine and
scarcely been assimilated by the end of lysine are readily taken up by the yeast.
fermentation, whereas over 95% of the other However, under nitrogen limiting conditions
amino acids have been used by the yeast. these two amino acids are inhibitory to growth
There are a number of permeases for the uptake and to fermentation. When usable nitrogen (such
of wort amino acids, some specific for individual as other amino acids) is in excess, the glycine
amino acids and one that is a general amino acid and lysine are not inhibitory. A balanced nitrogen
permease (GAP) with broad substrate specificity. source (i.e. a mixture of amino acids) is more
The uptake and regulatory mechanisms are efficient in providing nitrogen to the yeast for
complex. use in the various biochemical pathways than a
single nitrogen source.
116 I. Russell
Table 10. Classes of wort amino acids in order of assimilation during fermentation1.
1
Pierce, 1987.
It is worth noting that surface culture enzymes Stress tolerance during the fermentation of high
such as Rhizozyme™ contain enough natural gravity media has been shown to be very strain
protease activity to make further additions dependent. Although the issues of yeast nutrition
unecessary. in high gravity wort can be addressed by careful
selection of strains and fermentation conditions,
failure to address them can result in serious
Endogenous enzymes fermentation problems. Ingledew and coworkers
have published extensively on the opportunities
In malt and grain distilleries, the mash, as distinct for very high gravity fermentation (VHG) and
from the mash in a brewing operation, still has have extended the upper limits by successfully
residual enzyme activity and a high proportion producing 23.8% (v/v) ethanol in a batch
of the cereal starch is converted to fermentable fermentation (Ingledew, 1999; Wang et al.,
sugars throughout the fermentation by 1999a,b).
continuing starch hydrolysis. In fuel ethanol
fermentations, the addition of exogenous
enzymes to break down starches performs the Stress factors and synergism
same function throughout the fermentation.
Stress factors such as ethanol, temperature, and
acids can significantly impact the vitality and
High gravity fermentation viability of the yeast. These factors can act
synergistically and overpower the ability of the
High gravity fermentation employs worts/ yeast to recover from stress. The list of stress
mashes at higher than normal concentration. factors is long and yeast has an amazing ability
High gravity fermentation is progressively being to recover from them – but there is a limit! It is
introduced around the world and yeast nutrition important to remember that yeast is the heart of
must be addressed when the change is made the fermentation process and if reasonable care
from regular gravity to high gravity. Sufficient and attention are given to ensure its health the
oxygenation, higher nitrogen levels and higher payback will be huge.
yeast addition are the key success factors.
Increased osmotic pressure and ethanol levels ‘Select and treat your yeast well and it will work
can affect the yeast and the stress tolerance of well for you!’
the yeast becomes a much more important factor.
20 500
Yeast cells (million/g mash)
Dissolved solids (g/100 ml)
400
15
Protease
300
10
Control
200
5 Control
100
Protease
0 0
0 50 100 150 0 50 100 150
Figure 24. Benefits of protease addition to an alcohol fermentation (Thomas and Ingledew, 1990, by permission).
118 I. Russell
Reed, G. and T.W. Nagodawithana. 1991. In: Biotechnology of Malting and Brewing. 1991.
Yeast Technology, 2 nd Edition. AVI. Van J.S. Hough, Cambridge University Press.
Nostrand Reinhold, New York. Brewing Microbiology. 2002. F.G. Priest and I.
Slaughter, J.C. 2003. The biochemistry and Campbell. (eds). Kluwer Academic.
physiology of yeast growth. In: Brewing Beers and Coolers. 1994. Manfred Moll.
Microbiology, 3rd edition. Kluwer Plenum, Springer-Verlag, New York.
New York. pp. 19-62. Brewing Yeast and Fermentation. 2001. C.
Smart, K.A., K.M. Chambers, I. Lambert and C. Boulton and D. Quain. Blackwell Science.
Jenkins.1999. Use of methylene violet staining Brewing Yeast Fermentation Performance. 2000.
procedures to determine yeast viability and 1st Edition. K. Smart (ed). Blackwell Science.
vitality. J. Am. Soc. Brew. Chem. 57(1):18-23. Brewing Yeast Fermentation Performance. 2003.
Smart, K. 2000. The death of the yeast cell. In: 2nd Edition. K. Smart (ed). Blackwell Science.
Brewing Yeast Fermentation Performance. Handbook of Brewing. 1994. W. A. Hardwick.
Blackwell Science, United Kingdom, pp. 105- Marcel Dekker.
113. Standards of Brewing. 2002. C.W. Bamforth.
Thomas, K.C. and W.M. Ingledew. 1990. Fuel Brewers Publications.
alcohol production: effects of free amino Technology - Brewing and Malting, 2nd Edition.
nitrogen on fermentation of very high gravity 1999. W. Kunze. Verlag der VLB Berlin.
wheat mashes. Appl. Environ. Microb. Available from the VLB at www.vlb-berlin.org/
56:2046-2050. english/books/kunze/.
Walker, G. 1999. Yeast Physiology and The Alcohol Textbook: A Reference for the
Biotechnology. John Wiley and Sons Ltd. Beverage, Fuel and Industrial Alcohol
United Kingdom. Industries. 1999. K.A. Jacques, T.P. Lyons and
Wang, S., W.M. Ingledew, K.C. Thomas, K. D.R. Kelsall (eds). Nottingham University
Sosulski and F.W. Sosulski. 1999a. Press, Nottingham, UK.
Optimization of fermentation temperature and The Practical Brewer [MBAA], 3rd Edition. 1999.
mash specific gravity for fuel alcohol Available from the Master Brewers Association
production. Cereal Chem. 76:82-86. of the Americas at www.mbaa.com.
Wang, S., K.C. Thomas, K. Sosulski, W.M. Whisky: Technology, Production and Marketing.
Ingledew and F.W. Sosulski. 1999b.Grain 2003. I. Russell (ed). Academic Press, New
pearling and the very high gravity (VHG) York.
fermentation technologies for fuel alcohol Yeast Physiology and Biotechnology. 1998. G.
production from rye and triticale. Process Walker. John Wiley and Sons Ltd., New York.
Biochemistry. 34(5): 421-428.
Chapter 10
Introduction
Fermentation is the critical step in a distillery. It liquefied mash from cooking (1-2% glucose) to
is here that yeast converts sugar to ethanol; and ensure that pre-formed sugar and other bound
it is here that contaminating microorganisms find sugars (starch and other polysaccharides) are
an opportunity to divert the process to lactic acid, released in a timely fashion for maximum
glycerol, acetic acid, etc. The fermentation step conversion to ethanol by yeast.
(and to a much lesser degree the cooking
process) is also where yeast must be ‘coached’
to produce maximum levels of ethanol, Choosing a yeast strain
sometimes as high as 23%. Fermentation will
also have a direct bearing on downstream One of the most important decisions a distiller
processing. If sugar levels remaining in beer are must make is the selection of yeast strain. The
too high, evaporative capacity may be impaired best characterized yeast is Saccharomyces
or syrup contents increased. Both can lead to cerevisiae, but within this species there are
distillers grains being more difficult to dry and thousands of unique strains. Which one to use?
altered in color as Maillard reactions occur Several decisions must be made (Table 1). The
between proteins and sugars. Truly the first decision is whether or not to grow the strain
fermentation step is the heart of the distillery. in-house. In fact, yeast strain maintenance and
When reflecting on the overall flow diagram growth is a job best left to the experts. For the
of the distillery (Figure 1) the central role of yeast very small cost per gallon it represents (less than
is obvious. The process of converting sugars to 0.5 cents), there is just too much at stake.
ethanol takes between 10 and 60 hrs during
which heat is generated. The various factors Table 1. Choices among yeast strains.
affecting the efficiency of that process need to
be considered. In the chapters by Russell and • Grow in-house?
• Active dried or pressed?
Ingledew, the biochemistry of ethanol production
• High alcohol tolerance needed? Thermosacc™
by yeast is covered in detail. In this chapter we • Lower alcohol in fermentor: Superstart™
will discuss the factors involved in taking
122 D.R. Kelsall and T.P. Lyons
Milling
Starch Ethanol
(32 lbs) (17.6 lbs)
+
Liquefaction DDGS
Saccharification (17 lbs)
+
Fermentable Heat
sugar Fermentation (7450 BTU)
Recovery +
(36 lbs) Dryhouse CO2
(18.4 lbs)
the microorganism that competes with yeast for maintain good yeast growth and maximum
glucose. Heat also increases the impact of other alcohol productivity while avoiding high
stress factors such as mycotoxins and salt levels. temperature peaks. Glucoamylase (Allcoholase
II L400) provides glucose at the start of
Combining enzyme activities to control sugar fermentation while the Rhizozyme™, with pH
delivery to yeast and temperature optimums more closely aligned
to fermentation conditions, continues to work
How can we best control fermentation during fermentation. The net result is a steady,
temperature? The easiest way to reduce slow release of glucose and a steady formation
fermentor temperature would be to reduce the of alcohol by yeast.
sugar level going in and therefore reduce yeast RhizozymeTM also affects ethanol yield as it is
growth; however, less alcohol would be capable of hydrolyzing starch that is
produced. Since the objective is more, not less unconverted in the cook process and also
alcohol, this is not an option. However, sugar converts some of the cellulose in the corn to
level, or more importantly sugar type, is glucose.
important. The key is to consider cooking and
Yeast must be maintained in a growth fermentation together. Cooking must be
(budding) phase, since their ability to produce designed to release as much starch, both bound
alcohol is more than 30 times greater during and unbound, as possible without overloading
growth than in non-growth mode. Yeast however the yeast with glucose. It must also be recognized
can respond to high glucose levels in two ways. that corn has, in addition to starch, other
Cell growth is either inhibited by high glucose potentially useful carbohydrates including
levels or yeast may grow rapidly and then stop. cellulose and hemicellulose. Conventional
In either case there is an over-dosing effect on glucoamylases do not degrade either, whereas
fermentation efficiency and possibly a spike in glucoamylase produced by surface culture
temperature. Glucose must be ‘spoon-fed’ to the fermentation (Koji) also contains a range of
yeast; and for this reason it is recommended that carbohydrase enzymes as ‘side activities’ that
glucoamylases be used both sparingly and in degrade some of the cellulose thereby releasing
conjunction with an enzyme produced in surface more sugar, which boosts yield (Table 2,
culture called Rhizozyme™. The combination Figure 4).
provides ideally balanced sugar delivery to
20
Ethanol (% v/v)
15
10
0
Glucoamylase Rhizozyme
Figure 4. Comparison of alcohol yields from high solids corn mash with traditional and surface culture glucoamylases.
Practical management of yeast: conversion of sugars to ethanol 125
Table 2. Comparison of surface culture and vessel or the entire vessel. Recirculating
conventional glucoamylase. coolant through the jacket allows
fermentation temperatures to be monitored.
Production Surface culture Conventional
(Koji) fermentation deep tank
If jackets are spaced out over the fermentor
liquid sides, then cooling can commence as the
fermentor is filled.
Activities d) Heat exchanger with recirculation. Contents
Glucoamylase 40 400
of the fermentor are pumped through a heat
Cellulase 400 —
Hemicellulase 350 —
exchanger (typically spiral or plate-and-
α-amylase 5000 — frame) and returned to the fermentor. This
Phytase 300 has the added advantage of acting as a means
of keeping the fermentor agitated; and
pH profile Broad Narrow
Temperature
nutrients (sterol donors, peptides, oxygen,
optimum, °C 30-40 0-60 etc.) can be introduced at critical times. It is
Activity on suggested that optimum control of
gelatinized starch Rapid Slow temperature can be obtained by placing the
Activity on raw heat exchangers in such a position that
starch Reasonable Slow recirculation can start at the beginning of the
Yeast stimulation Significant Poor fill when the yeast starts producing heat.
Modeling provides a prediction of the increase Of the four types, the cooling jacket is best for
in yield to be expected. Predicted yields of 3.1 minimizing infection, but the heat exchanger is
gallons per bushel, 17-18% higher than the the most effective at heat control. As the distiller
standard 2.65, have been calculated based on moves towards very high levels of ethanol (over
increased sugar release. This is 18 gallons (>65 20% by volume), heat recovery and maintenance
liters) more per tonne of grain. The of low temperatures will become even more
recommendation is to use a 50/50 blend of important as temperature rises of 6-8°C will
Rhizozyme™ Koji and conventional become the norm. A word of caution: some
glucoamylase to maximize yield. Where distillers distilleries, particularly in continuous operations,
can reliably measure yield, the results in terms share external heat exchangers between
of costs per gallon reveal optimum returns. fermentors in order to economize. Based on the
experience of many existing distilleries using this
Equipment for heat control sharing concept, infection can easily be
transferred among fermentors as well as having
Despite the sequential feeding of glucose, heat inadequate cooling.
is generated during fermentation and must be
removed. Four types of heat exchangers are
typically used; and these are illustrated later in INFECTION: THE SECOND STRESS FACTOR
this chapter when fermentation design is
discussed. The various types of infection are described in
a) Internal cooling coils. These are the least detail elsewhere in this book, however several
desirable option because they are practically points are relevant in this context. Lactobacilli
impossible to clean and sterilize. consume glucose to produce lactic acid and
acetic acid, which is the second major factor
b) Internal cooling panels. Like cooling coils, affecting the yield of alcohol in fermentation,
panels are suspended in the fermentation which in turn has a major impact on distillery
vessel. Also like cooling coils, cooling is economics (Table 3). Yeast do produce some
adequate but sterilization may be difficult. organic acids during fermentation, but
c) External cooling jacket. Cooling jackets, concentrations are relatively low compared to
often dimpled for added contact with the those produced by lactobacilli and other
fermentation, can cover either part of the contaminating bacteria. As a general rule of
126 D.R. Kelsall and T.P. Lyons
thumb, if the titratable acidity of an uninoculated (typically 50-70%), the yeast uses the sugar as
mash is X, then the titratable acidity of the it becomes available and competitively excludes
fermented beer is usually about 1.5X. This is a lactic and acetic acid-producing bacteria by
general rule and seems to work quite well in reducing available sugar. Equally, when yeast
practice, although it is not based on known growth rate begins to decline (typically at 20-
scientific principles. The 1.5X represents the 24 hrs) sterol donors and oxygen, which at that
titratable acidity of the finished beer when there stage have become limiting, will revive activity
is no significant contamination from lactobacilli. by ensuring that fatty acids are available for
When lactobacilli are active, the generation of yeast reproduction.
lactic and acetic acids substantially increases the
titratable acidity and often the high acid content Table 4. Practical means of controlling infection.
will cause the yeast fermentation to stop or Avoid deadlegs in lines
dramatically slow down. Very early into an Avoid sharing fermentor heat exchangers
infection the titratable acidity reaches 2X. Proper cleaning programs
Periodic use of de-scaling chemicals
Practical means of controlling infection Control fermentation temperatures
Avoid the saccharification step in cooking
The most fundamental way to control infection Maximize yeast growth (budding)
is to control the fermentation environment (Table Use yeast yeast conditioning
4). Fermentors must be cleaned on a regular Use sterol donors and peptides as yeast foods in yeast
basis; and designs must avoid sharing of heat conditioning
Antimicrobials
exchangers and dead legs in piping. Fermentors AllpenTM (against Gram-(+) organisms)
should be designed to facilitate complete LactosideTM (against Gram-(+) and Gram-(-) organisms)
emptying. Scaling will occur regardless of the
fermentor design, and should be removed using
either a scale inhibitor (such as Scale-BanTM) or Regardless of precautions, a distillery is not a
a de-scaling chemical (an acid wash). sterile environment and opportunities for growth
Fermentations must be fast; and yeast should be of both Gram (+) and Gram (-) bacteria exist.
added either in a form that will quickly Because their growth rate is so much faster than
rehydrate or preferrably one that has been yeast, strict controls to prevent infection must
‘started’ in a conditioning tank. Saccharification be in place. As early as 1954 penicillin was used
tanks can be a source of infection as in ethanol fermentation; and owing to its quick
lactobacillus strains can adapt and grow at the inactivation caused no problems with regulatory
temperatures used in this process. The goal is to agencies. Other antibiotics such as
make conditions as suitable as possible for yeast virginiamycin are effective antimicrobials, but
growth (without jeopardizing alcohol yield) and some are now known to suppress yeast
as unsuitable as possible for infectious fermentation and in addition can leave residues
microorganisms. By restricting glucose release (2.6%) in spent grains. With the ban in Europe
in fermentation by using Rhizozyme™ and against use of virginiamycin in products destined
lower levels of glucose-releasing glucoamylase for animal feeds, caution is urged. Furthermore,
Practical management of yeast: conversion of sugars to ethanol 127
80 Adaptation
to mash
MAXIMIZE YEAST PERFORMANCE BY
40 Maximum Decline CONDITIONING
growth phase
10
Yeast is the powerhouse of any distillery; and
without healthy yeast, alcohol percentages in the
12 24 36 48 60 72 fermentor and alcohol yield will drop.
Time (hours)
Conditioning tanks are a way to ensure good
cell numbers in the fermentor (Table 5). At the
Figure 5. Typical yeast growth curve in distillery mash. start of batch fermentation there should be a
128 D.R. Kelsall and T.P. Lyons
of other feedstock-related stress factors such as industry has largely converted to rapid batch
mycotoxins or phytin content will also become fermentation with cylindro-conical or sloping
more important as we move toward higher bottom fermentors, or to cascade continuous
ethanol production. fermentation using cylindro-conical fermentors.
BATCH FERMENTATION
Fermentation systems used in distilleries
A typical cylindro-conical fermentor with skirt
Before considering the specific management
support is illustrated in Figure 8. These
tools available to the distiller, it is necessary to
fermentors are usually designed to ferment from
consider the differences among the fermentation
30,000 gallons to 750,000 gallons. Fermentors
systems in use. Over the last 30 years the
need to be fabricated on site if the diameter and
brewing and distilling industries have developed
length exceed a size that can be safely
new fermenting systems; and the distilling
transported by road.
Vent
20 in. Manway/
sampling port
CIP spray head
D A
Coolant
Agitator C
(if required)
Different cooling systems are available. Many and to use a conditioning tank so that the main
distilleries use chilled water to cool the fermentation starts immediately after the yeast
fermentors. From a microbiological standpoint is added.
the most desirable cooling system by far uses There is a choice of materials to use in the
external cooling jackets. The least desirable manufacture of fermentors. Stainless steel is
option employs internal cooling coils because generally the best choice as it is easier to clean
they are practically impossible to clean and and sterilize. It also lasts much longer than mild
sterilize from the top-mounted CIP (clean-in- steel or lined vessels; and when considered over
place) spray nozzles. The external recirculation the expected lifetime of a fermentor it is by far
heat exchanger is frequently used in distilleries the most economical option. If chloride levels
and is sometimes shared with other fermentors. are high, then special stainless steels should be
This is a poor option as there are times when considered to avoid stress corrosion.
more than one fermentor needs to be cooled. It is important that the slope of the bottom be
The main disadvantage of this system when sufficient for the mash to run out when
shared among fermentors is bacterial cross- discharging. This type of fermentor is much
contamination. With such a system, the better suited for use with clear or semi-clear
microbiological condition of all fermentors will mashes than with whole cereal mashes.
be determined by the fermentor with the most Fermentor cleaning is accomplished with
contamination. Even when these recirculating clean-in-place (CIP) equipment. CIP sprayheads
heat exchangers are not shared, they are difficult are high-pressure devices (100-120 psi), which
to clean. Throughput of at least 7 ft3/sec must usually have automated cleaning cycles. A
be obtained to ensure turbulent flow through the typical cleaning cycle would be (minimum):
tubes.
The agitator is required particularly at the start Pre-rinse with water 10 minutes
and at the end of fermentation. Frequently the Detergent circulation 20 minutes
agitators are badly designed and provide poor Post-rinse with water 10 minutes
mixing. A folding action is required to ensure Sterilization 10 minutes
proper mixing of the mash solids and to ensure
an even temperature throughout the fermentor.
Carbon dioxide (CO2) is removed through the The detergent used would be based on caustic
vent. Fermentors must be fitted with pressure soda, normally with added wetting agent,
relief valves and vacuum breakers to avoid antifoam and de-scaling agent. The caustic
serious accidents. The CO 2 is frequently strength should normally be in the 3-5% range.
collected and sold. In any case, CO2 should be Ideally, the detergent should be hot (80-90°C).
scrubbed to remove alcohol, which is returned The detergent and rinse waters should be
to the beer well. The CO2 header manifold can continually drawn off from the fermentor to
be a source of contamination as infected mash prevent accumulation of liquid during the cycle.
from one fermentor can migrate in the CO2 into The detergent and post-rinse should at least be
another non-contaminated fermentor. The CO2 recirculated and continually made up to strength.
manifold should be designed to be cleanable. Chlorine dioxide and iodophors make ideal
The lag phase is a great opportunity for sterilizing agents, but many distillers still use
lactobacilli to become established. Bacteria steam to sterilize the fermentors. This is time-
under optimum growth conditions can reproduce consuming and is probably not as effective as
every 20-30 minutes; however yeast, which are chemical sterilization.
much larger microorganisms, can only It is important to control the build up of
reproduce every 3 hrs. A single bacterium beerstone (calcium magnesium phosphate and
reproducing every 20 minutes would produce a calcium oxalate). Bacteria can penetrate
population of 256 in 3 hrs. Some yeast strains beerstone, which is insoluble in straight alkaline
have lag phases as long as 12 hrs, during which solutions. Beerstone thus protects the bacteria
time the lactobacilli can become heavily from detergent and sterilant. EDTA-based
established. Therefore, it is very important to chelating agents added to the detergent help
choose a yeast strain with a very short lag phase dissolve beerstone and will control
132 D.R. Kelsall and T.P. Lyons
accumulation. The level of chelating agent fermentors and a pre-fermentor. The majority of
needed can be calculated from the calcium level these plants in the US use whole mash fed into
in the mash. A problem recently encountered is the first two fermentors. Both are aerated
sodium contamination of process condensate continuously with sterile air and are cooled with
water. During CIP the caustic based detergent external coolers. The pre-fermentor feeds yeast
can be neutralized by the CO2 in the fermentor equally into both. The mash has been previously
producing large quantities of soluble sodium saccharified and enters the fermentors at a pH
carbonate that can pollute the process of 4.0 or slightly less. Usually these systems, if
condensate. Very high sodium levels can weaken yeast stress factors are taken into account,
the yeast during fermentation and care must be produce 13-15% beers in 24-30 hrs although
taken to reduce these sodium levels. clear mash systems operate in 10-14 hrs. Each
fermentor is individually cooled and agitated
either with air or CO2 or mechanically. The clear
CONTINUOUS FERMENTATION mash systems are capable of maintaining a fixed
yeast count although they have the potential to
The most successful continuous fermentation recycle.
system used in distilling is the cascade system Typically yeast cell counts in these fermentors
(Figure 9). The system illustrated in Figure 9 has are slightly higher than in batch systems, with
at least two fermentors and a beer well where counts in the range of 180-220 million cells/ml
the yeast are recycled. Yeast can only be recycled for the first two fermentors and slightly less as
when clear mashes are used. The yeast can then the mash proceeds through the system. The
be centrifuged, washed and reused. Typically, viability and percentage of budding cells
yeast can be washed using phosphoric acid at a decrease from one fermentor to the next.
pH of 2.2-2.4 and a holding time of 90 minutes. Typically, the beer entering the beer well could
These conditions are sufficient to kill the bacteria be down to 120 million cells/ml with a viability
without doing permanent damage to the yeast. of around 30%. Alcohol levels could be 10-11%
Chlorine dioxide at 40-50 ppm is an alternative in Fermentors 1 and 2, 12-13% in Fermentor 3
to acid washing, and seems to be gentler on the and up to 15% in Fermentors 4 and 5.
yeast. The main advantages of continuous
The system in Figure 9 has two fermentors, fermentation (and they are very important
whereas most cascade plants have five advantages) are rapid throughput and the fact
Agitator drive
Agitator drive
Foam breaker
Oxygen
column
Mash in CO2 outlet
Cooling
Pump Vessel Coil
No. 1
Vessel
No. 2
Beer
outlet
Sterilizer Yeast
outlet
Oxygen in
Chapter 11
W.M. INGLEDEW
Applied Microbiology and Food Science Department, University of Saskatchewan, Saskatoon,
Saskatchewan, Canada
Introduction
Fuel alcohol manufacture in North America is Table 1. Fuel alcohol plants using continuous or batch
conducted in approximately 70 (soon to be 78) fermentation (2003).
production plants. The annual output from North
American fuel alcohol distilleries will soon Numbers of US plants (70 total)
exceed 3 billion US gallons (11 billion liters). Volume Continuous Batch
Both continuous and batch fermentation
technologies are used, with the breakdown <11 million gal/year 3 15
between plants (Table 1) and on a volume basis 12-39 million gal/year 9 15
>40 million gal/year 14 14
(Figure 1) not exactly illustrating the fact that
older and larger wet mills predominantly use
continuous technology whereas the newer and In this article, continuous fermentation, the
smaller dry mills predominantly use batch prolonging of the logarithmic phase of yeast
fermentation.
Billion gallons
Continuous Batch
Figure 1. Volumes of alcohol made in 2003 in 70 US plants by batch and continuous technology.
136 W.M. Ingledew
growth seen in batch culture, will be the focus rates, and genetic stability of the culture used –
along with a discussion on the technology and all of these negating, in part, the advantages
information relating to problems that yeast have mentioned above. Contamination leads to losses
in such environments. in yields of ethanol and unplanned shutdowns
of the fermentation trains – factors that lead to
lower than expected yields of product and less
Continuous fermentation efficient conversion of sugars (from starch) to
ethanol (Bayrock and Ingledew, 2001b). In the
Continuous fermentation is said to be fuel ethanol industry, most continuous
advantageous because of elimination of the lag fermentation is done by the wet milling segment
phase of yeast growth, the yeast being ‘locked’ of the industry with the advantage that ethanol
into exponential phase growth where conditions production can occur over prolonged periods
exist for maximal ethanol formation, long term of time under optimal environmental conditions.
continuous productivity, higher volumetric
productivity, reduced labor costs while in steady
state, reduced downtime for cleaning, filling and SINGLE STAGE CONTINUOUS FERMENTORS
sanitation, easier process control and savings in
construction of smaller fermentors with higher This most simple form of continuous
output (Cyzewski and Wilkie, 1978; Kirsop, fermentation is diagrammed in Figure 2. In such
1982; Kosaric et al., 1987; Maiorella et al., a system there is a continued flow (F) of nutrients
1981; Sinclair and Cantero, 1990). Higher into the fermentor from the medium reservoir,
technological capability to operate such systems and a continual flow of product (P) out of a
is needed, but the primary disadvantages in fermentor to a harvest reservoir. The fermentor
continuous culture are the problems with is designed such that the internal volume of
contaminating bacteria and wild yeast that ‘medium’ remains constant.
disturb the balanced nature of the fermentors, The parameter describing this is called the
problems of maintenance of high fermentation dilution rate (D), measured in hr-1, calculated by
Pump
Air in Air out
Overflow weir
Nutrient
reservoir
Fermentor
Harvest
reservoir
Figure 2. A schematic diagram of a continuous fermentor. The flow rate of medium into the fermentor is controlled, thereby
determining the flow of product into the harvesting reservoir (Waites et al., 2001).
Continuous fermentation in the fuel alcohol industry: how does technology affect yeast? 137
dividing the medium flow rate (F) by the working MULTISTAGE CONTINUOUS CULTURE
volume of the vessel (V). The vessel is stirred
to ensure continued mixing of the contents. An In multistage continuous culture in the fuel
adequate, constant flow of the same medium alcohol industry, four or five fermentors linked
ensures that D is constant, and therefore that the in series are normally used. In such a
yeast cell growth is constant, the substrate level fermentation ‘train’, each fermentor is usually
(and product level into the harvest reservoir) are of the same working volume, and is ideally fed
constant. At that time, the vessel will be from the one before. The first fermentor is fed
balanced in a so-called steady state. Steady state by a mash stream that is under continuous
is governed by the level of a limiting nutrient in production and cooled aseptically to
the mash medium. It is normally achieved after fermentation temperature. Interestingly, if
three residence (replacement) times, where T = fermentors can be operated in balanced growth,
1/D. In some studies, six residence times are there is no need for added yeast, no need for
used to ensure steady state conditions. nutrient supplements or mash additions to the
When changes to dilution rate take place, the second to fifth fermentors and little need for
parameters above (P and S concentrations) as anything but monitoring. In such cases, each
well as the cell biomass all change, and time is fermentor is in balanced growth where the
needed to bring the vessel back to steady state. substrate level is constant, the product level is
The time required depends upon the size of the constant, the rate of fermentation is constant and
vessel(s) used and will also depend on any other so is the amount of yeast biomass (viable yeast)
environmental changes or chemical or biological that, of course, is the catalyst of the biochemical
contamination which can affect the pH, levels reaction. However, none of the fermentors in the
of inhibitory compounds present, temperature train have the same steady state values as
etc. The rate of cell growth is denoted by µ, the fermentors upstream or downstream. Dilution
specific growth rate of the yeast. Contaminants rates, however, will be identical if the volumes
have their own specific growth rates (µ) that, in contained are the same and if no additional mash
competition with the yeast, may result in faster or other additions are made distal of Fermentor
or slower growth rates than that of the yeast. In 1. Productivity on continuous processes can
such a case, one microbe will wash out of the exceed 6 g ethanol/L fermentation mash/hour
fermentor, normally leaving the other as the (>2.4 to 3.3-fold more than corresponding batch
dominant organism. Specific alcohol fermentation – Kosaric et al., 1987), and this
productivity in a simple continuous fermentor means that for similar plant outputs, vessels can
is ordinarily limited by ethanol inhibition and be 3-fold smaller, and operating and capital costs
by low cell concentration, necessitating substrate are reduced by ~50%.
feed concentrations of about 10% w/v (Kosaric Figure 3 shows a simple engineering diagram
et al., 1983). of a continuous train. This is a train with four
!" !# !$ !% &''(
)'**
Fermentors
!"#$%$&#'%(")'*+,-&.
A and B trains possible
fermentors and a beer well. The inoculum is Another advantage in operation of a multistage
added when needed, and saccharified mash is train is that the first vessel can be used to produce
continuously provided to Fermentor 1. Cooling cells (viable biomass) that will be used to
is provided to each fermentor and an effluent transform remaining substrate in Fermentors 2
takeoff is sent to the second fermentor at the to 4 to end product. For example, if the volume
same rate of addition as that of the mash entering of Fermentor 1 was twice the volume of the other
Fermentor 1. Carbon dioxide leaves each production fermentors, the dilution rate would
fermentor, and it is scrubbed to remove ethanol be lower (and average medium residence time
(in this case about 1%) and any dissolved would be higher) leading to more extensive cell
solubles or entrained particulates. Often in such growth. Conditions would be imposed on
designs, two identical trains are built. This gives Fermentor 1 to ensure maximum growth, while
a remarkable opportunity for any changes made conditions in Fermentors 2 to 4 (or 5) could be
in the A train to be compared to the unchanged more optimal for ethanol production. If a
B train – during and after a period of medium feed occurs in Fermentor 2 as well as
equilibration to steady state. in Fermentor 1 (called distributed feeding), the
Figure 4 is a photograph of a typically installed mathematics at steady state for substrate
multistage continuous culture. This fermentation utilization, product formation, dilution rate, etc.
train has been a model of steady state will be different and more complex. A
fermentation where for long periods of time, mathematical treatment of continuous
balanced growth was maintained, and the fermentation can be found in the PhD thesis of
fermentor during this time did not require D. Bayrock (2002).
additional yeast cells to be added.
There are many possible variations in a
designed multistage continuous fermentation. CONTINUOUS CULTURE WITH YEAST
Some involve a pre-fermentor, a yeast RECYCLE
propagator, or additions of nutrient mash streams
into more than one fermentor. In all such Figure 5 portrays a continuous culture system
schemes, balanced S, P, or biomass is more with provision for yeast recycle. This is the so-
difficult to maintain in any vessels distal to the called Biostil fermentor system where effluent
last one where nutrients are added or where from the continuous fermentor is centrifuged,
changes occur. and the collected yeasts are re-added to the
Figure 4. The continuous fermentation train previously operated by Williams Energy in Peoria, IL.
Continuous fermentation in the fuel alcohol industry: how does technology affect yeast? 139
fermentor to provide additional catalytic activity yeast concentration of up to 109 (1000 million)
and therefore faster production of ethanol cells/mL, the 6-9% v/v ethanol present, and the
(Danielsson, 1992). The advantages of this fact that internal pasteurization using a heat
design include the continuous removal of yeast exchanger can be accomplished. A one vessel
by centrifuge, an increase in yeast cell density system is also easier to shut down, sanitize and
in the fermentor of 6 to 10-fold, and a lesser restart.
synthesis of new yeast with the concomitant Continuous cultures, regardless of design, must
small increase in sugar available for ethanol operate simply and at maximum capacity and
production. Unfortunately, the viability of maximum efficiency. They must produce
recycled Biostil yeast is not good. Solids ethanol as well as by-products of consistent
accumulate with the yeast and build up due to quality. As said previously, the multistage systems
recycle (the reason why the design is more of in use in fuel alcohol production are often
importance in sugarcane juice fermentation than somewhat complex in that more than one
in corn mash fermentations), bacteria and wild fermentor is fed, a pre-fermentor or yeast
yeast increase in number and lead to propagator is often engineered into the system,
accumulation of inhibitory metabolites, and and the presence of continued infection leads to
centrifuge/separation is required. Operating abnormal stresses that take the fermentors out
costs and capital expenditures increase due to of balanced growth. Reestablishment of balanced
the need for cell separators and reinoculation. growth can take well over 48 hrs, and this means
The literature reports ethanol productivities as that the plant is continually destabilized and that
high as 50 g ethanol/L/hour in these fermentation the lab and plant staff are continually challenged
systems – directly due to increased yeast to find the key to stabilize conditions.
biomass. Yeast is the catalyst in the reaction It is especially a challenge to reduce or
converting sugar to ethanol, carbon dioxide and eliminate bacteria without losing productivity
glycerol. It is said that control of infection in (ethanol made per working volume per hr). It is
such a system takes place due to the low pH, normally not possible to clean and sanitize one
the low content of fermentable sugar (usually fermentor without later cross-contaminating it by
less than 0.1%), the high osmotic pressure, the the others in the train. This is because all are
Ethanol
CO2 Primary
distillation
Substrate
Concentrated stillage
Figure 5. The Biostil fermentor system (Chematur Engineering, Concept of Continuous Alcohol Production, 1993).
140 W.M. Ingledew
contaminated in the flow of the liquid through (insufficient time to multiply), which eventually
the train, and when Fermentor 1 is taken down leads to lowered productivity.
for cleaning and then restarted, it quickly It is also the belief of this author that
becomes contaminated from the CO 2 headers continuous systems may never be able to
that often connect the train in a countercurrent produce ethanol at levels over 15-16% v/v, due
direction. Headers are also very difficult to to washout of cells occurring because the yeasts
clean. In addition, the mathematics and are continually bathed in ethanol and because
engineering theory are so complex in the levels of other medium ingredients (some
comparison to batch fermentation that it is not recycled in backset) also increase providing
easy to obtain adequate advice on eradication continual, synergistic stresses that lower
of problems. Moreover, plants often make metabolic rates and decrease ethanol levels
changes to the parameters under which the train during the residence time of the fermentors. As
operates, and do not allow enough time to see if substrate levels are increased (concomitantly
the train re-establishes productivity (steady state) with alcohol concentrations), the dilution rates
before even more changes are made. become smaller such that for all practical
Another problem inherent in the operation of purposes, the fermentors approach batch mode
continuous multistage systems is that rather than remaining in a continuous mode.
saccharification of starch must be carried out at It should be mentioned from an academic
the optimal temperature for the enzyme prior to standpoint that other continuous systems have
addition of the mash medium to Fermentor 1. been developed, including: internal recycle
The saccharification stage is ideally suited for (flocculation of cells), tower fermentors,
growth of very heat tolerant Lactobacillus strains membrane reactors, vacuum fermentation
known to create the lactic acid in fuel alcohol (removing ethanol and carbon dioxide), solvent
plants that interferes with yeast metabolism. extraction (removing ethanol), and immobilized
Simultaneous saccharification and fermentation cell reactors. None of these have had much
(SSF) technologies were developed for batch importance in manufacturing ethanol except in
fermentation to eliminate this very important the brewing industry.
source of bacteria and the resultant loss of
ethanol yields. SSF is not possible in continuous
systems at this time. Moreover, in a multistage Contamination and stressful conditions in
train, saccharification, yeast propagation, pre- a fermentation plant
fermentation, and fermentation are the normal
sequence of events. This complicates the total The source of microbes in distilleries is not mash
understanding of changes that occur in the train prior to its cooling, due to the high temperatures
with time. As the section below and the chapter used for starch gelatinization and liquefaction.
in this volume on water usage show, balanced All points in the alcohol process after cooking,
growth is very susceptible to wastewater streams however, can be foci for infection including
and surges of such wastes (such as sodium coolers or heat exchangers, tankage, transfer
hydroxide (NaOH) in wash water, evaporator lines, unprocessed waters, yeast, enzymes and
condensate, thin stillage, still bottoms, or other additives, air, fermentor tankage or the beer
biomethanator discharge) and this, compounded well. Viable contaminants and their end products
by changes in mash composition, temperature, in recycled waters can all affect subsequent
osmotic pressure, and bioproducts of fermentations. Contaminants of concern to fuel
contaminants including acetic acid and lactic alcohol plants mainly include the Lactobacillus
acid all can lead to perturbations of the train that bacteria and wild yeasts. It is important that
are difficult to repair. In addition, our lab has lactobacilli do not grow well in corn mash and
shown that deficiencies in yeast-usable nitrogen are not able to grow at low pH. However, some
occur in mashes made from every conceivable are well adapted and grow exceedingly quickly
grain substrate. Such constituents vary from load under mash fermentation conditions. For the
to load of received grain, making continuous purpose of this chapter, it is important to realize
operation difficult due to the fact that yeast levels that bacteria and wild yeast compete with
in mash vary with nutrition. Even small changes Saccharomyces yeast for nutrients. Moreover,
in dilution rate can lead to washout of yeast and more importantly, the end products of the
Continuous fermentation in the fuel alcohol industry: how does technology affect yeast? 141
metabolism of the bacteria are normally lactic mash. That of the wild yeast and molds would
acid and smaller amounts of acetic acid, whereas approximate 2 cells per ml. At such levels, it
the wild yeasts (like Brettanomyces) produce appears obvious that the bacteria would need to
acetic acid (in addition to ethanol and carbon have a very definite competitive advantage in
dioxide). the fermentor (or a prolonged time) before
It is these contaminant end products that cycle significant interference in the process would
throughout the plant due to the fact that they are occur. This is why a batch fermentation process
not as volatile as ethanol and mostly remain in has little worry about contaminants in ADY. A
the thin stillage. For this reason, waste waters continuous fermentation train could be in
such as backset and evaporator condensate are difficulty using direct inoculation of ADY if the
critical components in mash manufacture bacteria, wild yeast or mold present in the ADY
because they contribute increasing amounts of grew faster in the mash provided under the
chemicals that inhibit yeast growth, and the environmental and chemical conditions (and
levels of such chemicals increase with every time of operation) of the fermentation train. In
subsequent cycle. Many, if not most, problems all likelihood the contaminants from equipment
in fermentation can be alleviated if a plant can would exceed these numbers by a considerable
eliminate backset and/or evaporator condensate, amount. The lowering of pH practiced in plants
replacing them in the mash bill with water for suffering contamination often lowers the growth
one or more cycles. This emphasizes the need rate of the bacteria present such that they wash
for excess evaporator capacity in newly designed out of the fermentation, leaving the more pH
facilities. tolerant yeast to multiply and persist. However,
the cost of this procedure is lowered alcohol
yield.
CONTAMINATION OF ACTIVE DRY YEAST Yeast repropagation in a fuel alcohol plant is
a different story especially when continuous
This subject of entrance of bacteria into mash propagation is practiced. Under such conditions,
via pipes, unclean tanks, or air, etc. is covered the very small number of bacteria mentioned
in the chapter by Larson and Power in this above that are present with the yeast in
volume and found in books discussing cleaning commercial ADY preparations are provided an
and sanitation. A word is needed, however, on opportunity to grow and compete. Because
levels of contamination in both active dry yeast bacteria normally have significantly shorter
and via plant-propagated yeast in order to generation times (times required to double),
understand how fermentations can become those few bacteria that can endure the
contaminated in-house. conditions of the fermentor (like Lactobacillus)
Yeast can be a source of contamination. Active will compete with Saccharomyces yeasts for
dry yeasts (ADY) cannot be grown, harvested nutrients and eventually increase in numbers to
and dried without a level of bacterial, mold and the extent that they begin to produce large
wild yeast contamination. Usual levels are given amounts of lactic acid and some acetic acid
in Table 2. which inhibit yeast and help them to take over
the population in the propagator, and in similar
Table 2. Normal contamination in active dry yeast fashion, the fermentors. These end products are
(organisms/g).
known to affect the growth of culture yeast and
Bacteria Wild yeasts and molds Culture yeasts reduce alcohol output in fermentors when these
vessels become contaminated at levels over 105
103-105 <102 - 103 2.2 to 5 × 1010 bacteria (100,000 ) per mL (Narendranath et al.,
1997). As acids build up to over 0.1% w/v (lactic
acid) and 0.05% w/v (acetic acid), we know that
Considering the levels of contaminants, and the length of the yeasts’ lag time after inoculation
usual levels of inoculation of active dry yeasts will increase significantly (Figure 6) as does
into fermentors (2.2 lb/1000 gal US), it can be yeast generation time (Narendranath and
estimated that the contribution of the bacteria in Ingledew, 2001a). For these reasons, continuous
the mash after a one-time inoculation of ADY is propagation is not recommended by this author.
less than 26 viable bacteria per ml of inoculated
142 W.M. Ingledew
%" Stress occurs >0.05% "#' &! Stress occurs >0.1% "#'
$! "#& %! "#&
9/8501:'5;<(7=
µ >'(5<(
µ >'(5<(
%"
$"
Figure 6. Specific growth rates decrease and lag times increase exponentially as acids increase in concentration. MIC (acetic) =
0.6%, MIC (lactic) = 2.5%. MIC = no growth at 72 hrs. = specific growth rate, = lag times; generation time = ln2/µ.
CONTAMINATION CONTROL STRATEGIES train at Williams Energy in Pekin, IL. The work
was conducted with penicillin G, an antibiotic
The effects of bacterial end products when more long used in the industry. Luckily, penicillin is
than one stress is present are additive, or more not degraded enzymatically by Saccharomyces
likely synergistic, and affect the yeast in yeasts. The penicillin was therefore added in
association with other stressful situations like pulses considering scientifically only its
high ethanol, high temperature and low pH. The degradation rate at the pH of the mash, its
yeasts are then in serious trouble (Narendranath degradation rate at the temperatures used over
et al., 2001a). The mechanism of action of the the time course of the continuous system, and
stressful chemicals is now better known the dilution rate of the fermentor (and each
(Narendranath et al., 2001b). For this reason, it component in it). It was known that in the system
is necessary to reduce the numbers of the used only 49.8% of the antibiotic would remain
microbes present in mash such that the levels of after 3.2 hrs of incubation in mash at pH 6 at
bacterial end products are too small to affect rates 28°C, with only ~1% remaining at 21 hrs. The
of ethanol production or ethanol yields. This is regime for addition therefore included antibiotic
normally done with low pH, antimicrobials or pulses every 6 hrs with an average penicillin
antibiotics. concentration of 2475 U/liter but with spikes of
The use of antibiotics in continuous culture larger values than the average such that bacteria
has been more of an art than a science. would be alternatively inhibited from growing -
Unfortunately, in the past most applications of and perhaps also encouraged to initiate growth
antibiotics have been done with the same thought again when concentrations were lowest. This was
process that had been in use for batch found to be more effective than the continuous
fermentations. Antibiotics were, therefore, added addition of the same antibiotic over time to the
all at once into continuous systems with the result same average value (Bayrock and Ingledew,
that a spike of antibiotic concentration took 2003). This technology is being translated to
place, which decreased over time at a rate industrial use. Other single antibiotics are
prescribed by the set dilution rate (D). In 2001, available to the industry as are commercial
the published work by Bayrock and Ingledew formulations with combinations of antibiotics
(2001a; 2001b) provided alternative strategies providing a broader target range against
for the provision of antibiotic to continuous contaminating bacteria. The latter reduce a
systems. This work was done in a bench scale number of persistent infections that detract from
multistage continuous culture system, which was ethanol yield.
designed to model the commercial fermentation
Continuous fermentation in the fuel alcohol industry: how does technology affect yeast? 143
Chapter 12
Introduction
Near-infrared (NIR) spectroscopy has been a fermentors, distillate analysis, dry house
reliable technique in the Hiram Walker & Sons operations, and finished product analysis. The
distillery quality assurance laboratory since exact locations where NIR spectroscopy is used
1997. It has proven to be powerful tool that can in the Hiram Walker & Sons distillery, and the
benefit distilleries, breweries, or other producers advantages and disadvantages of the technology,
of other types of ethanol in quality assurance will be discussed in more detail. Initially
programs and research facilities. As a research background material for NIR spectroscopy will
tool NIR spectroscopy can be used to evaluate be discussed, to help the reader understand what
raw materials, yeasts, enzymes, nutritional it takes to implement NIR spectroscopy in a
supplements, and production parameters to quality assurance or research program.
optimize conditions for a production plant. As a
quality assurance tool it can assist production
by monitoring and maintaining control of History of NIR spectroscopy
processes. If failures are not quickly identified
and a process becomes out of control, it can cost Infrared spectroscopy has been a well-
valuable time, product, and resources. NIR established research tool in academia and in
spectroscopy can help eliminate these problems. industry for several decades. The first
What makes NIR technology so attractive to experiments conducted with NIR light were by
the ethanol industry as a quality assurance and Herschel in 1800, in which he used a glass prism
research tool is that it can measure organic to demonstrate that there was radiation beyond
substances very quickly (5-10 seconds) and visible light (Herschel, 1800). NIR spectra are
consistently without the destruction of a sample. complex however, and only recently have
Traditional analysis can take a lab technician computers become powerful enough to analyze
several hours or even days to get reliable results, the spectra of complex samples. NIR technology
but with NIR spectroscopy the result is virtually is now also a choice for process monitoring
instantaneous and does not use hazardous equipment, because there is no sample
chemicals. The areas of production where rapid preparation required and fiber optics can be used
and reliable analysis are most beneficial include to transmit the light over significant distances.
monitoring of incoming grains, profiling This has allowed companies in the food and
146 D. Livermore, Q. Wang and R.S. Jackson
beverage industry such as ethanol producers, to not be the same wavelength as that of the
move from low technology to high technology incident light. Reflected light is similar to
production methods. scattered light, but the wavelength of reflected
light is always the same as that of the incident
light. Scattering and reflection of a particle are
NIR background dependent on the relation between the
wavelength of the light and the dimensions and
A detailed discussion of the theory of NIR orientation of the particle. Transmitted light is
spectroscopy is beyond the scope of this chapter, non-redirected light, which if it interacts with
but a review of the basics may help the reader matter maintains the same wavelength after
understand why NIR is a powerful tool in interaction.
industry today. Absorption of infrared light by a molecule is
Light consists of electromagnetic waves of due to interaction of the electromagnetic
varying wavelengths (Figure 1). Our eyes can radiation with the vibration of bonds between
only sense light at particular wavelengths, called atoms. To absorb infrared light there must be a
the visible spectrum, but the electromagnetic change in dipole moment during the vibration,
spectrum covers a much wider range. The near- so homonuclear diatomic molecules such as H2,
infrared region of the spectrum extends from 800 O2, and N2 cannot be monitored using infrared
to 2500 nm (12500 to 4000 cm-1), lying between spectroscopy. In the mid-infrared light is
the mid-infrared (MIR) region at longer absorbed if the frequency of the light is the same
wavelengths and the visible region (VIS) at as the fundamental frequency of vibration of the
shorter wavelengths. molecular bond. Molecular vibrations are
When light impinges on matter, each slightly anharmonic, however, and consequently
wavelength present will be selectively absorbed, higher frequency light in the NIR can also be
scattered, reflected, or transmitted. Absorption absorbed if its frequency is the same as that of
is the transition of a molecule from a lower one of the harmonics of the fundamental. It is
energy level to a higher energy level with the also possible that vibrations can interact and
transfer of energy from the light to the molecule. couple, if the vibrating bonds are joined to a
Scattering is the redirection of light due to the single, central atom. The absorption bands in
interaction with matter. Scattering may or may the NIR are thus referred to as either overtone
not occur with a transfer of energy and the bands, with frequencies of 2, 3 or more times
wavelength of the scattered light may or may the fundamental, or combination bands with
108 107 106 105 104 103 102 101 1 10-1 10-2 10-3
Wavelength (cm-4)
Inner Spin
Valance
Nudea shell Molecular Molecular orientation in
electron
Transitions electronic vibrations rotations magnetic
transitions
transitions field
Wavelength (m)
10-10 10-9 10-8 10-7 10-6 10-5 10-4 10-3 10-2 10-1 1
101
Figure 1. Location of the NIR region of the electromagnetic spectrum in relation to other wavelengths.
Understanding near infrared spectroscopy and its applications in the distillery 147
2.5
O-H C-H
2.0 Water
Absorbance units
Ethanol
1.5
1.0
O-H
0.5 C-H
0.0
10000 9000 8000 7000 6000 5000 4000
Wavenumber (cm-1)
Filter photometers lack the flexibility of other mechanical precision. This can affect instrument
instrument types, and can be prone to errors if stability, existing calibration models, and make
the temperature changes. They have the transfer of a calibration model to a new
advantage of low cost, however, and can be instrument difficult.
useful for dedicated analysis either in the
laboratory or on-line in a production facility. Detector array dispersive spectrophotometers
Filter Wheel
excited by sound waves using a piezo-electric spectrometers can be found in Griffiths et al.
transducer. The grating can be tuned to different (1986).
wavelengths by changing the frequency of the Fourier transform spectrometers offer high
sound. These instruments are fast and rugged, resolution, good speed, and high signal-to-noise
but have limited resolution and signal-to-noise ratios. Their biggest advantage, however, stems
ratio. They are also very temperature sensitive, from the fact that the position of the moving
and therefore typically have built-in mirror is controlled using a HeNe laser. The
thermostatting. inherent wavelength stability of the laser results
in very high wavelength accuracy and precision.
Fourier transform (FT) NIR spectrophotometers This in turn means a calibration model is very
stable over time, and permits easy transfer of
A schematic representation of a Fourier calibration models between instruments.
transform instrument is shown in Figure 5. The FT-NIR has been the instrument of choice in
broadband light source is directed to a the Hiram Walker & Sons laboratory because of
Michelson interferometer. The interferometer these advantages, especially the wavelength
consists of a beamsplitter, a fixed mirror, and a accuracy and precision. In the alcohol industry
mirror that moves back and forth very precisely. measuring spirit strength is very important to
The beamsplitter reflects half of the light to the regulatory agencies. If NIR were to be approved
fixed mirror, and transmits the other half to the by governments as an official method of alcohol
moving mirror. Both the fixed and moving determination, then the wavelength stability of
mirror will direct the light back to the FT-NIR systems might be advantageous because
beamsplitter, where they interfere. This of the resulting model stability. Currently
interference changes as the moving mirror is available systems also offer a full range of
displaced, and therefore the intensity of the light analytical techniques (i.e. solids, pastes,
at the detector changes. The intensity as a powders, liquids, probes, vial holders,
function of mirror displacement is called an integrating spheres) on a single instrument, with
interferogram, which must then be Fourier easy computer controlled switching between
transformed to obtain the spectrum. A detailed different modes.
description of the operation of Fourier transform
Fixed
mirror
Moving
Source Beam
mirror
splitter
Laser
Detector
Sample presentation liquid probe can be used for either off-line, at-
line, or in-line measurements. Light from the
There is a wide range of possibilities in the spectrophotometer is directed through fiber optic
alcohol industry for NIR analysis. There are cable to a pair of mirrors that direct the light
three main categories of analysis. The first of through a cavity in the probe head. The cavity
these is off-line measurement, where sample has a fixed pathlength that enables a liquid
materials are selected from production to be sample to enter (Figure 7). Figure 8 shows a
analyzed by an instrument in a laboratory setting. commercial probe.
This type of sampling places the least demand
on the operator developing the calibration
models. The second category is at-line sampling,
where the instrument is placed close to the
operation in order to decrease analysis time.
Depending on the environment, the instrument
may need to be in an enclosure. Finally, there
are sampling accessories that can be integrated
into the production line without having to collect
samples by hand. This is called in-line analysis.
This type of analysis allows the user to integrate
the spectrophotometer with existing control
systems. In each of these categories there are a
number of sampling accessories available, which
must be carefully chosen. There are three types
of measuring techniques that the accessories can
use: transmission, diffuse reflection, and
transflection.
Figure 7. Schematic diagram of a liquid probe, showing the
path of the light.
TRANSMISSION
lo l trans
IR Source
IR Energy Sample
Reflected IR Energy
Detector
S
Sample
Gold sphere
Detector
NIR beam
D
Figure 11. A rotating cup full of whole kernel corn on an integrating sphere.
lo
Source
l trans
Analyte
Detector Mirror
It is very important to acquire the proper Since the pathlength D is often fixed, in practice
instrument and sampling accessories as it directly the absorption coefficient and the pathlength are
affects the next stage involving NIR technology, combined into one coefficient k. Equation (1)
which is calibration development. When using can be then simplified to:
a NIR spectrometer, a large amount of time is
initially spent turning the raw spectra into useful A= kC (Equation 1)
information. The ultimate goal is to be able to
predict component concentrations in a matter of Solid or slurry samples are often measured in
seconds or minutes, so the time spent diffuse reflectance mode, using the accessories
performing wet chemical analysis of the shown in Figures 9 and 10. For reflectance
calibration samples is well worth the effort. measurements log (1/R) is used instead of
absorbance, where R denotes reflectance. This
is defined as:
THEORETICAL BASICS
log (1/R) = -log (Is/I0)
When infrared radiation passes through a sample,
some of the light is transmitted or reflected, Where I0 is the light intensity measured from
without interacting with the molecules, and the a reference material
remainder interacts with the molecules and is
absorbed. The wavelengths at which the Is is the light intensity measured from
absorption will occur and how much light will the sample
be absorbed depend on a given substance. This
is the basis of qualitative and quantitative The reference material used in the NIR region is
analysis by spectroscopy, including NIR usually a material with uniform scattering
analysis. The amount of absorbed light is properties and very little absorption over the
measured in absorbance. The absorbance, entire NIR region. Two kinds of material are used
generally denoted as A, is defined as: commercially. One is called Spectralon ® , a
material made of a special type of Teflon®; the
A = -log (Is/I0) other one is diffuse gold, a piece of metal with
finely roughened surface coated with gold.
Where I0 is the light intensity before passing There are many theories regarding quantitative
through a sample analysis in diffuse reflectance measurements.
Empirically, log (1/R) is found to be fairly linear
Is is the light intensity after passing with the concentrations of chemical components
though the sample in the NIR, hence,
Spectroscopic quantitative analysis is based on log (1/R)= kC (Equation 2)
the Beer-Lambert law (also know as Beer’s law),
which states that there is a linear relationship The similarities between equations 1 and 2 are
between the absorbance and concentration of apparent.
absorbing molecules. Mathematically, Beer’s law Quantitative analysis generally has two steps:
can be written as: calibration and prediction. During the calibration
stage, NIR spectra of a set of samples will first
A=εCD be measured so that the absorbance values in
154 D. Livermore, Q. Wang and R.S. Jackson
Equation (1) can be determined. The concen- 1989). PLS is the most commonly used method
trations of the chemical components can be for building multivariate quantitative models and
analyzed with traditional lab methods such as is the method used in all data analysis presented
wet chemical or HPLC methods. Once the in this chapter. This chapter will only deal with
absorbance values A and concentrations C are practical aspects of development of calibration
known, the values of the coefficients k can be models and how to interpret common
determined by the least squares method. Once terminology used in the PLS method.
the coefficients are determined, the spectrometer
is calibrated and ready to predict the
concentrations of unknown samples. Equation CALIBRATION SAMPLE SET, NIR
1 only works in a simple system where the MEASUREMENTS AND REFERENCE ANALYSIS
absorption at a particular wavelength comes from
only one component. Such a method is often To build a quantitative calibration model using
called univariate calibration. NIR data, the user must have a reference analysis
As described earlier, NIR spectra are highly method to determine the concentrations of the
overlapped, meaning that absorbance at a components of interest in a sample. The
particular wavelength contains contributions reference methods commonly used in QC
from multiple chemical components in the laboratories in the distillery industry today are
sample. The univariate calibration can therefore HPLC, GC, DMA and various wet chemical
rarely be applied to NIR applications. A more analyses. The accuracy and precision of a NIR
sophisticated multivariate analysis is usually calibration model is strongly affected by the
required to build the calibration model. When quality of the reference analysis method. A few
multiple components are present, the total examples are given in the section of this chapter
absorbance at a given wavelength is the dealing with NIR applications in the distillery.
summation of absorbances of each individual The reference methods should be closely
component. The absorbance at a given examined and understood so that the user does
wavelength can then be expressed as follows: not have unrealistic expectations of NIR
calibration models.
A = ε1 C1 D + ε2 C2 D + …+ εn Cn D (2) When building the calibration model, the
reference analysis and NIR measurements
where should be performed as close to the same time
as possible if samples are not stable. For
A =Total absorbance at a specific wavelength example, live fermented mash samples taken
ε1 =Absorption coefficient of component 1 from a fermentor are unstable. The longer a sample
C1 =Concentration of absorbing component 1 sits between the time of NIR measurement and the
D =path length time of the reference measurement, the bigger the
ε2 =Absorption coefficient of component 2 error to be expected.
C2 =Concentration of absorbing component 2 Preferably, samples from production that span
εn =Absorption coefficient of the nth a period of time should be incorporated in the
component calibration model, with the goal of incorporating
C n = Concentration of absorbing component n all possible variations in the production process.
One must be cognizant of production parameters,
and collect unique samples when they occur. For
There are four multivariate methods used to build example, DDGS can vary in a whisky distillery
the calibration model, Classical Least Squares with the many different mash bills, which can
(CLS), Multiple Linear Regression (MLR), include grains such as rye, barley, wheat, and
Principle Component Regression (PCR) and corn. The NIR user must recognize when different
Partial Least Squares (PLS). How these methods mash bills are being fermented and collect
work is beyond the scope of this chapter, but samples that do not occur very often and
there are excellent review articles describing how incorporate them into a calibration model. This
all these methods work as well as the pros and means it may take an extended period of time to
cons of each (Haaland, 1988; Martens et al., obtain a representative set of samples, but it is
Understanding near infrared spectroscopy and its applications in the distillery 155
important to do this because it makes a by the reference method, the user can then
calibration model more robust, accurate, and develop and validate the quantitative calibration
precise. Another occasion when a user should models. Once the calibration model is validated,
incorporate unique samples is when processing the user can finally enjoy the fruits of all this
equipment breaks down. Usually these are times labor: the prediction of multiple parameters from
when unique or extreme samples occur. For a single NIR measurement in a matter of seconds
example, when a belt conveyer breaks while to minutes. Validation is the key to the
delivering corn to a pre-mix tank it can cause development of PLS models. Validation is used
mash to be more dilute than usual. The user can both to optimize many parameters during the
collect the mash in these fermentors for analysis, development of the model, and to identify
collect the syrup in the evaporators, and collect abnormal samples (outliers). There are two types
the DDGS at the end, because the broken of validation: cross validation and external
conveyer affects all of these processes. In that validation or test-set validation.
way, the next time the conveyer malfunctions
the NIR user can quickly identify the source of External validation
the problem. To make the calibration model
more robust, it is recommended that any odd External validation is sometimes referred to as
samples be included at least three times in the test-set validation. The sample set used in
calibration model (Williams and Norris, 2001). external validation consists of samples with
The calibration samples should also span the known reference data that were not used in the
concentration range as smoothly as possible, calibration set for the model development. The
avoiding large numbers of calibration samples goal for the external validation is to test the
at just a few concentrations. For example, refer predictive ability of the model. At the end of
to the calibration model for crude protein in external validation, the NIR predicted values are
DDGS in Figures 13a and Figure 13b. In Figure compared with those from the reference method,
13a there are not enough samples of DDGS with often expressed in a plot of NIR predicted vs.
low (21 to 22%) and high crude protein (27 to reference values. The value of R2 in this plot is
31%). To make a calibration model more robust one of two important indicators for how good
and precise, the sample distribution should be the model is. The other one is the Root Mean
similar to that of Figure 13b. This type of sample Square Error of Prediction (RMSEP), which is
distribution, where samples are evenly used as an indicator of the average error for NIR
distributed over the calibration set, is often prediction on future unknown samples.
described as a ‘boxcar’ effect (Williams, 2001).
The calibration range should be larger than the Cross validation
specified range and not too small in view of the The advantage of external validation is that it is
error of the reference method (EMEA 2003). If a true independent validation. The main
the required accuracy is greater than 0.1% by disadvantage of external validation is that it
weight, generally it can be achieved with NIR. wastes the valuable calibration data in order to
If the required accuracy is between 0.1 to 0.01% check the predictive ability of the calibration
by weight, then the NIR application is likely to models. When the number of available
be feasible for a simple system with very few calibration samples is small, a procedure called
components, an accurate reference method, a cross validation is often used. In a cross
wide enough concentration range, and a lot of validation, the first calibration sample is removed
samples. If the required accuracy is 0.01% wt to from the calibration set. PLS models are
a few ppm, the application is unlikely to be generated using the remaining calibration
feasible. samples. The PLS models are then used to
predict the first sample that is left out and the
error of prediction for this sample is calculated.
BUILDING CALIBRATION MODELS
The first sample is then re-included into the
After the calibration samples are collected, calibration set and the next sample is removed
measured by a NIR spectrometer and analyzed from calibration. The whole process is repeated
156 D. Livermore, Q. Wang and R.S. Jackson
Number of samples
8
7
6
5
4
3
2
1
0
20 21 22 23 24 25 26 27 28 29 30 31
8
7
6
5
4
3
2
1
0
20 21 22 23 24 25 26 27 28 29 30 31
until the last sample is removed and analyzed. external validation result is often shown in a plot
The number of samples left out can be just one of NIR predicted values vs. reference method
sample or a small set of samples. At the end of values. In an ideal situation, this plot should be
cross validation, the Root Mean Squared Error a 45 o line, where the prediction should equal
of Cross Validation (RMSECV) is calculated and the reference value. The closer the R2 value is to
used as an indicator of the future predictive 1 or 100%, the more accurate the model. Figures
ability of the models. 14a and 14b show a PLS cross validation result
and an external validation for prediction of
Results of validation ethanol concentrations in corn mash samples.
Errors exist in all measurements and cannot
As discussed earlier, the cross validation or be avoided. It is important to identify the
Understanding near infrared spectroscopy and its applications in the distillery 157
12
11 Rank: 5
10 R2 = 99.86
Figure 14a. Cross validation result for ethanol prediction in fermented corn mash samples. The values of R2 and RMSECV were
0.9986 and 0.139 % v/v, respectively. The number of samples in the calibration set is 111.
13
12
11 Rank: 5
R2 = 99.86
NIR predicted value
10
9
RMSEP = 0.144
8
7
6
5
4
3
2
1
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13
Reference method value
Figure 14b. External validation result for ethanol prediction in fermented corn mash samples. The values of R2 and RMSECV
were 0.9986 and 0.144 % v/v, respectively. The number of samples in the validation set is 40.
accuracy that is needed for a plant to keep a number of samples in the calibration set. In
process under control. For example in protein general, one should incorporate as many as are
in DDGS, there is no need to strive for an error available. Whether or not there are enough
of 0.05% when an error of 0.5% is satisfactory samples to build a stable calibration model can
for plant operations; however, a 0.05% error for only be determined by performing validation
determining alcohol strength in a finished blend tests with an independent validation set. As a
can cost a company money in terms of ethanol rule of thumb, one should start with a minimum
losses and taxes. of 10 samples for each independent chemical
component or other source of variation. ASTM
How many samples in a calibration data set? provides guidelines that can be used to judge
whether or not a calibration set contains
It is very important to include an adequate sufficient samples (ASTM, 1997).
158 D. Livermore, Q. Wang and R.S. Jackson
A sample should always be collected by this measuring incoming corn directly without any
consistent and statistically reliable method. sample preparation such as grinding. Since the
Sampling with this method can also detect absorption coefficient in the NIR region is low,
loads containing grain that is out of NIR light can penetrate corn kernels and whole
specification on the bottom that have been kernels can therefore be used for model
‘topped up’ with good quality grain. development. The NIR spectrum can be
collected with the aid of an appropriate sampling
3. Production efficiencies can be tracked more device such as a rotating cup or a flow-through
closely. Accurate determination of funnel. Depending on the available sampling
production efficiency is important, because devices and how a particular calibration model
it helps the distiller make improvements and was developed, some instruments today still
reduce costs. require users to grind incoming grain in order
4. More crop data accumulated each year. to achieve a reliable measurement.
Calibration models for other grains including
5. Variety selection for desired characteristics
rye, rye malt, barley, and barley malt are
is possible.
currently being developed. The components of
interest include starch, oil, protein, moisture, and
The calibration model Hiram Walker & Sons enzymatic activity.
currently uses to monitor incoming corn was
provided by Bruker Optics with the purchase of
the NIR spectrometer. The expected value of FERMENTATION MONITORING
each of the four major components and the
predictive error of each component expressed Fermentation optimization can be complex,
in RMSEP are listed in Table 1. because many parameters can affect the final
alcohol content and overall yield. With the
Table 1. Expected values and predicted errors for starch, traditional analyses available today it can be time
oil, protein and oil content of incoming corn expressed in consuming for laboratory staff to keep up with
RMSEP. the demand for fermentation analysis. Optimum
efficiency can be achieved if all parameters are
Component Expected result Approximate RMSEP kept in control prior to and at fermentation, but
(% as is basis) (+/- %)
this does not always happen as equipment
Starch 60–62 1.2
failures can occur hours before being identified
Oil 2–3 0.5 and corrected. NIR technology can identify non-
Protein 7-8 0.5 optimum conditions in the fermentation almost
Moisture 12-15 0.3 immediately, allowing correction of problems.
In addition, NIR analysis can aid distillers in
evaluating the new enzyme and yeast nutritional
A major advantage of using an FT-NIR system supplements.
is that it has high wavelength accuracy and In the industry today, most distilleries use an
precision, which allows calibration models to be HPLC for fermentation monitoring, including
easily transferred from instrument to instrument measurements of sugars, acids, glycerol, and
via the internet, without having to adjust for alcohol. All these components are interrelated
slope or bias on each instrument. As a precaution, and it is very important to keep track of them at
the downloaded models are validated with a set each stage of fermentation, but there are
of 40 corn samples that are not from the original disadvantages to using HPLC as the primary
calibration set. The predicted error of the 40 monitoring tool:
validation samples measured by the instrument
in the Hiram Walker & Sons plant was roughly 1. The technician must be highly skilled.
the same as the error in the original model, which
met requirements. 2. Accessories can be quite expensive (i.e.
Another advantage of this calibration model columns, chemicals, filters).
is that reliable results can be obtained by 3. The amount of time needed to determine the
160 D. Livermore, Q. Wang and R.S. Jackson
12
11
10
9
Ethanol (% abv)
8
7
6
1999 2003
5
4
3
2
1
0
0 15 30 45 60 75 90 105 120
Hours of fermentation
Figure 15. Fermentation profiles from 1999 to 2003 during which period NIR has assisted in optimizing fermentors and thus
improved the distillery bottom line.
Understanding near infrared spectroscopy and its applications in the distillery 161
reference data available in order to determine value of RMSECV for the NIR models using the
when an enzyme, nutritional supplement or HPLC and distillation-DMA method is +/- 0.67%
other change in fermentation conditions is and 0.18%, respectively. This demonstrates the
effective. greater accuracy of the distillation-DMA method
Figures 16a and b show the two calibration for the calibration of NIR models for corn mash.
models for determining ethanol in corn mash. It is recommended that the distillation-DMA
The first NIR model was built using HPLC as procedure be used for ethanol calibrations in
the reference method (Figure 16a) and the second corn mash instead of the HPLC method.
was built using the distillation – DMA reference In contrast to alcohol, for sugar calibration
method (Figure 16b). The distillation – DMA models HPLC is recommended as the reference
method is to distill 100 mL of the corn mash, method. Even though the prediction error for
collect 100 mL of the distillate and then sugar is roughly +/- 0.5%, it is still acceptable.
determine the percent alcohol via the DMA. The The only concern to a distiller is that sugar is
14
13
12
Rank: 5
R2 = 96.2
11
RMSECV = 0.674
NIR predicted value
10
9
8
7
6
5
4
3
2
1
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Figure 16a. Cross validation result for ethanol content in 30 fermented corn mash samples. The reference method of analysis was
HPLC. The values of R2 and RMSECV were 0.962 and 0.67% v/v, respectively.
12
11
Rank: 6
10 R2 = 99.73
NIR predicted value
9 RMSECV = 0.176
8
7
6
5
4
3
2
1
0
0 1 2 3 4 5 6 7 8 9 10 11 12
being converted and is used up by the yeast. If optimize the amount of wash water and
sugar values are not decreasing then the detergent necessary for clean fermentations by
fermentation parameters must be quickly correlating the wash out procedure with the
investigated and proper actions taken to correct levels of lactic acid produced at the end of
the problem. Figure 17 shows the carbohydrate fermentation, thus determining which wash out
concentration over a 140 hr period, monitored procedure is most cost effective.
using NIR spectroscopy. NIR spectroscopy is a flexible analytical tool
Another important parameter in corn because its results are based on the best primary
fermentation that NIR spectroscopy can monitor methods available, distillation – DMA for
is the production of lactic acid. The primary ethanol, and HPLC for sugars and lactic acid.
method of calibrating the NIR for lactic acid is The predictive errors of the calibration models
HPLC. Lactic acid is a compound produced by currently used to monitor corn fermentations at
Lactobacillus sp., bacteria which compete with the Hiram Walker & Sons plant are shown in
yeast for sugar. The more lactic acid produced, Table 2.
the more bacterial contamination is present.
Figure 18 illustrates the amount of lactic acid Table 2. RMESCV values for fermentation parameters.
produced over time in a typical batch
fermentation process. It was noted that as the Component Approximate RMESCV
backset stillage use rate increased in the
Ethanol 0.14
fermentor, so did the starting level of lactic acid. Dextrins 0.50
By monitoring lactic acid, a distiller can therefore Dextrose 0.46
adjust backset stillage rates in order to optimize Maltose 0.52
fermentations. For continuous fermentations, the Lactic acid 0.11
lactic acid values can be used to determine when Glycerol 0.07
antibacterial products are required or wash out
procedures should commence.
A potential application for NIR technology is Troubleshooting is one of the largest paybacks
optimization of fermentor cleaning. One can for an NIR spectrometer. Over the course of one
4.0
3.5
3.0 Dextrose
Carbohydrate (%)
Maltose
2.5
Dextrins
2.0
1.5
1.0
0.5
0.0
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140
Hours of fermentation
Figure 17. Typical reduction in carbohydrates in corn fermentations over a 140 hr period.
Understanding near infrared spectroscopy and its applications in the distillery 163
1.10
1.00
0.90
Lactic acid (%)
0.80
0.70
0.60
0.50
0.40
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140
Hours of fermentation
Figure 18. Production of lactic acid during the course of corn fermentations.
year, one lab technician can measure 5600 corn fermentor with a nitrogen supplement added
mash samples with just one NIR spectrometer. finished in 50 hrs while the control fermentors
There are 30 batch fermentors at the Hiram finished in 80 hrs. A downside of the nitrogen
Walker & Sons facility, and all are kept full of supplement, however, was that if the control
mash until the mash is sent to the beer well. fermentors were given enough time, they would
Today Hiram Walker & Sons laboratory monitors finish with more alcohol than the nitrogen-fed
the fermentors at 12–40 hr intervals and can fermentors. This is because the nitrogen induced
predict the final percent alcohol. If ethanol yeast cell growth from the carbohydrate source
values are lower than expected, a course of action instead of alcohol production. In the end, it was
can be determined. Troubleshooting of not economical for this distillery to use nitrogen
fermentors involves checking records for correct as a supplement, since 80 hrs of fermentation
enzyme or yeast addition, or auditing the plant was acceptable. However, if production capacity
for mechanical failures. An example of ever needed to be increased, or if the gravity of
mechanical failure is an increase in lactic acid fermentation was increased, then a nitrogen
values and a decrease in ethanol in one supplement could be an asset to this operation.
fermentor, in which case the Butterworth washer The NIR analysis allowed the distillery get
must be closely inspected to determine if it is analytical results quickly, utilize personnel to
rotating properly. As another example, if the their maximum potential, and optimized
alcohol content is too low, check for small leaks production.
in the cooling coils. NIR technology can assist NIR is an asset in determining the efficiency
in identifying and fixing problems in the early of fermentations. For example, incoming corn
stage before yield is compromised. was monitored by NIR spectroscopy and found
The use of NIR spectroscopy for fermentation to contain 62.6% starch (as is). This means that
analysis allows the distillery to rapidly perform there were 626 kg of starch per metric tonne of
research on fermentation supplements. One corn. The 626 kg of starch should produce 696
example is experimenting with various forms of kg of dextrose if it were completely converted.
nitrogen. Nitrogen is the limiting substrate for The theoretical ethanol yield from yeast is 51.1%
yeast growth in fermentations. Figure 19 shows ethanol; therefore, the theoretical yield from
NIR results from nitrogen supplement 1000 kg of this grain is 451 liters. If scales are
experiments. It was determined that nitrogen at in place to measure the amount of corn that enters
a specific level decreased fermentation time the fermentor, then the efficiency of
(increased rate of fermentation) by 30 hrs. A fermentations can be calculated. If a 200,000
164 D. Livermore, Q. Wang and R.S. Jackson
11
10
9
Ethanol (%) 8
7
6
Nitrogen
5 Supplement
4 Control
3
2
1
0
0 10 20 30 40 50 60 70 80
Hours of fermentation
liter fermentor contains 50 metric tonnes of corn or the condensed syrup discharged from the
and finishes at 10.0% alcohol, then final yield is evaporators. After the third effect of a falling
400 liters of absolute alcohol per tonne of grain. film evaporator, the desired level of solids is
But theoretical yield is 451; therefore, this around 40%. If the solids level gets much higher,
fermentation was 89% efficient. it can cause fouling and eventual plugging of
The efficiency of the distillation column can the discharge lines. It is very expensive to clean
also be determined. If the above fermentor pipes plugged with solid material; therefore it is
contained 20,000 liters of ethanol, and the tank highly desirable to have a quick and easy
that recovered the ethanol contained 20,500 liters method to determine the percent solids. NIR has
at 96.0% ethanol, then the distillation efficiency the ability to measure the percent moisture of
was 98.4%. the syrup in a Pyrex beaker at very warm
Calibration models for rye mash, barley mash, temperatures with the integrating sphere
and molasses have also been developed at accessory. After the percent moisture is
Hiram Walker & Sons distillery. Optimization determined, simply subtract this value from 100
studies, efficiency determinations, and to get the percent solids. At Hiram Walker &
troubleshooting currently are carried out for Sons, the RMSECV for this calibration model is
these types of fermentations. A future calibration +/- 0.8%.
model also possible using NIR technology is The most important application for NIR
determination of the percent solids in mash prior analysis in the dryhouse area is DDGS analysis.
to exiting the pre-liquefaction tank or prior to After the rotary driers, most distilleries will
fermentation. This can help quickly identify if guarantee to their customers minimum levels for
there are milling problems or belt scale protein and fat; while there are maximum levels
problems. for moisture, fiber, and ash. Protein levels for
DDGS are probably the most important
guarantee for the end customer. The most
DRYHOUSE OPERATIONS accurate wet chemical method for protein is the
Kjeldahl method. Figure 20 illustrates the cross
There are several applications for NIR analysis validation for a calibration model using the
in the dryhouse. The first application is the Kjeldahl method as the reference method.
determination of the percent solids in solubles, Moisture analysis is also important for the
Understanding near infrared spectroscopy and its applications in the distillery 165
quality of DDGS. To get optimum results, the Table 3. RMSCV for DDGS components.
most accurate wet chemical procedure for
moisture analysis is the Bidwell-Sterling method, Component Number of Approximate RMSCV
as compared to the oven moisture method. The factors (+/- on a % as is basis)
procedure for the Bidwell-Sterling method Protein 7 0.46
involves refluxing 50 g of DDGS in 200–250 Moisture 5 0.35
mL of xylene for 2 hrs. There is a graduated Fat 6 0.35
side arm that collects the water (more dense) Fibre 5 0.49
from the DDGS, and the remaining (less dense) Ash 6 0.13
xylene falls back into the boiling flask. At Hiram
Walker & Sons the RMSECV of the NIR
calibration model for moisture using Bidwell- A future opportunity for NIR analysis in the
Sterling as the reference method is +/-0.35%. dryhouse is the determination of lactic acid in
The three remaining components in DDGS that the backset stillage that is recycled to the
can be measured by NIR are fat, fiber, and ash. fermentors. The toxic effects of backset stillage
A fat extraction unit is used to obtain the on yeast are well documented. If a distiller can
reference value for fat content in DDGS, which monitor lactic acid and control the amount of
is a 2-day test. Similarly, fiber content is backset stillage in the mash bill, then alcohol
determined with a fiber extraction unit, which production can be optimized.
requires three days. These two components will Another opportunity for NIR analysis in the
take an NIR technician a much longer time to dryhouse is to measure the residual starch in
develop the calibration models. In addition, the DDGS. After some initial preliminary trials with
cost of each of these extraction units can be quite starch analysis by cooking, converting, and
high and they require hazardous chemicals. The measuring sugar levels, it was found that this
reference method for ash is a simple test. A 2 g method can be quite complex, lengthy, and
sample is burnt in an oven for 2 hrs; and the inaccurate. If fermentations are good, then
weight loss is measured at the end. The RMSECV residual starch will not be a concern. Time is
obtained at Hiram Walker & Sons for each better spent optimizing fermentation than
DDGS component is listed in Table 3. creating calibration models for starch in DDGS.
31
30 Rank: 7
29 R2 = 91.75
NIR predicted value
28 RMSECV = 0.462
27
26
25
24
23
22
21
20
20 21 22 23 24 25 26 27 28 29 30 31
BLENDING OPPORTUNITIES AND ETHANOL the laboratory occur. The temperature effects on
DETERMINATION NIR spectra and the NIR calibration model for
ethanol in a liqueur were observed during a
In the whisky or liqueur industry it is imperative period when the Hiram Walker & Sons
that the finished product contain the correct laboratory temperature control system
strength of alcohol. Most products will have a malfunctioned. The normal liqueur samples were
target specification. For example, most whiskies often detected as abnormal samples (outliers)
have a specification of 40.0% alcohol by during this period. To resolve the issue, NIR
volume, however, governments have tolerances spectra of the liqueur samples were measured
on these specifications. For the Bureau of over a range of temperatures and incorporated
Alcohol, Tobacco and Firearms (BATF), the into the calibration model. This calibration model
specification is 40.0% minus 0.15% alcohol by is shown in Figures 21a and b. Today, the
volume, which gives a very small target range calibration can be used to determine percent
of 39.85-40.00%. Excise Canada has a larger alcohol of the liqueur accurately even when the
tolerance window of 40.0% ±0.2%, which is a laboratory temperature changes a few degrees.
target range of 39.8%-40.2%. This leaves a very Alcohol content of a liqueur can be calibrated
tiny window for a blender to make sure of the to an RMSEP of +/- 0.036%, which is inside the
correct proof of the final product. tolerance range allowed by the BATF.
To make matters more difficult, when blending To illustrate further the importance of the
whiskies, rums, tequilas, or liqueurs there are reference method, the alcohol content of
suspended solids that obscure the final percent approximately 40 whisky samples was measured
alcohol when measured by a DMA. The by three different methods: the DMA, the GC,
obscuration has to be determined prior to final and HPLC. These data were used to generate
proofing in order to determine the true percent NIR calibration models. The results are shown
alcohol. In order to compensate for the in Figures 22a, b and c. This clearly shows that
obscuration, the blender must either distill the the DMA is the most accurate and reliable
sample and proof the distillate by DMA, or oven method for a blending department, and should
dry the alcohol and determine the solids by be used as the reference method for NIR
weight and factor it into the ‘as is’ DMA result. calibrations.
Either method can be time consuming, and can For the future, the next step is to implement
hold up production for hours for the true percent in-line measurements to determine the proof of
ethanol reading. If the true percent ethanol the spirit. Currently in a blending operation, the
reading is outside the tolerance reading, then the spirit is brought in house at high proof, and then
blender must make adjustments and repeat the tested for strength. Components are mixed and
process. Laboratory analysis in a blending again tested for strength. The sample is then
department is the bottleneck of the operation, filtered for final polishing, then stored in a third
and the longer the product stays in a tank the tank. Water is the last component to be added to
more money it costs the company. In order to bring the product down to the final strength,
reduce the time in blending, the NIR technology which requires even larger tanks. If NIR were
can be a considerable asset. Currently NIR an approved method of determining alcohol
measurements are collected using 8 mm strength, it could be used in-line as a method
disposable glass vials in transmission mode; for alcohol strength in the dilution process
however a fiber optic probe can be used if immediately before bottling, thus requiring less
desired. tank space and reducing the associated costs.
Another benefit of NIR analysis is that it was An additional benefit of NIR technology is that
found that the spectrometer could be calibrated different components can be determined from
to measure ethanol over various temperature one measurement. For example, Figure 23 shows
ranges without losing much accuracy. This is the measurement of titratable acidity for an
possible only if data taken at different alcoholic spirit with added malic and citric acids.
temperatures are included in the model. Extra The problem with this product was that when it
samples should be included in the calibration was distilled for proofing, it required an
when failures in the environmental controls of additional prior step of neutralization in order
Understanding near infrared spectroscopy and its applications in the distillery 167
26
25 Rank: 3
R2 = 99.9
RMSEP = 0.038
23
22
21
20
20 20.4 20.8 21.2 21.6 22 22.4 22.8 23.2 23.6 24 24.4 24.8 25.2 25.6 26
Figure 21a. External validation of the initial model for prediction of percent ethanol on an independent set of 94 liqueur samples.
The values of R2 and RMSEP for the external validation are 0.9990 and 0.038% vv, respectively. The number of calibration
samples is 98.
26
25 Rank: 5
R2 = 99.84
NIR predicted value
24
RMSEP = 0.0416
23
22
21
20
20 20.4 20.8 21.2 21.6 22 22.4 22.8 23.2 23.6 24 24.4 24.8 25.2 25.6 26
Figure 21b. The external validation result after incorporating additional variances such as temperature. The values of R2 and
RMSEP for the external validation are 0.9984 and 0.042% vv, respectively. The number of samples in the calibration set and
independent validation sets are 123 and 120, respectively.
to achieve accurate results, which added Other examples of components that could be
significant time in the laboratory. NIR analyzed with NIR spectroscopy are fructose or
spectroscopy can measure the titratable acidity sucrose when blending liqueurs.
and the proof of the product at the same time.
168 D. Livermore, Q. Wang and R.S. Jackson
41.7
Rank: 2
41.3
R2 = 21.33
40.1
39.7
39.3
38.9
38.5
38.1
38.1 38.5 38.9 39.3 39.7 40.1 40.5 40.9 41.3 41.7
Reference method value
Figure 22a. Cross validation of 45 whisky samples where the reference percent alcohol was measured by HPLC. The RMESCV
+/- was 0.66% v/v and R2 is 0.213.
41
Rank: 2
40.8
R2 = 30.31
NIR predicted value
Figure 22b. Cross validation result of 45 whisky samples where the reference percent alcohol was measured by GC. The
RMSECV was +/- 0.28 and the R2 is 0.303.
40.85
40.75
Rank: 2
40.65 R2 = 90.51
NIR predicted value
40.55 RMSECV = 0.0663
40.45
40.35
40.25
40.15
40.05
39.95
39.85
39.75
39.65
39.55
39.55 39.65 39.75 39.85 39.95 40.05 40.15 40.25 40.35 40.45 40.55 40.65 40.75 40.85
Figure 22c. Cross validation result of 45 whisky samples where the reference percent alcohol was measured by DMA. The
RMSECV was +/- 0.067 and the R2 is 0.905.
12
11
Method:
Predicted total acids (g/l)
Rank: 12
10
R2 = 99.52
RMSECV = 0.106
9
5
5 5.5 6 6.5 7 7.5 8 8.5 9 9.5 10 10.5 11 11.5 12
Figure 23. Cross validation of the titratable acidity of a high proof alcohol sample. The RMSECV is +/-0.106 with an
R2 of 99.52.
production. There are exciting times ahead as In the foreseeable future, it is possible that
the next generation of NIR spectrometers will instrument vendors or companies that specialize
continue to improve as hardware and software in NIR model development could provide ‘turn-
become more advanced and the precision and key’ calibration for the distilling industry, leaving
accuracy of NIR technology steadily increase. more time for distillers to focus on optimization
One of biggest hurdles in implementing NIR of operations. NIR spectroscopy is here to stay;
methods is to generate reliable calibration and companies that do not utilize this technology
models after the NIR instrument is purchased. will run the risk of being left behind.
170 D. Livermore, Q. Wang and R.S. Jackson
Chapter 13
CHARLES A. ABBAS
Director of Yeast and Renewables Research, Archer Daniels Midland, Decatur, Illinois, USA
Table 1. Selected current and potential biotechnological and industrial products derived from nonconventional yeast.
Amino acids L-cystine, L-glutamic acid, L-serine, L-lysine, L-threonine, D and L-methionine; L-tryptophan, N-
acetyl-glycine, L-phenylalanine
Carotenoids Astaxanthin, carotene, lycopene
Chemicals Acetoin, adipic acid, acetanilide, dicarboxylic acids, 1,3-propanediol, ethylacetate, 2,3-butanediol, glycerol,
1-ß-hydoxybutyric acid, α-ketoglutaric acid, lactic acid
Enzymes Invertase, galactosidase, lipase, phytase, pectinases, trehalase, inulinase, glucoamylase, proteases, esterases
Feed additives SCP, SCO, phytase, ß-glucans, riboflavin, amino acids
Fermented foods Bread doughs, miso, soy sauce, ethanol
Flavors 2-phenlyethanol, nucleotides, ethylacetate, acetaldehyde, citric acid, furanones, methyl ketones
Food ingredients Citric acid, acetic acid, lactic acid, ribonucleotides, biosurfactants-sophorolids
and additives
Heterologous and Enzymes, pharmaceuticals-vaccines, immunoproteins
other proteins
Organic acids Acetic, lactic, citric, iso-citric, formic, pimelic, homogentisic acids
Nutraceuticals, N-acetylglucosamine, inositol, S-adenosyl methionine, melanin, tumor necrosis factor, human serum
pharmaceuticals factors, ß-glucans
and cosmetics
Polyols and Erythitol, glycerol, xylitol, D-arabitol, L-arabitol, D-mannitol
sweeteners
Simple sugars and Xylulose, D-psicose, D-lyxose, mannans, phosphomannans, phosphogalactans, pullulans, mannoproteins,
polysaccharides glucans, heteroxylans, exo-and acidic polysaccharides
Vitamins Riboflavin (B2), B12, thiamin, ergosterol, ubiquinone
Figure 1. Morphological differentiation and characterization of six new species of Candida. Budding cells are depicted in A, D,
G, I, K,and M. Pseudohyphae with blastoconidia are depicted in B, C, E, F, H, J, L and N. Budding cells are from 5% malt extract
agar after 3 days at 25°C while pseudohyphae are from Dalmau plate cultures on yeast morphology agar following 7 days of
incubation at 25°C. Light microscopic photo was provided courtesy of Dr. Cletus Kurtzman.
174 C.A. Abbas
Figure 2. Cell division in the ascomycetous yeast. A) Multilateral budding in Saccharomyces cerevisiae; B) Bipolar budding in
Saccharomycodes ludwigii; C) Fission in Schizosaccharomyces octosporus. Phase contrast photo was provided by Dr. Cletus
Kurtzman.
Figure 3. Representative ascospores of A) Hat-shaped (galeate), Pichia bispora; B) Spheroidal, Saccharomycopsis capsularis; C)
Elongate with a whip-like extension of the cell wall, Eremothecium (Nematospora) coryli. Bar=5 microns. Bright field photomi-
croscopic photo provided by Dr. Cletus Kurtzman.
Biotechnological applications of nonconventional yeast 175
stress the significant contribution of yeast protein electrophoresis, and biochemical and
biology to the study of evolutionary ecology metabolic flux analyses has been adopted as part
and the valuable role played by yeast in the of strain development strategies to improve NCY
interaction of the microbial communities in the strains for commercial applications. While the
ecosystem. As such, one cannot ignore the role application of these approaches in NCY lags
of yeast in climatic changes and the need for behind that of S. cerevisiae, great progress in
the research community to utilize this area has been facilitated by the major effort
multidisciplinary approaches to bridge the gaps to sequence a number of NCY of industrial and
in the fundamental knowledge of the biotechnological importance as a part of the
significance of yeast to ecological studies. Genolevures undertaking in France. This
impressive sequencing program has sought to
delineate and compare the functional genomes
NCY industrial strain development of hemiascomycetous yeast using random
programs sequence tags (RST) to obtain the maximum
biologically relevant data with a minimum of
Historically strain development programs with effort. The selected list of 13 ascomycetous yeast
NCY have followed the same path successfully included several NCY yeast of industrial and
taken with the yeast Saccharomyces cerevisiae biotechnological significance (Figure 4). A
(Benitez et al., 1996; Penttila, 1997; Beckerich detailed and a complete description of the results
et al., 1982; Phaff, 1985). These approaches of the initial phase of this effort was published
have often consisted of classical mutagenesis in FEBS Letters in December of 2000 and a
followed by screening and selection, yeast recent update of the latest progress was presented
hybridization using mating followed by at Yeast 2003 in Göteborg, Sweden (Dujon,
sporulation and tetrad analysis (ploidy 2003).
dependent) or rare mating (ploidy independent),
and the employment of parasexual genetics
using protoplast fusions (Phaff, 1985; Panchal Yeast physiology and biochemistry
et al., 1984; Beckerich et al., 1982; Benitez et
al., 1996). The advent of molecular biology NCY yeast physiology and biochemistry
gave rise to the use of new approaches in NCY continue to be active and attractive areas of
strain development with increased use of research (deWinde, 2003; Wolf, 1996; Wolf et
recombinant DNA to transform yeast via al., 2003). The greater recent attention to NCY
incorporation of genes on plasmids acting as has resulted from great strides made in our
shuttle vectors or by direct chromosomal understanding of the genetics, physiology and
integration. The development of tools for the biochemistry of the yeast S. cerevisiae. In S.
genetic manipulation of NCY is described in cerevisiae this has given rise to what is currently
greater detail in several recent texts that cover referred to as the discipline of system biology.
progress in this area (deWinde, 2003; Wolf, System biology seeks to integrate knowledge
1996; Wolf et al., 2003). Using the listed gained from genetic, biochemical, structural, and
approaches in this section numerous physiological research to develop a
homologous and heterologous genes have been comprehensive approach to our understanding
cloned and successfully expressed in most NCY of the metabolism of yeast and other organisms
of industrial importance as well as those of (Nielsen and Olsson, 2002). An essential
current interest (deWinde, 2003; Wolf, 1996; component of this approach is to provide a
Wolf et al., 2003). As newer yeast strain detailed description of the interactions among
development approaches are increasingly more all components within a cell. In order to pave
reliant on genomic and metabolic engineering the way for the use of this approach to NCY,
analyses, these approaches are also emerging research must address many of the areas listed
as new tools with NCY as well. The in Table 2. Due to space limitations the reader is
combination of genome sequencing with gene urged to consult with many of the excellent
expression profiling using microarray chips, 2D references provided at the end of this chapter.
The successful application of this approach to
176 C.A. Abbas
S. cerevisiae
S. bayanus
S. exiguus
S. servazzii
Z. rouxil
S. kluyverl
S. thermotolerans
K. lactis
K. marxianus
P. angusta
D. hansenii
P. sorbitophila
C. tropicalis
Y. lipolytica
0 20 40 60
Common genes % of total
Ascomycetes specific genes
NCY will undoubtedly have a significant impact animal sources and from distillery and yeast
on the increase in industrial and biotechnological factories (Phaff, 1985; Kurtzman and Fell, 1998;
uses of these yeast. Barnett et al., 2000). Vegetative reproduction in
this yeast occurs by budding with filaments or
Table 2. List of biochemical, cellular, metabolic, simple pseudohyphae occasionally reported.
nutritional and physiological topics currently studied in Sexual reproduction has been recognized; and
yeast. as a result of this its relatedness to Pichia jadinii
Cell wall chemistry, organization, and flocculation accepted by taxonomists. Due to some concerns
Growth kinetics about the potential pathogenicity of this yeast it
Nutritional requirements and metabolism of nutrients has been classified at a biological safety level I
Regulation of C/N metabolism (Barnett, 2000). This is in spite of the long safety
Stress responses and signal transduction record attributed to this yeast and its acceptance
Sugar utilization profiles
Sugar transport and energetics
by the US FDA as a safe substance and its
Protein synthesis, trafficking, secretion and turnover approval as a food additive (Miura et al., 1999).
Yeast organelles biogenesis and function C. utilis has been reported to produce a variety
Yeast cell structure, compartmentalization, and organization of products ranging from amino acids (L-serine;
Yeast cell cycle, control of growth and cellular division D-methionine; L-glutamic acid), glutathione, S-
Yeast aging and apoptosis adenosyl methionine, RNA (9-ß-D-ribo-
Genomics: structural, functional, comparative, evolutionary fuarnoside-5´-phosphoric acid esters of 2-
substituted-6-hydroxypurine), coenzyme A, single
cell proteins (SCP), polyols (xylitol), organic acids
NCY of biotechnological importance (acetic acid, adipic acid), sophorolipids,
ethylacetate, methyl ketone, ethanol and
CANDIDA UTILIS (ASEXUAL STATE OF PICHIA acetaldehyde. Most recently this yeast has been
JADINII) genetically engineered to produce a number of
carotenoids, which include lycopene, ß-carotene
Candida utilis (also known as Torula utilis or and astaxanthin as well as for the heterologous
Torula yeast) has been isolated from human and
Biotechnological applications of nonconventional yeast 177
production of a number of proteins including Tanaka et al. (1971) recognized the ability of C.
the enzyme α-amylase and the sweet potato tropicalis to grow on n-alkanes and most recently
protein, monellin (Miura et al., 1999; Miura et this yeast has been genetically engineered to
al., 1998; Shimada et al., 1998). A number of efficiently produce long-chain dicarboxylic acids
enzymes of this yeast have been isolated and from alkanes and fatty acids (Fabritius et al.,
characterized. These include alcohol 1996; Picataggio et al., 1992). These acids are
dehydrogenase, allantoin racemase, 2,3- of commercial interest as they can serve as
butanediol dehydrogenase, dimethylamine precursors of polyamides, fragrances, and
mono-oxygenase, dihydroxyacetone reductase, polyhydroxyalkenoates (Fabritius et al., 1996).
glutamine synthetase, invertase, maltose C. tropicalis has also been widely studied for
permease, trimethylamine mono-oxygenase, the production of xylitol from xylose derived
trehalase, urate oxidase, and urease just to name from lignocellulosics, where it has shown great
a few. C. utilis is known to assimilate inexpensive promise in achieving high rates of conversion,
sugars produced by the hydrolysis of ligno- high yield, and high concentration in pilot plant
cellulose feedstocks such as alfalfa wastes, studies (Yahashi et al., 1996; Choi et al., 2000;
cellulose, hemicellulose, pineapple waste, potato Kim et al., 1999; Oh and Kim, 1998). A number
waste, and sugarcane bagasse. It can also utilize of enzymes from this yeast have been isolated
and degrade petroleum waste, rapeseed oil meal, and characterized including alcohol oxidase,
sauerkraut waste streams and tallow. Since C. carnitine acetyl transferase, catalase, and urate
utilis, unlike S. cerevisiae, cannot produce oxidase. C. utilis has been shown to utilize and/
ethanol under strict aerobic conditions, it can or degrade hydrocarbons, kraft black liquor from
be grown in continuous culture to a high density phenol, D-xylose, petroleum, and paper mill
(Miura et al., 1999). These are desirable features waste. The inclusion of this yeast in the initial
that have been exploited in SCP, in the Genolevures comparative genomic exploration
production of extracellular enzymes and other is described by Blandin et al. (2000a). A detailed
proteins as well as in the production of update taxonomic description is found in Barnett
intracellular proteins and other metabolites such et al. (2000) on pages 270-271.
as carotenoids by this yeast. A recent taxonomic
description of this yeast can be found in Barnett
et al. (2000) on pages 532-533. CANDIDA FAMATA (ASEXUAL STATE OF
DEBAROMYCES HANSENII)
and salt tolerances as well as its ability to grow mannitol, xylitol and erythritol, lipases active at
on hydrocarbons and the pentose sugars D- pH 5.5 and 7.5, lipases active at pH 9.0 and
xylose and L-arabinose that are present in 60°C, lipase activators, proteases, alkaline
hydrolysates of lignocellulosics (Roostita and protease, melanin, and single cell protein have
Fleet, 1999). Flocculation has been described all been reported for this yeast (Kyong and Shin,
for this yeast in the presence of peptone and its 2000; Delgenes et al., 1998; Silva et al., 1996;
involvement in the formation of biofilms has Rodrigues et al., 1998). One of the many
been noted (Martinez et al., 1996). Recently a interesting characteristics of this yeast is the
transformation system has been developed and ability to hydrolyze and reduce zearalenone, as
optimized for this yeast (Voronovsky et al., well as the ability to degrade other aromatic
2002) and the use of PCR to identify and compounds, fatty acids, and alkanes such as
distinguish this yeast from other Candida has azelaic acid, benzo (α) pyrene, biphenyls,
been published (Nishikawa et al., 1997; 1999). hydrocarbons, naphthalene, petroleum, rapeseed
Due to the inclusion of D. hansenii in the oil, and tallow (Hassan anbd El-Sharkway, 1999;
Genolevures comparative genomic project, the De Felice et al., 1997). The above properties
genomic sequence of this yeast has been have been exploited in the production of organic
delineated and compared to a number of other acids and enzymes using ethanol, glucose, fatty
ascomycetous yeast (Lepingle et al., 2000; acids, or hydrocarbons as the primary carbon
Dujon, 2003). Recently, attempts to characterize substrates (Arzumanov et al., 2000; Antonucci
and isolate genes responsible for salt tolerance et al., 2001; Wache et al., 2003; Pazouki et al.,
in D. hansenii have been reported (Ramos et al., 2000). More recently the development of an
2003). Understanding the underlying efficient transformation system for Y. lipolytica
mechanism for salt tolerance in this yeast can has aided further in the expansion of the uses of
have a significant impact on its further this yeast to the production of proteins such as
commercial exploitation as it has been used for enzymes and pharmaceuticals (Chang et al.,
over a decade for the large-scale commercial 1998; Barth and Gaillardin, 1996). The inclusion
fermentation production of the vitamin riboflavin of this yeast in the Genolevures 1 and 2
(B 2 ) (Abbas, 2001). A complete recent comparative genomics project has aided in its
description of D. hansensii is provided by genomic analysis and comparison to other
Barnett et al. (2000) on pages 336-337. hemiascomycetous yeast (Dujon, 2003;
Casaregola et al., 2000). A complete recent
taxonomic description is provided by Barnett et
YARROWIA LIPOLYTICA (ASEXUAL STATE OF al. (2000) on pages 785-786.
CANDIDA LIPOLYTICA)
citric acid, xylitol from D-xylose, L-arabitol from information to support its relatedness to the other
L-arabinose, as well as tea cider (Saha and hemiascomycetous yeast, K. marxianus
Bothast, 1996). Two enzymes that have been (Bolotin-Fukuhara et al., 2000; Dujon, 2003).
isolated and well characterized from this yeast The taxonomic description of this yeast has been
are alcohol dehydrogenase and deaminase. Due described by Barnett et al. (2000) on pp. 417-
to the interest in the production of xylitol by C. 418.
guilliermondii, the impact of acetic acid and
sugar degradation products present in a number
of lignocellulosic hydrolysates from the two KLUYVEROMYCES MARXIANUS (SYNONYM K.
enzymes involved in D-xylose metabolism in FRAGILIS AND ASEXUAL STATE CANDIDA
this yeast, xylose reductase and xylitol KEFYR)
reducatase, has been extensively studied.
Growth of this yeast in brine and on cellulose K. marxianus is often associated with spoilage
hydrolysates has been reported. The recent use of commercial yogurt and has been commonly
of this yeast as a biocontrol agent to reduce isolated from other dairy products such as milk
postharvest fruit spoilage has been described. and cheese, pressed commercial yeast, effluent
Genetic transformation of C. guilliermondii has of sugar refineries, bread doughs, beer,
been reported (Boretsky et al., 1999) and a fermenting figs, sewage, infected humans and
detailed description of the genetics of this yeast infected milk cows, where it causes bovine
provided by Sibirny (1996a). A complete mastitis. In spite of the association of K.
description of the taxonomy of this yeast is marxianus with infected humans and dairy cows,
provided in Barnett et al. (2000) on pages 523- it is considered GRAS and frequently referred
524. to as the ‘dairy yeast’. Vegetative reproduction
in this yeast occurs via budding; and simple to
elaborate pseudohyphae are produced. Due to
KLUYVEROMYCES LACTIS the production of ascospores, K. marxianus is
currently classified as the perfect state of C. kefyr,
Kluyveromyces lactis is another milk-sugar while the closely related K. fragilis is listed as
fermenting yeast that has been isolated from a one of its many synonyms (Barnett et al., 2000).
wide range of sources, which include dairy When cultured on corn meal or potato dextrose
factories, wineries, human sources, and plant agar for 8-11 days, long filaments are observed
sources such as the slime of an oak tree. This microscopically. K. marxianus has been
yeast been shown to produce ethanol from whey reported to produce ethanol from cheese whey,
produced from cheese manufacturing. Some oligonucleotides for use as flavor enhancers,
strains of this NCY are known to harbor yeast polyphosphates, metallothionein, glycerol,
protein killer factors. Genetic studies of this yeast single cell protein from biomass, several
have led to the development of efficient enzymes for commercial application (lactase and
transformation systems for the production of pectinases), pectin extraction from fruits,
proteins such as the enzyme ß-galactosidase and polyphosphate, and glycerol (Parrondo et al.,
the proteinase calf chymosin, as well as many 2000; Ferarri et al., 1996; Belem et al., 1997;
other proteins that are listed in the extensive Belem et al., 1998). In addition to lactases (ß-
review by Wesolowski-Louvel et al. (1996). galactosidase) and pectinases, several other
Most of these proteins are commercially enzymes have been isolated, characterized, and
important in pharmaceutical applications. The cloned including inulinase, alcohol
metabolism of galactose, high cell density dehydrogenase, acetoin dehydrogenase, and
cultivation, the pattern of N-protein UDP glucose-4-epimerase (Schwan et al., 1997).
glycosylation, and its codon usage choice, which Interesting characteristics of this yeast include
is similar to that of S. cerevisiae, have been well the production of killer factors and growth and
studied (Wesolowski-Louvel et al., 1996). production of ethanol from cheese whey and
Another recent development with this NCY is inulin at 45°C to a final concentration of 7.0%
its inclusion in the Genolevure comparative (w/v) (Balletsteros et al., 1993; Singh et al.,
sequencing efforts, which has provided further 1998; Banat et al., 1995). K. marxianus can
180 C.A. Abbas
grow on a wide range of carbon sources and Castoria et al., 1997). The involvement of 1,3-
waste streams such as lactic acid, sauerkraut ß-glucanases as the mode of action against
waste, orange peel, paper mill sludge, and cocoa spoilage fungi has been implicated as the
pulp. A simultaneous saccharification and protective mechanism. A complete description
fermentation process to produce ethanol that of this yeast is provided by Barnett et al. (2000)
combines fungal cellulase enzymes with this on pages 315-316.
yeast has been reported (Lark et al., 1997). The
results have indicated that further development
with the genetic engineering of K. marixanus to ZYGOSACCHAROMYCES ROUXII
ferment lignocellulosics to ethanol is feasible
and attractive given this yeast’s ability to grow Zygosaccharomyces rouxii is one of the most
on a variety of feedstocks and its thermo- common food and beverage spoilage yeast.
tolerance. K. marxianus has been shown to Vegetative reproduction of this yeast occurs via
accumulate toxic compounds including budding with the presence or absence of
cadmium, copper, and silver, and has also been filaments or simple pseudohyphae. Persistent
used in adhesive testing. The controversy over asci formation containing up to four rough or
the relatedness of this yeast to K. lactis is cited smooth sound ascospores has been noted for this
by Wesolowski-Louvel et al. (1996) and Barnett yeast. Among the many products of this yeast
lists K. lactis and K. marxianus separately as are acetic acid, formic acid, ethanol, alkyl esters,
several lines of evidence clearly distinguish L-glutamine, mannans, miso, 4-hydroxy-
between these two yeasts. Both of these yeasts furanone, soy sauce, lipases that are active at
were analyzed as a part of the Genolevures pH 9.0 at 60°C, UDP-N-acetylglucosamine, and
comparative genomic sequencing and results 5´-nucleotides (Hayashida et al., 1998).
reported (Bolotin-Fukuhara et al., 2000; Llorente Important characteristics of this yeast are its
et al., 2000). A complete description of this NCY resistance to high salt, sugar, sorbic acid, and
is provided in Barnett et al. (2000) on pages tolerance to high pressure, which contribute to
420-421. its survival in and spoilage of preserved foods
and beverages. Due to this property, this yeast
is increasingly being targeted for
CRYPTOCOCCUS LAURENTII biotechnological applications (Limtong et al.,
1998). The inclusion in the Genolevures
Cryptococcus laurentii is commonly associated comparative genomics effort will propel further
with cereal grains, grapes, soil, birds, decaying rapid developments in the genetic manipulation
plant matter, air, seawater, and tropical plants. and exploitation of Z. rouxii (De Montigy et al.,
Vegetative reproduction occurs by budding and 2000; Sychrova et al., 2000). The isolation of
filaments are present or absent with the formation ura3 mutants and the development of an efficient
of simple and/or septate hyphae. Sexual transformation system were recently described
reproduction in this yeast has not been reported. by Pribylova and Sychrova (2003). A complete
C. laurentii is known to produce a number of description of the taxonomy of this yeast is
exo- and acidic polysaccharides, xylanases and provided in Barnett et al. (2000) on page 797.
glycosidases, a thermostable glutaminase, killer
toxins, L-lysine, and a number of diol
compounds. Other characteristics of this yeast CANDIDA BOIDINII
include its alkali tolerance and its ability to
degrade benzene derivatives. Due to C. Candida boidinii have been isolated from a wide
laurentii’s ability to produce copious levels of range of geographic locations from a number
extracellular polysaccharides, there has been of sources such as tanning solutions, fermented
great interest in developing food and non-food acidic products, swamps and other water habitats
applications for their use (De Baets and and flowers. Vegetative reproduction occurs by
Vandamme, 1999; Nadezda et al., 1997). budding with the formation of simple to
Another area that has received commercial elaborate pseudohyphae. No sexual reproduction
interest is use of this yeast as a biocontrol agent stage has been described for this yeast. C.
to retard fruit spoilage (Giuseppe et al., 1998; boidinii grows on and degrades methanol as well
Biotechnological applications of nonconventional yeast 181
as pectin and can produce a number of products Dauglis, 2000; Minning et al., 2001; Trujillo et
of interest such as the polyols xylitol and L- al., 2001; Rodriguez et al., 1996; Sberna et al.,
arabitol, xylulose, SCP, nucleotides such as ATP, 1996). In addition to its ability to grow on
ethanol, D-amino acids, as well as a number of methanol and other oxygenated hydrocarbons,
pigments (Vandeska et al., 1996; Suryadi et al., P. pastoris can degrade hydrogen peroxide and
2000; Yamada et al., 2001). Several enzymes can be used to produce acetaldehyde and SCP.
have been isolated and characterized from this Due to space limitations, the reader is directed
yeast including NAD-linked alcohol to excellent recent reviews of this yeast that
dehydrogenase, polyamine oxidase, formate describe in greater detail its biotechnological
dehydrogenase, diamino acetyltransferase, applications (Gellissen, 2000; Sreekrishna and
dihydroxyacetone synthase, and S- Kropp, 1996). Comparative genomic sequence
formylglutathione hydrolyase (Labrou et al., data comparison from inclusion of this yeast in
2001). Due to C. boidinii’s ability to grow on Genolevures was reported (Blandin et al., 2000)
methanol, the metabolism of this carbon source A recent description of this yeast is provided by
by this yeast has been extensively studied Barnett et al. (2000) on page 554.
(Yurimoto et al., 2000, Sakai et al., 1998;
Gabrinska-Loniewska, 1997; Aggelis et al.,
1999). The development of genetic tools and a PHAFFIA RHODOZYMA (ANAMORPH OF
transformation system have enabled the cloning XANTHOPHYLLOMYCES DENDRORHOUS)
and expression of a number of heterologous
proteins in this yeast (Sakai et al., 1996; Phaffia rhodozyma is a carotenogenic yeast that
Nakamura et al., 2000). Significant interest has was originally isolated from the stump of birch
also been shown in the use of this yeast for the trees in Japan, Finland and Russia and from an
commercial production of dicarboxylic acids. A alder tree in Japan. This yeast reproduces
recent taxonomic description of C. boidinii can vegetatively by budding with formation of
be found in Barnett et al. (2000) on page 145. filaments with simple pseudohyphae present or
absent. Sexual spores or basidiospores are borne
on basidia. The current taxonomic classification
PICHIA PASTORIS of P. rhodozyma as an anamorph of X.
dendrorhous has not been accepted by Kurtzman
Pichia pastoris can be found in exudates of and Fell (1998). Due to its ability to produce the
hardwood trees such as black oak and chestnut. carotenoid astaxanthin, which is a valuable
Vegetative reproduction in this yeast is by additive in salmon diets, this yeast has been of
budding and no filaments are formed. The important commercial interest. The availability
formation of asci containing up to 4 hat-shaped of improved strains of P. rhodozyma that can
ascospores have been noted in this yeast (Barnett economically produce astaxanthin intra-
et al., 2000). P. pastoris has attracted much cellularly has aided in the development of
interest in biotechnological applications that industrial scale high cell density fermentations
involve heterologous proteins. Its ability to grow (Abbas, 2001). The production of other
on methanol as a carbon source has been carotenoids by P. rhodozyma has also been
extensively characterized and the peroxisomal reported. Cultivation of this yeast to produce
system described in great detail. The astaxanthin from several nonconventional
development of genetic tools for this yeast has carbon sources such as wood hydrolysates, date
led to the rise in the use of commercial laboratory juice, glycerol, D-xylose, a number of
kits and the demonstration of high-level disaccharides, as well as on dried distillers grains
production of a number of pharmaceutical and produced from ethanol dry mill plants has also
other proteins by P. pastoris. In a properly been noted (Vazguez et al., 1998; Parajo et al.,
designed fermentation process this yeast can be 1998a,b; Fontana et al., 1997; Santopietro et al.,
cultivated to high density, which is key to 1998; Ramirez et al., 2000; Chan and Ho, 1999;
demonstrating commercial feasibility of Cruz and Parajo, 1998; Florencio et al., 1998;
processes that are dependent on optimizing cell Kesava et al., 1998). A transformation system
mass production (Chung, 2000; D’Anjou and has also been developed for this yeast and
Dauglis, 2001; Thorpe et al., 1999; D’Anjou and improved recently (Wery et al., 1998; Visser,
182 C.A. Abbas
2003). A detailed taxonomic description of this as palm wine, arak, beer breweries, and from
yeast is provided in Barnett et al. (2000) on page sulphited grape juice, grapes, cane sugar, and
784. molasses. This yeast is fairly distinguishable
using microscopy by its large size and by its
vegetative reproduction by splitting or fission.
HANSENULA POLYMORPHA (SYNONYM PICHIA The production of filaments in this yeast has not
ANGUSTA) been observed. Sexual reproductive asci are
produced with up to four smooth, oval, or round
Hansenula polymorphia has been isolated from ascospores reported (Barnett et al., 2000). Due
a number of sources in different parts of the to its unique morphological, biochemical,
world, which include spoiled orange juice, maize physiological, and metabolic features, S. pombe
meal, the fruit fly Drosophila pseudoobscura, has received much recent attention for
milk from cows with mastitis, and soil irrigated biotechnological applications. These
with a distillery effluent. Vegetative reproduction applications have taken advantage of its high
occurs by budding and no filaments are formed. osmotolerance, high temperature tolerance and
Asci that contain up to 4 hat-shaped ascospores the development of genetic tools for its
have been described for this yeast. H. exploitation (Dhamija et al., 1996; Chaudhari
polymorpha is characterized by its excellent and Chincholkar, 1999). Among the many
growth on short chain oxygenated hydrocarbons products that can be produced with S. pombe
such as ethanol and methanol as well as other are ethanol, SCP, and heterologous proteins for
sugar alcohols or polyols, which include use in food and pharmaceutical applications
glycerol, erythritol, xylitol, and L-arabitol (Amanchukwu et al., 1989; Fukuda et al.,
(Sanchez et al., 1998; Suryadi et al., 2000). Due 1984). An excellent recent text by Giga-Hama
to its growth on methanol, the physiology and (1997) describes many of the heterologous
biochemistry of this yeast has been well studied proteins that can be produced by this yeast. A
and the metabolism of this carbon source is well detailed taxonomic summary is provided by
understood. Based on the ability of this yeast to Barnett et al. (2000) on pages 678-679.
grow on methanol and ethanol, alcohol
biosensors have been described that utilize
coupled oxidase-peroxidase enzymes OTHER NONCONVENTIONAL YEAST OF
(Vijayakumar et al., 1996). The development of INTEREST
genetic tools for H. polymorpha has given rise
to the genetic manipulation of this yeast and the In addition to the NCY described in detail in the
production of a number of heterologous proteins previous section there are many others that are
of commercial interest such as maltase, L- of current interest to the biotechnology industry.
rhamnosidase, and phytase (Liiv et al., 2001; These include several that can produce ethanol
Mayer et al., 1999; Yanai and Sato, 2000). This from sugars present in lignocellulosic
has led to numerous biotechnological hydrolysates such as Candida shehatae,
applications of this NCY and due to space Candida wickerhamii, Pichia stipitis, and
limitations the reader is urged to consult with Pachysolen tannophilus (Sreenath and Jeffries,
excellent recent reviews cited at the end of this 1999; Sanchez et al., 1999; Amartey and Jeffries,
chapter (Hansen and Hollenberg, 1996; 1996; Possoth et al., 1996; Kruse and Schuegerl,
Hollenberg et al., 1997; Gellissen, 2000). A 1996; Skory et al., 1996; Freer et al., 1998). The
detailed taxonomic description is provided in amylolytic NCY Debaromyces (Schwann-
Barnett et al. (2000) on page 493. iomyces) occidentalis, Lipomyces starkeyi and
Aruxla adeninivorans have been recently tapped
for their ability to grow on starch and have other
SCHIZOSACCHAROMYCES POMBE (SYNONYM attributes that can be exploited further for a wide
SACCHAROMYCES POMBE) range of biotechnological applications (Kunze
and Kunze, 1996; Wartmann and Kunze, 2000;
Schizosaccharomyces pombe was originally Laires et al., 1983). The yeast Candida
isolated from East African millet beer and since bombicola, C. rugosa, and C. maltosa can grow
then from a variety of alcoholic beverages such
Biotechnological applications of nonconventional yeast 183
on n-alkanes, fatty acids, waste animal fat or employment of metabolic controls aided by
waste vegetable oils to produce products such increased automation will fuel much of the
as single cell oils, biosurfactants, and 1,3- progress in the uses of NCY in industrial and
butanediol (Sharyshev et al., 1997; Montesinos biotechnological applications. With these
et al., 2003; Matsuyama et al., 2001; Deshpande developments in mind, the next decade promises
and Daniels, 1995; Rau et al., 1999; Daniel et to be an exciting era for the biotechnological
al., 1998; Rau et al., 2001; Lang et al., 1996; and industrial uses of NCY.
Daniel et al., 1999; Mauersberger et al., 1996).
The NCY Candida krusii, Candida magnolia
and Pichia farinosa can produce polyols such References
as glycerol and erythritol (Abbas, unpublished;
Sung-Wook et al., 1999; Nakano et al., 2000; Abbas, C.A. 2001. Industrial applications of
Yamazaki et al., 2000). The methylotrophic yeast nonconventional yeast. Plenary presentation
Pichia methanolica can grow on methanol abstract, ISSY 2001, Lviv, Ukraine.
(Sibirny, 1996). The yeast Pichia fermentans can Aggelis, G., S. Fakas, S. Melissis and Y.D. Clonis.
produce the flavoring compound, 2- 1999. Growth of Candida boidinii in a
phenylethanol (Huang et al., 2000). Several methanol-limited continuous culture and the
NCY, which include Rhodotorula aurantiaca, formation of methanol-degrading enzymes. J.
R. rubra, and R. glutinis have been evaluated as Biotech. 72(1/2):127-139.
biocontrol agents to prevent fruit harvest decay Ahmed, Z., H. Sasahara, S.H. Bhuiyan, T. Saiki,
while the yeast Pichia anomala has been shown T. Shimonishi, G. Takada and K. Izumori.
to inhibit mold growth during cereal grain 1999. Production of D-lyxose from D-glucose
storage (Chand-Goyal et al., 1996; Sugar and by microbial and enzymatic reactions. J.
Spotts, 1999; Lima et al., 1998). Rhodotorula Bioscience and Bioeng. 88(6):676-678.
glutinis and R. rubra are also known for their
Amanchukwu, S.C., A. Obafemi and G.C.
production of the carotenoids ß-carotenes and
Okpokwasili. 1989. Single-cell protein
lycopene (Bhosale and Gadre, 2001; Matelli et
al., 1990; Buzzini, 2000; 2001). Several NCY production by Schizosaccharomyces pombe
are well known pathogens such as Candida isolated from palm wine using hydrocarbon
albicans and Filobasidiella (Cryptococcus) feedstocks. Folia Microbiol. (Prague)
neoformans and therefore have received for 34(2):112-119.
many decades much attention from the medical Amartey, S. and T. Jeffries. 1996. An
and the pharmaceutical industries (Demain et al., improvement in Pichia stipitis fermentation of
1998). Finally, many NCY have been recognized acid-hydrolyzed hemicellulose achieved by
for their role in food spoilage. Some of the many overliming (calcium hydroxide treatment) and
NCY that are involved in food spoilage are strain adaptation. World. J. Microbiol.
members of the genera Brettanomyces, Pichia, Biotechnol. 12(3):281-283.
Candida, Torulaspora, Metschnikowia, and Antonucci, S., M. Bravi, R. Bubbico, A.D.
Hansenisapora (Grbin and Henschke, 2000). Michele and N. Verdone. 2001. Selectivity in
citric acid production by Yarrowia lipolytica.
Enzyme and Microbial Tech. 28(2/3):189-195.
Future predictions and conclusions Arzumanov, T.E., N.V. Shishkanova and T.V.
Finogenova. 2000. Biosynthesis of citric acid
Over the next decade the commercial value of by Yarrowia lipoltyica repeat-batch culture on
all products derived from the biotechnological ethanol. Appl. Microbiol. Biotechnol. 53
applications of NCY will surpass all those (5):525-529.
derived from the current uses of food/fuel/feed/ Balletsteros, J., J.M. Oliva, M. Ballesteros and
chemical as well as pharmaceutical and
J. Carrasco. 1993. Optimization of the
nutraceutical uses of S. cerevisiae (Abbas, 2001).
simultaneous saccharification and
Rapid advances in metabolic engineering in
combination with bioinformatics and improved fermentation process using thermotolerant
fermentation processes from the greater yeasts. Appl. Biochem. Biotech. 39/40:201-
211.
184 C.A. Abbas
Chung, J.D. 2000. Design of metabolic feed De Montigy, J., M-L. Straub, S. Potier, F. Tekaia,
controllers: application to high-density B. Dujon, P. Wincker, F. Artiguenave and J-L.
fermentations of Pichia pastoris. Biotech. Souciet. 2000. Genomic exploration of the
Bioeng. 68(3):298-307. hemiascomycetous yeasts: 8.
Cruz, J.M. and J.C. Parajo.1998. Improved Zygosaccharomyces rouxii. FEBS Lett.
astaxanthin production by Xanthophyllomyces 487(1):52-55.
dendrorhous growing on enzymatic wood Deshpande, M. and L. Daniels. 1995. Evaluation
hydrolysates containing glucose and of sophorolipid biosurfactant production by
cellobiose. Food Chem. 63(4):479-484. Candida bombicola using animal fat.
Daniel, H-J, R.T.Otto, M. Reuss and C. Syldatk. Bioresource Technology 54:143-150.
1998. Sophorolipid production with high De Winde, J.H. (ed.). 2003. Topics in Current
yields on whey concentrate and rapeseed oil Genetics: Functional Genetics of Industrial
without consumption of lactose. Biotech. Yeasts. Springer-Verlag, Berlin, Germany, pp.
Letters 20(8):805-807. 1-367.
Daniel, H-J., R.T. Otto, M. Binder, M. Reuss and Dhamija, S.S., D. Bhaskar and R. Gera. 1996.
C. Syldatk. 1999. Production of sphorolipids Thermotolerance and ethanol production (from
from whey: development of a two-stage 15% sucrose at 37°C) are at variance in mutants
process with Cryptococcus curvatus ATCC of Schizosaccharomyces pombe. Biotech.
20509 and Candida bombicola ATCC 22214 Letters 18(11):1341-1344.
using deproteinized whey concentrates as Dujon, B. 2003. Comparative genomics of
substrates. Appl. Microbiol. Biotechnol. 51:40- Hemiascomycetous yeasts: the systematic
45. sequencing of Candida glabrata,
D’Anjou, M and A.J. Daugulis. 2000. Mixed- Kluyveromyces lactis, Debaromyces hansenii
feed exponent ional feeding for fed-batch and Yarrowia lipolytica. Plenary lecture 2-5
culture of recombinant methylotrophic yeast. abstract.
Biotech. Letters 22:341-346. Fabritius, D., H.J. Schaefer and A. Steinbuechel.
D’Anjou, M and A.J. Dauglis. 2001. A rational 1996. Identification and production of cis-3-
approach to improving productivity in hydroxy-1, 18-octadec-9-enedioic acid by
recombinant Pichia pastoris. Biotech. Bioeng. mutants of Candida tropicalis. Appl.
72(1):1-11. Microbiol. Biotechnol. 45(3):342-348.
De Baets, S. and E.J. Vandamme. 1999. Yeasts Ferrari, M.D., L. Loperena and H. Varela. 1996.
as producers of polysaccharides with novel Ethanol production from whey permeate using
application potential. SIM News 49(6):321- fed-batch culture of Kluyveromyces fragilis.
328. Biotech. Letters 16(2):205-210.
De Felice, B., G. Pontecorvo and M. Carfagna. Fleet, G. 2000. The biodiversity of yeasts in the
1997. Degradation of waste waters from olive production of foods and beverages. Plenary
oil mills by Yarrowia lipolytica ATCC 20255 presentation abstract, ISY 2000, Arnhem, The
and Pseudomonas putida. Acta Biotechnol. Netherlands.
17(3):231-239. Florencio, J.A., C.R. Soccol, L.F. Furlanetto and
Delgenes, J.P., M.C. Escare, J.M. Laplace, R. T. Bonfim.1998. A factorial approach for a
Moletta and J.M. Navarro. 1998. Biological sugarcane juice-based low cost culture
production of industrial chemicals, i.e. xylitol medium: increasing the astaxanthin production
and ethanol from lignocelluloses by controlled by the red yeast Phaffia rhodozyma.
mixed culture systems. Industrial Crops and Bioprocess Eng.19(3):161-164.
Products. 7(2/3):101-111. Fontana, J.D., M.B. Chocial, M. Baron, M.F.
Demain, A.L., H.J. Phaff and C.P. Kurtzman. Guimaraes, M. Maraschin, C. Ulhoa, J.A.
1998. The industrial and agricultural Florencio and T.B. Bonfim. 1997.
significance of yeasts. In: The Yeasts, A Astaxanthinogenesis in the yeast Phaffia
Taxonomic Study (C.P. Kurtzman and J. W. rhodozyma-optimization of low-cost culture
Fell, eds). 4th edition, Elsevier, Amsterdam, The media and yeast cell-wall lysis. Appl. Biochem.
Netherlands, pp. 13-19. Biotech. 63-65:305-314.
186 C.A. Abbas
Freer, S.N., C.D. Skory and R.J. Bothast. 1998. Huang, C. Jr., S.L. Lee and C.C. Chou. 2000.
Production of fuel ethanol from cellulosic Production and molar yield of 2-phenylethanol
biomass. Recent Res. Dev. Microbiol. by Pichia fermentans L-5 as affected by some
2(1):201-210. medium components. J. Biosc. Bioeng.
Fukuda, H., T.Fujii and T. Ogawa. 1984. 90(2):142-147.
Microbial production of C-2 hydrocarbons, Kesava, S.S., G-H. An, C-H. Kim, S-K. Rhee
ethane, ethylene and acetylene. Agric. Biol. and E-S. Choi. 1998. An industrial medium
Chem. 48(5):1363-1365. for improved production of carotenoids from
Gabrinska-Loniewska, A., E. Pajor and E. a mutant strain of Phaffia rhodozyma.
Slavikova. 1997. Comparative analysis of Bioprocess Eng. 19(3):165-170.
yeast-like species potentially useful to Kim, J-H., Y-W. Ryu and J-H. Seo. 1999.
environmental biotechnology. 1997. Acta Analysis and optimization of a two-substrate
Mycologica 32(1):99-106. fermentation for xylitol production using
Gellissen, G. 2000. Heterologous protein Candida tropicalis. J. Indust. Microb. Biotech.
production in methylotrophic yeasts. Appl. 22(3):181-186.
Microb. Biotech. 54(6):741-750. Kruse, B. and K. Schuegerl. 1996. Investigation
Giga-Hama, Y. 1997. Fission yeast Schizo- of ethanol formation by Pachysolen
saccharomyces pombe: an attractive host for tannophilus from xylose and glucose/xylose
heterologous protein production. In: Foreign co-substrates. Process Biochem. 31(4):389-
Gene Expression in Fission Yeast: 407.
Schizosaccharomyces pombe (Y. Giga-Hama Kunze, G. and I. Kunze. 1996. Arxula
and H. Kumagai, eds). Springer-Verlag, adeninivorans. In: Nonconventional Yeasts in
Germany, pp. 3-28. Biotechnology (K. Wolf, ed). Springer, Berlin,
Giuseppe, L., D.C. Filippo, C. Raffaello and D.C. Germany, pp. 389-409.
Vincenzo. 1998. Activity of the yeasts Kurtzman, C.P. and J.W. Fell. 1998. The Yeasts,
Cryptococcus laurentii and Rhodotorula A Taxonomic Study (C.P. Kurtzman and J. W.
glutinis against post-harvest rots on different Fell, eds), 4th edition. Elsevier, Amsterdam, The
fruits. Biocontrol Sci. Tech. 8(2):257-267. Netherlands.
Grbin, P.R. and P.A. Henschke. 2000. Mousy Kurtzman, C. 2001. Six new anamorphic
off-flavour production in grape juice and wine ascomycetous yeasts near Candida
by Dekkera and Brettanomyces yeasts. Aust. tanzawaensis. FEMS Yeast Research 1:177-
J. Grape Wine Res. 6(3):255-262. 185.
Hansen, H. and C.P. Hollenberg. 1996. Kurtzman, C. and J. Sugiyama. 2001.
Hansenula polymorpha (Pichia angusta). In: Ascomycetous yeasts and yeastlike taxa. In:
Nonconventional Yeasts in Biotechnology (K. The Mycota VII Part A. Systematics and
Wolf, ed). Springer-Verlag, Berlin, Germany, Evolution (McLaughlin, McLaughlin and
pp. 293-311. Lemke, eds.) Springer-Verlag, Berlin,
Hassan, M.A. and S.H. El-Sharkway. 1999. Germany, pp. 179-200.
Microbial production of useful fatty acids from Kyong, S.H. and C.S. Shin. 2000. Optimized
used sunflower oil. Zagazig J. Pharm. Sci. production of L-ß-hydroxybutyric acid by a
8(2):46-49. mutant of Yarrowia lipolytica. Biotechnol. Lett.
Hayashida, Y., K. Nishimura and J.C. Slaughter. 22(13):1105-1110.
1998.The importance of the furanones HDMK Labrou, N. and D.J. Rigden. 2001. Active-site
and HEMF in the flavour profile of Japanese characterization of Candida boidinii formate
barley miso and their production during dehydrogenase. Biochem. J. 354(2):455-463.
fermentation. J. Food Sci. Agric. 78(1):88-94. Lachance, M-A. and W.T. Starmer. 1998.
Hollenberg, C.P., and G. Gellissen.1997. Ecology and yeasts. In: The Yeasts, A
Production of recombinant proteins by Taxonomic Study (C.P. Kurtzman and J. W.
methylotrophic yeasts. Current Opinions in Fell, eds), 4th edition. Elsevier, Amsterdam, The
Biotechnology 8:554-560. Netherlands, pp. 21-30.
Biotechnological applications of nonconventional yeast 187
Laires, A., I. Spencer-Martins and N. van Uden. hemiascomycetous yeasts: 12. Kluyveromyces
1983. Use of D-glucosamine and 2- marxianus var. marxianus (FEBS 24376) FEBS
deoxyglucose in the selective isolation of Letters 487(1):71-75.
mutants of the yeast Lipomyces starkeyi Martinez, X.C., A. Narbad, A.T. Carter and M.
derepressed for the production of extracellular Stratford. 1996. Flocculation of the yeast
endodextranase. Zeitschrift fur Allgemeine Candida famata (Debaromyces hansenii): an
Mikrobiologie 23(9):601-603. essential role for peptone. Yeast 2(5):415-423.
Lang, S., A. Brakemeier, U. Rau and F. Wagner. Matelli, H.L., I.M. da Silva, N.O. Souza and D.
1996. Biotechnological production of Pomeroy. 1990. Production of ß-carotene by
interfacial active glycolipids. Oils-Fats-Lipids a Rhodotorula strain grown on sugar cane
1995, Proc. World. Congr. Int. Soc. Fat. (21st) juice. Biotechnology Letters 12(3):207-208.
1:53-54. Matsuyama, A., H. Yamamoto, N. Kawada and
Lark, N., Y. Xia, C-G. Qin, C.S. Gong and G.T. Y. Kobayashi. 2001. Industrial production of
Tsao. 1997. Production of ethanol from R-1,3-butanediol by new catalysts. J. Mol.
recycled paper sludge using cellulase and Catal. B: Enzym. 11:4-6.
yeast, Kluyveromyces marxianus. Biomass and Mauersberger, S., M. Ohkuma, W-H. Schunck
Bioenergy 12(2):135-143. and M. Takagi. 1996. Candida maltosa. In:
Lepingle, A., S. Casaregola, C. Neuveglise, E. Nonconventional Yeasts Biotechnol (K. Wolf,
Bon, H-V. Nguyen, F. Artiguenave, P. Wincker ed.). Springer, Berlin, Germany, pp. 411-580.
and C. Gaillardin. 2000.Genomic exploration Mayer, A.F., K. Helmuth, H. Schlieker, R. Lopez-
of the hemiascomycetous yeasts: 14. Ulibarri, S. Oertel, U. Dahlems, A.W.M.
Debaromyces hansenii var hansenii. FEBS Strasser and A.P.G.M. van Loon. 1999. An
Letters 487(1):82-86. expression system matures: a highly efficient
Liiv, L., P. Parn and T. Alamae. 2001. Cloning and cost-effective process for phytase
of the maltase gene from methylotrophic yeast, production by recombinant strains of
Hansenula polymorpha. Gene 265(1/2):77-85. Hansenula polymorpha. Biotech. Bioeng.
Lim, S.H., J.S. Choi and Y.E. Park. 2001. 63(3):373-381.
Microbial production of riboflavin using Minning, S., A. Serrano, P. Ferrer, C. Sola, R.D.
riboflavin overproducers, Ashbya gossypii, Schmid and F. Valero. 2001. Optimization of
Bacillus subtilis and Candida famata. An the high-level production of Rhizopus oryzae
overview. Biotechnol. Bioprocess. Eng. lipase in Pichia pastoris. J. Biochem. 86(1):59-
6(2):75-88. 70.
Lima, G., F. De Curtis, R. Castoria and V. De Miura, Y., K. Kondo, T. Saito, H. Shimada, P.
Cicco. 1998. Activity of the yeasts Fraser and N. Misawa. 1998. Production of the
Cryptococcus laurentii and Rhodotorula carotenoids lycopene, ß-carotene and
glutinis against post-harvest rots in different astaxanthin in the food yeast Candida utilis.
fruits. Biocontrol Sci. Tech. 8(2):257-267. Appl. Envt. Microbiol. 64(4):1226-1229.
Limtong, S., S. Deejing and W. Santisoparsi. Miura, Y., K. Kondo, H. Shimada, T. Saito, K.
1998. Construction of high ethanol Nakamura and N, Misawa. 1998. Production
halotolerant hybrid by intergeneric protoplast of lycopene by the food yeast, Candida utilis
fusion of Saccharomyces cerevisiae and that does not naturally synthesize carotenoid.
Zygosaccharomyces rouxii. Kaseterat J. Nat. Biotech. Bioeng. 58(2/3):306-308.
Sci. 32(2):213-223. Miura, Y., M. Kettoku, M. Kato, K. Kobayashi
Linardi, V.R., J.C.T. Dias and C.A. Rosa. 1996. and K. Kondo. 1999. High level production
Utilization of acetonitrile and other aliphatic of thermostable α-amylase from Sulfolobus
nitriles by a Candida famata strain. FEMS solfataricus in high-cell density culture of the
Microbiol. Lett. 144(1):67-71. food yeast Candida utilis. J. Mol. Microbiol.
Llorente, B., A. Malpertuy, G. Blandin, F. Biotechnol. 1(1):129-134.
Artinguenave, P. Wincker and B. Dujon. 2000. Montesinos, J.L., E. Dalamu and C. Casas. 2003.
Genomic exploration of the Lipase production in continuous culture of
188 C.A. Abbas
Candida rugosa. J. Chem. Tech. Biotech. Parrondo, J., L.A. Garcia and M. Diaz. 2000.
78:753-761. Production of an alcoholic beverage by
Nadezda, K., M. Maria and C. Peter. 1997. fermentation of whey permeate with
Structure of glucomannan-protein complex Kluyveromyces fragilis. I. Primary metabolism.
from the yeast Cryptococcus laurentii. J. J. Inst. Brewing 106(6):367-375.
Carboh. Chem. 16(5-4):609-623. Pazouki, M., P.A. Felse, J. Sinha and T. Panda.
Nakamura, T., T. Suzuki, J. Junko, N. Kato, Y 2000. Comparative studies on citric acid
Sakai, D. Mochizuki and H. Takahashi. 2000. production by Aspergillus niger and Candida
Production of recombinant phytase by fed- lipolytica using molasses and glucose.
batch fermentation of Candida boidinii. US Bioprocess Eng. 22(4):353-361.
Patent 6,140,077. Penttila, M. 1997. In: Biotechnology in the food
Nakano, K., R. Katsu, K. Tada and M. chain. VTT Symposium 177. pp. 109-118.
Matsumura. 2000. Production of highly Phaff, H.J. 1985. Biology of yeasts other than
concentrated xylitol by Candida magnoliae Saccharomyces. Biotech. Series 6:537-562.
under a microaerobic condition maintained by Picataggio, S., T. Rorer, K. Deanda, D. Lawning,
a simple fuzzy control. J. Bios. Bioeng. R. Reynolds, J. Mielenz and L.D. Dudley.
89(4):372-376. 1992. Metabolic engineering of Candida
Nielsen, J. and L. Olsson. 2002. An expanded tropicalis for the production of long-chain fatty
role for microbial physiology in metabolic acids. Biotech. 10:894-898.
engineering and functional genomics: moving Possoth, V., M. Zimmermann and U. Klinner.
towards systems biology. FEMS Yeast 1996. Peculiarities of the regulation of
Research 2:175-181. fermentation and respiration in the Crabtree-
Nishikawa, A., T. Sugita and T. Shinoda. 1997. negative, xylose-fermenting yeast Pichia
Differentiation between Debaromyces stipitis. Appl. Biochem. Biotechnol. 57/58:201-
hansenii/Candida famata complex and 212.
Candida guilliermondii by polymerase chain Pribylova, L. and H. Sychrova. 2003. Osmotic
reaction. FEMS Immunol. Med. Microbiol. stress resistant yeast Zygosaccharomyces
19(2):125-129. rouxii-isolation of auxotrophic mutants and
Nishikawa, A., T. Sugita and T. Shinoda. 1999. efficient transformation by electroporation.
Rapid identification of Debaromyces hansenii/ Abstract 11-7. XXIst International Conference
Candida famata by polymerase chain reaction. on Yeast Genetics and Molecular Biology,
Med. Mycol. 37(2):101-104. Gothenburg, Sweden.Yeast 20(S1).
Oh, D-K. and S-Y. Kim. 1998. Increased xylitol Ramirez, J., M.L.Nunez and R. Valdivia. 2000.
yield by feeding xylose and glucose in Increased astaxanthin production by a Phaffia
Candida tropicalis. Appl. Microbiol. Biotech. rhodozyma mutant grown on date juice from
50(4):419-425. Yucca fillifera. J. Indust. Microb. Biotech.
Panchall, C.J., I. Russell, A.M. Sills and G.G. 24(3):187-190.
Stewart. 1984. Genetic manipulation of Ramos, J. V. Montiel, C. Prista, R. Garcia-
brewing and related yeast strains. Food Tech., Salcedo and M.C. Loureiro-Dias. 2003.
February, pp. 99-111. Abstract 11-46. Yeast 20(S1), XXI
Parajo, J.C., V. Santos and M. Vasquez. 1998a. International Conference on Yeast Genetics
Optimization of carotenoid production by and Molecular Biology, Gothenburg, Sweden.
Phaffia rhodozyma cells grown on xylose. Rau, U., R. Heckmann, V. Wray and S. Lang.
Process Biochem. 33(2):181-187. 1999. Enzymatic conversion of a sophorolipid
Parajo, J.C., V. Santos and M. Vazquez. 1998b. into a glucose lipid. Biotech. Letters 21:973-
Production of carotenoids of Phaffia 977.
rhodozyma growing on media made from Rau, U., S. Hammen, R. Heckmann, V. Wray
hemicellulosic hydrolyzates of Eucalyptus and S. Lang. 2001. Sophorolipids: a source
globules wood. Biotechnol. Bioeng. 59 for novel compounds. Industrial Crops and
(4):501-506. Products 13:85-92.
Biotechnological applications of nonconventional yeast 189
Rodriguez, E., K. Sanchez, H. Roca, B. Garcia, Schwan, R.F., R.M. Cooper and A.E. Wheals.
J. Cremata and J. Delgado. 1996. Procedure 1997. Endopolygalacturonase secretion by
for the high-level recombinant dextranase Kluyveromyces marxianus and other cocoa
production. Biotecnol. Apl. 13(3):201. pulp-degrading yeasts. Enzyme and Microbial
Rodrigues, D.C.G.A., S.S. Silva, A.M.R. Prata Tech. 21:234-244.
and M.G.A. Felipe. 1998. Biotechnological Sharyshev, A.A., E.V. Peskova and G.N.
production of xylitol from agroindustrial Komarova. 1997.Effect of cultivation
residues. Evaluation of bioprocesses. Appl. conditions on the level of enzymes of n-alkane
Biochem. Biotech. 70-72:869-875. metabolism in Candida maltosa cells.
Roostita, R. and G.H. Fleet. 1999. Growth of Microbiology (Moscow) 66(6):652-656.
yeasts isolated from cheese on organic acids Shimada, H., K. Kondo, P. Fraser, Y. Miura, T.
in the presence of sodium chloride. Food Saito and N. Misawa. 1998. Increased
Technol. Biotechnol. 37(2):73-79. carotenoid production by the food yeast
Saha, B.C. and R.J. Bothast. 1996. Production Candida utilis through metabolic engineering
of L-arabitol from L-arabinose by Candida of the isoprenoid pathway. Appl. Envt.
entomaea and Pichia guilliermondii. Appl. Microbiol. 64(7):2676-2680.
Microbiol. Biotechnol 45(3):299-306. Sibirny, A.A. 1996a. Pichia guilliermondii. In:
Sakai, Y, M. Akiyama, H. Kondoh, Y. Shibano Nonconventional Yeasts in Biotechnology (K.
and N. Kato. 1996. High-level secretion of Wolf, ed). Springer-Verlag, Berlin, Heidelberg,
fungal glucoamylase using the Candida New York, pp. 255-275.
boidinii gene expression system. Biochimica Sibirny, A. 1996. Pichia methanolica (Pichia
et Biophysica Acta 1308(1):81-87. pinus MH4). In: Nonconventional yeasts in
Sakai, Y., Y. Tani and N. Kato. 1998. biotechnology (K. Wolf, ed.) Springer-Verlag,
Biotechnological application of cellular Berlin, Germany, pp. 277-291.
functions of the methylotrophic yeast. J. Silva, S.S., I.C. Roberto, M.G.A. Felipe and I.M.
Molecular Catal. B Enzymatic 6(3):161-173. Mancilha. 1996. Batch fermentation of xylose
Sanchez, S., V. Bravo, E. Castro, A.J. Moya and for xylitol production in a stirred tank
F. Camacho. 1998. The production of xylitol bioreactor. Process Biochem. 31(6):549-553.
from D-xylose by fermentation with Singh, D., P. Nigam, I.M. Banat, R. Marchant
Hansenula polymorpha. Appl. Microb. and and A.P. McHale. 1998. Review: ethanol
Biotech. 50(5):608-611. production at elevated temperatures and
Sanchez, S., V.Bravo, E. Castro, A.J. Moya and alcohol concentrations: Part II-use of
F. Camacho. 1999. Comparative study of the Kluyveromyces marxianus IMB3.World J.
fermentation of D-glucose/D-xylose mixtures Microb. Biotech. 14(6):823-834.
with Pachysolen tannophilus and Candida Skory, C.D., S.N. Freer and R.J. Bothast. 1996.
shehatae. Bioprocess Eng. 21(6):525-532. Properties of an intracellular beta-glucosidase
Sasahara, H., M. Mine and K. Izumori. 1998. purified from the cellobiose-fermenting yeast
Production of D-talitol from D-psicose by Candida wickerhamii. Appl. Microbiol.
Candida famata R28. J. Ferment. Bioeng. Biotechnol. 46(4):353-359.
85(1):84-88. Sreekrishna, K. and K.E. Kropp. 1996. Pichia
Santopietro, L.M.D., J.F.T. Spencer and F. pastoris. In: Nonconventional Yeasts in
Sineriz. 1998. Fed-batch and continuous Biotechnology (K.Wolf, ed). Springer-Verlag,
culture of Phaffia rhodozyma (Xantho- Berlin, Germany, pp. 203-253.
phyllomyces dendrorhous). Folia Microbiol. Sreenath, H.K. and T.W. Jeffries. 1999.
43(2):169-172. Production of ethanol from wood hydrolyzate
Sberna, G., R. Cappai, A. Henry and D.H. Small. by yeasts. Bioresour. Technol. 72(3):253-260.
1996. Advantages of the methylotrophic yeast Stahmann, K-P, J.L. Revuelta and H. Seulberger.
Pichia pastoris for high-level expression and 2000. Three biotechnical processes using
purification of heterologous proteins. Ashbya gossypii, Candida famata, or Bacillus
Australas. Biotechnol. 6(2):82-87. subtilis compete with chemical riboflavin
190 C.A. Abbas
Yamazaki,T., A. Kubiki, H. Ota, T.M. Mitzel, Yurimoto, H., T. Komeda, C.R. Lim, T.
L.A. Paquette and T. Sugai. 2000. Nakagawa, K. Kondo, N. Kato and Y. Sakai.
Yamadazyma farinosa IFO 10896-mediated 2000. Regulation and evaluation of five
reduction of 4,4-dimethoxy-2-butanone as the methanol-inducible promoters in the
key step for the preparation of 1,3- diols with methylotrophic yeast Candida boidinii.
unsymmetrical substituents. Synth. Commun. Biochimica et Biophysica Acta 1493(1-2):56-
30(16):3061-3072. 63.
Yanai, T. and M. Sato. 2000. Purification and
characterization of an α-L-rhamnosidase from
Pichia angusta X349. Bioscience, Biotech.
Biochem. 64(10):2179-2185.
Beverage alcohol production
Production of Scotch and Irish whiskies: their history and evolution 193
Chapter 14
T. PEARSE LYONS
President, Alltech Inc., Nicholasville, Kentucky, USA
Introduction
Whisky is the potable spirit obtained by Within the blend, there may be as many as 20-
distillation of an aqueous extract of an infusion 30 individual malt whiskies and grain whiskies.
of malted barley and other cereals that has been These blends are, by law, matured for at least
fermented with strains of Saccharomyces three years but in practice this period is much
cerevisiae yeast. Various types of whisky are longer. Unblended Scotch malt whiskies are
produced in a number of different countries in usually matured for a minimum of eight years.
the world. They differ principally in the nature The cereals used in the manufacture of Scotch
and proportion of the cereals used as raw grain whisky are malted barley, together with a
materials along with malted barley, and also in high proportion (up to 90%) of wheat or corn
the type of still used for distillation. (maize). Currently wheat is the main cereal,
The principal types of whisky are also chosen on the basis of cost and the attraction of
characteristic of particular geographical regions using a Scottish-grown cereal. All whiskies are
of the world. In Scotland, the characteristic legally protected and defined, mainly because
product is manufactured using only malted of the huge revenues that governments obtain
barley as the raw material; and the fermented from their sale. The Scotch Whisky Order (1990)
malt mash is distilled in relatively small pot stills. and the Scotch Whisky Act (1988), in defining
The product, known as Scotch malt whisky, is Scotch Whisky, state that to be called Scotch
produced in small distilleries, of which there are Whisky spirits must be:
over 100 in Scotland. Scotch malt whisky is
marketed both as a straight malt whisky, many 1. produced at a distillery in Scotland
brands of which have recently become
extremely popular throughout the world, and 2. produced from water and malted barley to
also as a blend with another type of whisky which only whole grains or other cereals may
produced in Scotland known as ‘Scotch grain be added
whisky’ or (because it is distilled continuously 3. processed at that distillery into a mash
in Coffey-type patent stills) as ‘patent-still
4. converted to fermentable carbohydrate only
whisky’. Most Scotch whiskies available on the
by endogenous enzymes
international market consist of blends with 20-
30% malt whisky and 70-80% grain whisky. 5. fermented only by the addition of yeast
194 T.P. Lyons
English in 1170. In all likelihood the art of riots occurred. The English, meanwhile, had
distillation was imported into Scotland by cultivated a taste for French brandy, there being
missionary monks from Ireland. Two of today’s very little whisky consumed at that time outside
main centers of Scotch distilling, namely the of Scotland. However around 1690 William III
island of Islay and the Speyside town of began to wage commercial war against the
Dufftown, were the sites of early monastic French, and imposed punitive taxes on imports
communities. of French brandy into England. The English
Whisky, principally Scotch whisky, has for reacted by acquiring a taste for gin, which was
many years been one of the most popular distilled locally. The scale of drunkenness that
distilled beverages in the world; and it was in developed with the popularity of gin had to be
Scotland rather than in Ireland that its qualities controlled by law; and the Acts of 1736 and 1713
came to be extensively appreciated. This has levied high taxes on gin manufacturers. Both of
continued to the present day and in the these acts contained clauses exempting Scotland,
intervening period many Scotsmen have felt but not for long. The Parliament in London saw
compelled to record for posterity their thoughts the prospect of a rich harvest of taxes in the
and inspirations on the potable spirit. There are distilleries of Scotland; and in a series of acts
numerous histories of whisky distilling in starting in 1751, production of whisky in
Scotland, some more comprehensive than others. Scotland was increasingly subjected to taxation.
For good general accounts, the reader is referred The outcome of these punitive measures was
to the texts by Brander (1975), Daiches (1969), not surprising. An extensive and thriving
Ross (1970) and Ross Wilson (1970). business in illicit distillation of whisky grew up
Whisky distilling flourished in Scotland not in Scotland as described by Sillett (1965).
least because consuming the spirit helped the Curiously, illicit production of Scotch hardly
inhabitants to withstand the climatic rigors of extended over the border into England, although
this northern region of Britain. The first recorded there are a few records of the operation of illicit
evidence of whisky production in Scotland is stills in the Cheviot Hills west of Newcastle-
an entry in the Exchequeur Rolls for the year upon-Tyne. Following the Act of 1823, which
1494. It reads “To Friar John Cor, by order of introduced much stiffer penalties for illicit
the King, to make aquavitae, eight bolls of malt”. distillation, and to some extent because of the
Production of whisky was therefore being increased standards of living in northern
controlled; and an Act of 1597 decreed that only Scotland, illicit manufacture of whisky declined.
earls, lords, barons and gentlemen could distill Indeed, many erstwhile illicit distillers emerged
for their own use. To many Scots of this era to become legal and registered distillers of
whisky was a medicine, and in 1506 King James Scotch whisky.
IV of Scotland had granted a monopoly for In 1826, Robert Stein of the Kilbagie distillery
manufacture of ‘aqua vitae’ to the Guild of in Clackmannanshire, Scotland, patented a
Barber Surgeons in the City of Edinburgh. continuously operating still for whisky
Taxation on whisky production first appeared production. However, this invention was
in the 17th century. Breaches of the monopoly superseded in 1830 with the introduction by
regulations and the need to raise money to send Aeneas Coffey of an improved version of this
an army into England to help the English type of still. The appearance of continuous stills
Parliament in its war against Charles I led to the sparked off a period of turmoil in the Scotch
Act of 1644, which fixed a duty of two shillings whisky industry, it being claimed that the product
and eight pence Scot’s on a pint of whisky (the from the continuous distillation of a mash that
Scot’s pint was then about 1.5 litres). But the contained unmalted grain (described as neutral
tax was short-lived and was replaced by a malt or ‘silent spirit’) could not be called whisky,
tax that later was also repealed. since it had not been distilled in the traditional
At the time of the Treaty of Union between pot still. The battle was waged for about three
Scotland and England in 1707 there was a tax quarters of a century; and in 1908 a Royal
on malt in England, but not in Scotland. The Commission decided that malt whisky and grain
English were irate; and in 1725 when Lord neutral spirit, when blended, could be labeled
Walpole’s administration decided to enforce the whisky.
tax in Scotland, the first of a series of Malt Tax
196 T.P. Lyons
The major factors which have affected the is entitled to be called a highland malt whisky,
development of the whisky distilling industry while whiskies distilled in areas south of the line
in Scotland in this century have been economic. are designated as lowland whiskies. Of the 104
The industry has had to endure the privations of malt whisky distilleries in Scotland, 95 are
two world wars, the economic depression in highland malt whisky distilleries. Of these, no
Great Britain during the 1920s and prohibition fewer than 49 are situated in an area measuring
in the United States from 1920 to 1933, which 50 miles east to west and 20 miles southwards
greatly affected export of Scotch to North from the Moray Firth. This area of Speyside has
America. Since 1945, however, the industry in been called the ‘Kingdom of Malt Whisky’
Scotland has consolidated and expanded. The (Cameron Taylor, 1970). Classification of the
20th century has also witnessed a considerable four whiskies distilled on the islands of Jura,
improvement in the quality of whisky distilled Orkney and Skye is disputed. Some authorities
and blended in Scotland as a result of the list them along with the Islay whiskies as ‘island’
acceptance of blending malt whiskies with grain whiskies, others as highland whiskies, which is
whisky and the amalgamation of several smaller geographically correct. There are also eight grain
distilleries into combines. whisky distilleries in Scotland.
Scotch malt whiskies can be divided into Whiskey distilling in Ireland was, as has been
‘highland’, ‘lowland’, ‘Islay’ and ‘Campbeltown noted, first recorded in the 12th century. By
whiskies’ (Simpson et al., 1974). The ‘highland 1556, it had become sufficiently widespread to
line’, which separates the areas in Scotland in warrant legislation to control it. A statute
which the first two types of spirit are distilled, is proclaimed that year stated that a license was
a straight line which runs from Dundee in the required to manufacture the spirit, but that peers,
east to Greenock in the west (Figure 1). It then gentlemen owning property worth £10 or more
extends southwards, below the Isle of Arran. and borough freemen were exempt (McGuire,
Any whisky produced north of this line, 1973). Taxation of whisky distilling gradually
including those from Campbeltown and Islay, became more excessive and collection of taxes
Northh Sea
ENGLA
LAND
AND
Table 1. Raw materials and unit processes used in the production of Scotch and Irish whiskies.
Raw materials Peated and unpeated Wheat or corn and a small Unmalted barley and unpeated,
malted barley proportion of malted barley malted barley
Conversion Infusion mash Mash cook followed by Infusion mash
conversion stand
Fermentation Distillers yeast and Distillers yeast Distillers yeast
brewers yeast
Distillation Two pot stills Continuous still Three pot stills
o o
Maturation At 62 GL in used charred Up to 67 GL in used charred At 70o GL in used charred oak
oak bourbon whiskey oak bourbon whisky barrels bourbon whisky barrels or sherry
barrels or sherry casks or sherry casks for at least casks for at least three years
for at least three years three years
198 T.P. Lyons
Fermentation is conducted with strains of the species Hordeum vulgare L. and Hordeum
yeast Saccharomyces cerevisiae that are usually distichon (Hough et al., 1971). Malted barley is
specially propagated for the purpose, although employed as a source of enzymes (principally
Scotch malt whisky distillers may use some amylolytic) that catalyze the hydrolysis of
surplus brewers yeast (Table 1). The process is starches and in some instances serves as a source
allowed to proceed to a point at which the of starch that is converted ultimately into ethanol.
specific gravity of the fermented mash has These two demands must be finely balanced. In
usually dropped to below 1.000. In pot still the manufacture of Scotch malt whisky, where
distilleries, the fermented mash or ‘wash’ (beer) only malted barley is used, care must be taken
is fed directly to a still known as the wash still, when the grain is sprouting during the malting
from which the distillates are redistilled in the process to ensure that only a limited amount of
second or low wines still. In Ireland, and in one enzyme activity is produced. This is because
Scotch malt whisky distillery, a third distillation enzyme is produced at the expense of the
is carried out. Finally, the freshly distilled whisky fermentable materials in the grain (referred to
is stored in charred oak barrels for minimum as ‘extract’). However in the manufacture of
periods of time that depend on the legislation in other types of whisky, malted barley is often
the producing country (Table 1). Scotch malt and used as the only source of amylolytic enzyme
Irish whiskies are customarily matured for much in a mash bill that contains a high proportion of
longer than the legal minimum period. unmalted grain. In this type of whisky production
the enzyme activity of the malted barley must
be greater than that used in Scotch malt whisky
Individual operations manufacture.
Traditionally, barley used in the production of
RAW MATERIALS
Scotch malt whisky was malted on the distillery
Malted cereals premises using a floor malting system and dried
over fires of coke and peat in the pagoda-shaped
Malted barley is the principal malted cereal used kilns which are still a feature of these distilleries.
in whisky production. Like the brewer, the To a large extent, this system has been
whisky distiller uses barley cultivars of the superseded by mechanical maltings, which
Yeast
Malt
storage Hot
bins water Condensers
Washback
(fermenter)
Spirit or middle run
Foreshot and feints
(Head & tails)
Malt
mill
Mashing
machine
(mixer)
Maturing warehouse
Wash still Low wines
Heat still
exchanger (pot still)
Low
wines Oak casks
(barrels)
Figure 2. Flow diagram showing the principal operations during production of Scotch malt whisky.
Production of Scotch and Irish whiskies: their history and evolution 199
produce malt for groups of distilleries. In order have a nitrogen content of 1.8% or higher
not to destroy the enzyme activity developed (compared with a brewer’s malted barley with a
during malting, a balance must be achieved in nitrogen content in the region of 1.5%) (Hough
the kiln between drying the green malt to a et al., 1971). Malting aids, such as gibberellic
suitably low moisture level for storage, curing acid and bromate, are not normally used in
to give it the appropriate flavor and retaining production of malted barley for whisky
sufficient enzyme activity (Simpson, 1968). In manufacture.
maltings attached to the distillery, the kiln Because of the high cost of malted cereals,
temperature is increased slowly over a 48 hr considerable effort has been expended by the
period to achieve an even rate of drying and the whisky distiller in attempting to devise methods
desired flavor. The latter character is achieved that allow prediction of the yield of alcohol
by fuelling the furnace with peat during the early expected using different proportions of malted
part of the kilning period when the green malt is barley in the mash bill. Unfortunately, the
moist and readily absorbs the peat smoke or methods customarily used by the brewer, such
‘reek’. In mechanical maltings the green malt is as those recommended in Great Britain by the
dried at a faster rate with a forced-air draught, Institute of Brewing Analysis Committee (1975),
but a supplementary peat-fired kiln is often used have proved of limited value. The brewer has
to produce flavored malts. The amount of peat used measurements of diastatic power,
used varies with different maltings. Some of the expressed as the Lintner value, which is a
distilleries on Islay in Scotland specialize in measure of the extent of saccharification of
producing a whisky with a very pronounced peat soluble starch present in a cold water extract of
flavor, and they therefore use heavily-peated the malt (Lloyd Hind, 1948) as an indicator of
malts. Malted barleys used in the manufacture malt quality. Diastatic activity measured in this
of Scotch grain and Irish whiskies are not dried way includes contributions from both α- and ß-
over a peat fire. They generally have a greater amylases. However, Preece and his colleague
enzyme activity, so the relatively small (Preece, 1947; 1948; Preece and Shadak-
proportion of malted barley used in the mash sharaswamy, 1949 a,b,c) have shown that high
contains sufficient enzyme activity to convert ß-amylase activity, as determined by the Lintner
all of the starch in the mash (principally supplied value, is not always accompanied by high α-
by unmalted cereals) into fermentable sugars. amylase activity. Pyke (1965) showed that the
The greater enzyme activity in these malted Lintner value of a malt is only useful for
barleys is reflected in their nitrogen content. predicting the spirit yield in manufacture of
Malts used in production of Scotch grain whisky Scotch grain whisky when the proportion of
375
Spirit yield (Liters absolute
per tonne of grain)
350
325
300
0 10 20 30
Malt enzyme activity (°Lintner)
Figure 3. Relationship between the diastatic activity of a Scotch grain whisky mash bill and spirit yield. Laboratory fermentations
were conducted using mash bills containing different proportions of malt mixed with corn, and therefore with different Lintner
values. It can be seen that only when the proportion of malt is rate-limiting can spirit yield be correlated with the Lintner value
(Pyke, 1965).
200 T.P. Lyons
malted barley in the mashbill is low (Figure 3). chromatography (GLC). The peat smoke oil was
Determination of α-amylase activities of the malt more complex, and no fewer than 30 peaks,
gave a less satisfactory correlation than the some of them created by more than one
Lintner value, an observation which agrees with compound, were obtained by GLC. The
that made earlier by Thorne et al. (1945). Further compounds included hydrocarbons, furfural
evidence for the unsuitability of employing derivatives, benzene derivatives and phenols.
traditional malt specifications for predicting The authors stressed that using their GLC
performance in whisky manufacture has come techniques, 3,5-xylenol is masked by the peaks
from Griffin (1972). of m-ethylphenol and ρ-ethylphenol, two
Although the degree to which a malted barley compounds thought to make an important
has been peated can to some extent be assessed contribution to peat aroma and taste. Figure 4
by smell, such is the importance of this character shows gas liquid chromatograms of extracts of
in malt that it must be determined in a more a peated and a unpeated malt.
rigorous fashion. Peat smoke or ‘reek’ contains
a wide range of compounds, but it is generally Unmalted cereals
held that the peaty character is imparted to the
malt largely as a result of absorption of phenols. Fewer problems are encountered in arriving at
For some years, Scotch distillers used a method specifications for the unmalted cereals used in
based on a reaction of phenols with diazotized whisky manufacture, namely wheat, corn, rye
sulphanilic acid. A lack of specificity in this and barley. The corn (varieties of Zea mays) used
method, coupled with the instability of the in mash bills for manufacturing Scotch grain
diazonium salt, prompted MacFarlane (1968) to whiskies is usually of the yellow dent type,
recommend an alternative method involving generally obtained from France. Occasionally
extraction of phenols from malt with diethylether white corn is used, and it is reputed to give a
under acid conditions with absorptiometric higher alcohol yield. Corn is a popular grain
measurement of the color developed when the because it has a high content of starch that is
phenols are reacted with 4-aminophenazone. readily gelatinized and converted into
MacFarlane applied the method to a wide variety fermentable sugars. The US has imposed controls
of malts, both peated and unpeated, and reported on the quality of corn used for whisky
values ranging from zero (for an unpeated grain) manufacture. In the US there are three grades of
to as high as 9.4 ppm for a malt produced on corn, with only grades 1 and 2 being approved
Islay in Scotland. (It has been calculated that to for spirit manufacture. In Great Britain, on the
obtain a malt with 10 ppm phenols, one tonne other hand, the corn used is normally grade 3
of peat must be used for drying each tonne of on the US scale.
malted barley). Unmalted barley used in manufacture of Irish
Scotch whisky distillers have been concerned whiskey has a quality intermediate to that used
by the possibility that colorimetric methods, such for malting and that used for cattle feed. In this
as those recommended by MacFarlane (1968) way, the maltster can select the best barley
and Kleber and Hurns (1972), may not assay all available on the market at the time of purchase.
of the organoleptically-important compounds For many years, a small percentage (about 5%
that malt acquires as a result of peating. To of the total) of unmalted oats (Avena spp.) was
examine this possibility, MacFarlane et al. included in mash bills for manufacturing Irish
(1973) produced peat smoke condensate on a whiskey. It was contended that these grains, with
laboratory scale and separated the oil and their large husks, improved the texture of the
aqueous phases from the wax fraction, as only grain bed in the mash tun (to assist in straining
the two former would contain components that off the clear wort from the mash solids) and that
might appear in the distilled whisky. Six oats influenced the flavor of Irish whiskey.
compounds, namely furfural, 5-methylfurfural, Whether either or both of these effects were
guiacol, phenol, ρ-cresol and 5-xylenol, were important will probably never be known, for oats
detected in the aqueous fraction by gas liquid are no longer used in the production of Irish
whiskey (Court and Bowers, 1970).
Production of Scotch and Irish whiskies: their history and evolution 201
Unpeated malt
Peated malt
ρ-Ethylphenol
ρ-Cresol
Figure 4. Gas-liquid chromatograms of extracts of unpeated and peated Scoth barley malts showing the contribution peating
makes to the content of phenols in the malt. Peating leads to an increase in the size of the peak corresponding to phenol, ρ-cresol
and guiacol, and of the peaks which lie to the right of the mixed phenol peak (attributable to furfurals and hydrocarbons). ρ-cresol
is also detectable on the chromatogram of extracts of peated malt. The ratio of the area of the total phenol peaks to that of ρ-
ethylphenol is used as an indication of the peatiness of the malt.
202 T.P. Lyons
MASHING AND COOKING The meal from overhead bins is mixed with hot
water at 60-65oC in the proportion of one part
Mash production in Scotch and Irish distilleries meal to four parts water, and the mash is
involves a process not unlike that used in homogenized by action of revolving rakes. In
breweries to prepare wort for beer manufacture. the traditional mashing process, the mash is
However, where cereals other than malted barley usually loaded into the tun to a depth of about
are used, the malt mashing is preceded by a high one meter and allowed to stand for about one
temperature cooking process. hour, after which the wort is drained off from
under the grain bed. This liquid extract, which
Mashing has a specific gravity (SG) of 1.070-1.060
Regardless of whether the mash bill contains (Figure 5) is collected in an intermediate vessel
cereals other than malted barley, the main known as an ‘underback’. After being cooled to
biochemical changes that take place during around 25oC in a heat exchanger, the wort is
mashing are hydrolysis of starch, protein and pumped into the fermentation vessel. The bed
other biopolymers in the meal to produce water of grains in the tun is then re-suspended in water
soluble low molecular weight compounds that at 75oC and a second batch of wort is drawn off
form a fermentable substrate (or wort). The major at a specific gravity of around 1.030 and passed
starch-liquefying and saccharifying enzymes are into the underback. This process is known as
α- and ß-amylases, while the limit dextrins the first aftermash and is repeated twice more;
formed by action of amylases on amylopectin except that the dilute worts drawn off are not
are further hydrolyzed by limit dextrinases. passed to the underback, but are returned to the
Barley malt used in the manufacture of Scotch hot water tank to be used in the next mash. The
malt whisky is coarsely ground in a roller mill wort in the underback has a pH value of about
adjusted to give a grind no finer than the malt 5.5, a specific gravity of 1.045-1.065, and an
warrants. Too fine a grind can give rise to a ‘set amino nitrogen content of about 150-180 mg/
mash’ that settles on the bottom of the mash tun L. The spent grain residue or ‘draff’ is removed
to block and impede drainage of the liquid from the mash tun and sold as animal feed. Many
(Simpson, 1968). The mash tun is usually distilleries have now reduced this process to just
preheated with water and the perforated bottom three water additions. Larger distilleries are now
plates are flooded before mashing commences. installing semi-lauter tuns in place of the
traditional mash tuns. These have vertical knives,
Malt meal
Figure 5. Flow diagram showing the mashing cycle in a Scotch malt whisky distillery.
Production of Scotch and Irish whiskies: their history and evolution 203
enabling the structure of the mash bed to be loss of sugar and decreased spirit yield. Pyke
maintained whilst speeding wort draining, and (1965) has provided an account of the cooking
sparging rings to allow simultaneous draining of corn in its production of Scotch grain whisky
and sparging. The result is a faster more efficient and a similar process is still used today. The
extraction without loss of clarity or changing the conventional cooker is a horizontal cylindrical
composition of the wort. vessel capable of working at pressures up to 90
Mash preparation in the manufacture of Irish psi (63 x 104 Pa), and fitted with stirring gear. A
whiskey is very similar to that used in Scotch typical cycle in cooking corn might consist of
production, but there are certain differences. Use 1-1.5 hrs of heating with live steam injection to
of a high percentage of raw barley in the mash reach a temperature of 120oC and a pressure of
bill (up to 60%) has necessitated the use of stone 15 x 104 Pa. The mash is held at this temperature
mills or hammer mills to achieve the required and pressure for a further 1.5 hrs. The liquid
grind. The unmalted barley is sprayed with water used for the mash is often that from the third
to give a moisture content of about 14% and aftermash mentioned earlier. At the end of the
then dried to around 4.5% moisture before cooking time, the pressure is released and the
grinding. The malted barley used is roller-milled. hot cooked corn mash is blown directly into a
In recent years a plant has been installed in saccharification vessel containing the malted
Ireland which uses a wet milling process that barley suspended in hot water. Cold water is then
eliminates the need for watering and drying of added to bring the temperature in the mash tun
the unmalted barley. Mash tuns used in Irish to around 63oC. Good mixing is also essential
whiskey distilleries differ from those used to at this stage; and failure to achieve it can lead to
make Scotch in that they are larger, simply the entire mash solidifying. The need for rapid
because these distilleries have stills with a greater cooling of the corn mash was emphasized by
capacity than those in Scotland. (This difference Murtagh (1970), who showed that with slow
dates back to when there was a flat rate of tax cooling or holding at a temperature around 82oC
per still in Ireland, rather than a tax per gallon lipid-carbohydrate complexes could be formed
of product.) Moreover, the mashing cycle differs that are not fermentable and can lead to a loss
in that the weak wort from the first aftermash is of 1-2% in spirit yield.
not passed to the underback, but is mixed with After a suitable mashing period, the mash may
worts from the second and third aftermashes to be filtered as described previously. More
be used in the subsequent mash (Lyons, 1974). commonly, the entire mash may be pumped via
a heat exchanger into a fermentor. Pyke (1965)
has published the composition of a typical
Cooking followed by malt mashing Scotch grain whisky mash (Table 2).
In the manufacture of Scotch grain whisky, the
The wheat, corn and rye used in production of proportion of malted barley in the mash bill is
Scotch grain whisky must be cooked before usually around 10-15%, a proportion far greater
being added to the mash containing malted than that required merely to provide a source of
barley in order that the starch in these grains amylolytic enzymes. It is a practice of long
can become gelatinized and accessible to the standing, and is done to obtain the required malt
malt enzymes. Traditionally, cooking was carried flavor in the grain whisky. This was noted as
out as a batch process, but it has to some extent early as 1902 by Schidrowitz and prompted
been superseded by continuous processes. further comment from Valaer in 1940.
For batch cooking, the grain is freed from Continuous cooking has been adopted in
extraneous material by passage through screens recent years in Scotland. Stark (1954) listed the
and then ground either in a pin-type mill or a advantages of using continuous cooking. While
hammer mill. It is then conveyed to the cooker practices vary to some extent in different
and subjected to a cooking cycle involving high distilleries, essentially the cereal is slurried at
temperatures and pressures designed to bring around 50oC and then pumped through a steam-
about complete gelatinization of the starch. jet heater into the cooking tubes where the
Inadequate cooking will sometimes leave starch residence time is normally 5-10 minutes. The
granules intact in the mash, while excessive continuous cooker has a narrow tube (6-16 cm
heating can cause caramelization and therefore
204 T.P. Lyons
in diameter) which reduces the incidence of fermentors were made of wood, usually of larch
charring and carbon deposition. The mash or Oregon pine, but in recent years they have
passes through the tube at about 20 meters/ been constructed of steel or aluminum. In some
minute, with the temperature reaching near 65oC Scottish distilleries, timber is still used as a
and the pressure 65 psi (40 x 104 Pa). covering material.
When the fermentor is partly filled, the mash
Table 2. Chemical composition of a typical scotch grain is inoculated with a suspension of
mash*. Saccharomyces cerevisiae. The source of yeast
varies with the location and size of the distillery.
% Distilleries in Scotland and Ireland, particularly
Total soluble carbohydrate (as glucose) 9.00
the pot whisky distilleries, rarely have their own
Insoluble solids 2.20 yeast propagation equipment. They rely on
Fructose 0.13 specially-propagated pressed yeast for use in
Glucose 0.29 fermentations. In Scotch malt whisky distilleries,
Sucrose 0.28 this pressed yeast is augmented, usually on an
Maltose 4.65 equal weight basis, with surplus brewery yeast
Maltotriose 0.96 typically supplied in compressed form from
Maltotetraose 0.45 breweries that may be as far as 500 km from the
Dextrin 2.54 distillery.
Amino nitrogen (as leucine) 0.09
Ash 0.27
The requirements for a strain of Saccharo-
containing P2O5 0.09 myces cerevisiae yeast used in distillery practice
K 2O 0.09 have been described by Harrison and Graham
Mg O 0.02 (1970). Apart from the obvious need to select a
strain that maintains very high viability in the
µg/ml
Thiamin 0.46
pressed state (containing 25% dry weight), other
Pyridoxine 0.61 very important properties of these strains are
Biotin 0.01 ability to tolerate concentrations of ethanol of
Inositol 236 the order of 12-15% (v/v) and the capacity to
Niacin 11.1 metabolize oligosaccharides such as maltotriose
Pantothenate 0.71 and maltotetraose in order to maximize the
conversion of starch into ethanol and carbon
*Pyke, 1965. dioxide.
Whisky worts usually have a specific gravity
in the range 1.050-1.080, a pH value of around
Fermentation 5.0, a total acid content of 0.1% and an optical
rotation of +30 o. After inoculation, the yeast
The objectives in fermenting a whisky distillery content is 5-20 million cells/ml. The bacterial
mash with strains of Saccharomyces cerevisiae count varies with the cleanliness of the plant and
are to convert the mash sugars to ethanol and the extent to which the raw materials were
carbon dioxide while producing quantitatively endowed with a microbial flora. Scotch grain
minor amounts of other organic compounds whisky fermentations create little if any foam
which contribute to the organoleptic qualities of because of their large content of suspended
the final distilled product. Fermenting vessels solids. However, in most Scotch malt whisky
vary considerably in volume, depending on the fermentations only a small proportion of the
distillery. The small Scotch whisky distillery at suspended solids in the mash is retained in the
Edradour near Pitlochry, for example, has fermentation vessel. These fermentations tend
fermentors with a capacity of only 4,500 liters. to foam and the distillers have resorted to the
In contrast, other Scotch and Irish pot whiskey use of various types of antifoams.
distillers have fermentors with a capacity in the Changes over the time course of a typical
range 50,000 to 150,000 liters. Much larger fermentation in a Scotch malt whisky distillery
fermentors are found in Scotch grain whisky are depicted in Figure 6. Fermentation proceeds
distilleries. Traditionally, the smaller pot distillery vigorously for the first 30 hrs, during which time
Production of Scotch and Irish whiskies: their history and evolution 205
1.060
5.0 +30 0.30
1.050 pH
pH
1.030
3.5 +15 0.15
Acidity
1.020
3.0 +10 0.10
1.010
2.5 Specific Optical rotation +5 0.05
0 gravity
2.0 0 0
0.995
0 10 20 30 40 50
Figure 6. Changes in specific gravity, optical rotation, pH value and acidity during fermentation of a mash
in the production of Scotch malt whisky (Dolan, 1976).
the specific gravity falls to 1.000 or below and (Pyke, 1965). Over the first 30 hrs the pH value,
the optical rotation to around zero. The sugars after declining to around 4.2, rises to about 4.5.
in the wort are utilized in a particular sequence During the first 30 hrs the specific gravity drops
with glucose and fructose being fermented first, at a rate of about 0.5o per hour accompanied by
followed by maltose and then maltotriose. The a massive evolution of heat. While many of the
removal of sugars during fermentation of a larger grain whisky distilleries have fermentors
Scotch grain whisky mash is shown in Figure 7 fitted with cooling coils, these are absent, or if
50
Concentration of fermentable sugar (g/l)
Maltose
30
Maltotriose
10
Glucose
0
10 30 50
Fermentation time (hrs)
Figure 7. Time course removal of fermentable sugars from a Scotch grain whisky mash (Pyke, 1965).
206 T.P. Lyons
fitted are relatively inefficient in most malt During the first 30 hrs or so of malt whisky
whisky distilleries where temperature can rise fermentation there is vigorous fermentation and
by the end of the fermentation to as high as 35- the majority of the aerobic bacteria die. This,
37 o C. The distiller is concerned about the however, provides ideal conditions for growth
temperature rise during fermentation since this of anaerobic or microaerophilic bacteria,
can cause the fermentation to stop or become principally lactic acid bacteria (mainly strains
‘stuck’. Temperature rise can be controlled by of Lactobacillus brevis, L. fermenti and
using a lower starting temperature or, because Streptococcus lactis) with the result that the
glycolysis of sugar is a heat-producing process, concentration of lactic acid in the fermented mash
by using a lower initial concentration. Strains of can be as high as 30 mg/L (MacKenzie and
Saccharomyces cerevisiae are well suited for Kenny, 1965). A wide range of lactobacillus
malt whisky distillery fermentations since they species have been identified in Scotch whisky
can ferment efficiently over a wide temperature fermentations including L. fermentum, L. brevis,
range. Fermentation is usually continued for at L. delbrueckii, L. plantarum, L. casei and a
least 36 hrs and frequently longer, at which time bacterium resembling L. collinoides, in addition
the ethanol content of the wash is 7-11% (v/v). to Leuconostoc spp., Streptococcus lactis and
In larger distilleries, particularly those in the US, Pediococcus cerevisiae (Bryan-Jones, 1976).
the carbon dioxide evolved is collected, liquefied More recently Barbour (1983) isolated many
and sold. Smaller distilleries, particularly the malt species that did not conform to recognized
whisky distilleries in Scotland, usually do not species of lactic acid bacteria, a point
have this facility. emphasized by Walker et al. (1990) who used
It should be noted that mashes in malt whisky DNA hybridization techniques to classify
distilleries are not boiled, so any enzyme activity distillery bacteria. Growth of lactic acid bacteria
manifested at the temperature of the mash and is probably enhanced by yeast excretion of
any microorganisms that can survive at that nitrogenous nutrients at the end of a vigorous
temperature will continue to be active during the fermentation. Kulka (1953) demonstrated the
fermentation. The continued activity of limit ideal nature of yeast autolysate for growth of
dextrinases in unboiled distillery mashes lactobacilli. Bacterial activity in the fermenting
increases the concentration of sugars available wort also leads to removal of some acids.
for fermentation by the yeast. Hopkins and Actively growing yeasts secrete citric and malic
Wiener (1955) calculated that with amylases acids, but MacKenzie and Kenny (1965)
alone the yeast cannot metabolize the equivalent attribute the lower concentrations of these acids
of the final 12-16% of the starch. in malt distillery worts (as compared to brewery
Another important consequence of using non- worts) to their partial removal by bacteria.
sterile conditions in distillery fermentations is Occasionally, the extent of the bacterial flora
the activity of bacteria that pass through in the in the fermenting wort can become too large.
mash, which are encouraged to some extent by This causes problems due to sugar utilization
the relatively high temperatures to which the by the bacteria that lead to an overall decrease
fermentations can rise. In addition to lactic acid in spirit yield. In addition, the bacteria may
bacteria, the flora can include other Gram- produce organoleptically-undesirable
positive as well as Gram-negative strains. The compounds and also release hydrogen ions
concentration of the flora depends on a number causing the pH value of the wort to fall too low,
of factors including the extent to which the lactic thereby providing suboptimal conditions for
acid bacteria grew during yeast propagation, the action of certain enzymes. Examples of
extent of the flora on the cereal raw materials undesirable compounds that may be excreted
and on the standard of hygiene in the distillery. by bacteria are hydrogen sulfide and other
There is no doubt, however, that the controlled sulfur-containing compounds (Anderson et al.,
activity of this bacterial flora, and particularly 1972). Lactobacilli can also metabolize glycerol
of the lactic acid bacteria, is accompanied by (excreted by the yeast during fermentation) to
excretion of compounds that contribute to the produce ß-hydroxypropionaldehyde, which
organoleptic qualities of the final whisky subsequently breaks down on distillation to give
(Geddes and Riffkin, 1989). acrolein (Harrison and Graham, 1970). Acrolein
Production of Scotch and Irish whiskies: their history and evolution 207
imparts a pungent, burnt and often peppery odor hrs and a lower optical rotation of the mash after
to the whisky (Lyons, 1974). In a later paper, about 40 hrs. In the fermentations there is often
Dolan (1976) concentrated on the problems a difference of up to 4 hrs from the time a rise in
arising in malt whisky distilleries when there is the acid content is detected to the point when
an unacceptably high concentration of bacteria the pH value of the fermentation begins to fall.
in the mash. Table 3 shows changes in the Dolan (1976) attributes this to the buffering
concentrations of Gram-negative and Gram- capacity of the mash. The data in Table 4 show
positive bacteria and (separately) of lactobacilli the effect of different levels of infection after 30
during fermentation of a minimally-infected hrs fermentation of a malt distillery mash on
mash and of a heavily-infected mash. The time spirit yield and the associated financial losses to
course of fermentation of an unacceptably- the distiller. Dolan (1976) recommends upper
infected malt distillery mash (Figure 8) shows, limits of 1,500 bacteria, 50 Gram-positive and
in comparison with similar data for fermentation 10 lactic acid-producing bacteria per million
of an acceptable mash (Figure 6), a greater rise yeast cells in the mash at the start of fermentation.
in the acid content of the mash after about 35
Table 3. Changes in the concentration of bacteria during fermentation of a minimally infected and a heavily infected
Scotch malt whisky wort.*
*Dolan, 1976.
**None detected.
1.060 0.30
5.0 +30
Optical rotation (a °D)
1.050 0.25
Specific gravity
1.030 0.15
3.5 Acidity +15
1.020 0.10
3.0 +10
1.010 Specific Optical rotation 0.05
2.5 +5
gravity
0 2.0 0
0
0.995 -2.5
0 10 20 30 40 50
Figure 8. Changes in specific gravity, optical rotation, pH value and acidity during fermentation of a Scotch malt whisky mash
containing an unacceptably high concentration of bacterial infection (Dolan, 1976).
208 T.P. Lyons
Table 4. Effect of the level of bacterial infection on the loss of spirit incurred following fermentation of a Scotch malt
whisky mash.*
*Dolan, 1976.
Much less has been published on the effect of major higher alcohols was increased in the
retaining solid material in the fermenting mash. presence of solids, the effect being particularly
However, marine microbiologists have long noticeable with isobutanol and 2-methylbutanol.
known that the presence of solid particles in a The effect of low insoluble solids content is a
liquid medium can affect bacterial growth, factor relevant to congener levels in malt whisky
probably because of the concentration of fermentations. In grain whisky production,
nutrients at the solid-liquid interface (Heukele- where ‘all-grains-in’ fermentations are generally
kian and Heller, 1940; Zobell, 1943). Moreover, used, the degree of rectification during
Cromwell and Guymon (1963) found that form- distillation is the principle determinant of higher
ation of higher alcohols during fermentation of alcohol levels in the spirit.
grape juice is stimulated by the presence of grape
skins or inert solids. Beech (1972) made similar
observations on cider fermentations. Merritt DISTILLATION
(1967), in the only detailed report on the role of
solids in whisky distillery fermentations, states Whether the fermented beer is distilled in a pot
that a dry solid concentration of 50 mg/100 ml still, as in production of Scotch malt and Irish
might typically be expected, although much will whiskies, or in a continuous still based on the
clearly depend on the design of the mash tuns Coffey design as in the manufacture of Scotch
used in individual distilleries. Merritt went on to grain whiskies, the objectives are the same:
report that a concentration of dry solids as low selective removal of the volatile compounds
as 5 mg/100 ml causes an increase in yeast (particularly the flavor-producing congeners)
growth, and that solids also enhance the rate of from the non-volatile compounds and to create
production of ethanol and glycerol. There was additional flavor-producing compounds as a
also an effect on production of higher alcohols result of chemical reactions that take place in
by the yeast (Table 5). With the possible the still. Nevertheless, it is still most convenient
exception of n-propanol, production of all of the to discuss whisky distillation under the separate
headings of pot still and continuous distillation.
Insoluble solids
content 2-methyl butanol 3-Methyl butanol
Mash (mg/100 ml) n-Propanol Isobutanol (amyl alcohol) (isoamyl alcohol) Total
Pot still distillation condenser. In Irish pot stills this purifier function
is effected by a trough fitted around the lyne
The copper pot still, which is the feature arm through which running water is circulated.
dominating any Scotch malt or Irish whiskey Pot stills in Scotch and Irish distilleries are
distillery, has changed hardly at all over the traditionally constructed of copper. The reason
centuries, except of course in size. Traditionally, for this adherence to copper is more than
the onion-shaped stills were fired from beneath tradition. It has been established that copper
and had the vapor pipe or ‘lyne arm’ from the fixes some volatile sulfur-containing
still projecting through the distillery wall to compounds that are produced during
connect with a condenser in the form of a coil fermentation but undesirable in the distilled
immersed in a water tank fed from a local stream spirit.
(Figure 9). Internal steam-heated calandria are Early whisky distillers realized that although
now preferred to direct firing because this the objective of distilling was to separate volatile
decreases the extent of pyrolysis of the still constituents from the beer, collecting the
contents and results, for example, in a lower distillate not as a whole but in several fractions
concentration of furfural in the whisky. Variations and combining certain of these fractions gave a
in still design include expansion of the surface much more acceptable product. Pot still
area of the column into a bulbous shape, water distillation in Scotland and Ireland differ not only
jacketing and return loops from the first stage in the size of the stills (25,000-50,000 liters in
of the condensation. (Nettleton (1913) provided Scotland vs 100,000-150,000 liters in Ireland),
a valuable account of early still design). In many but also in the different ways in which fractions
distilleries, condenser coils have been replaced are collected from the stills.
by tubular condensers that have an advantage In Scotland, the beer is subjected to two
in that they are designed to conserve the heat distillations. In the first, carried out in the beer
extracted from the distillates. Yet other pot stills still, the beer is brought to a boil over a period
are fitted with ‘purifiers,’ which consist of a of 5-6 hrs and the distillate is referred to as low
circular vessel cooled by running water wines. This distillation effects a three-fold
interposed between the neck of the still and the
Lyne arm
Condensate
Rousing return pipe
gear Worm
rub
Heating
coils
Product
Pot still
Figure 9. Diagram of an Irish distillery pot still. Designs used in Scotch malt whisky distilleries are similar except that the pot is
onion-shaped and the still usually has a shorter lyne arm not surrounded by a lyne-arm tank (Lyons, 1974).
210 T.P. Lyons
concentration of the alcohol in the beer (Figure whisky distilled over in the middle fraction has
10). The residue in the wash still, known as ‘pot an alcohol content of 63-70o GL.
ale’, is either discharged to waste or evaporated The manufacture of Irish whiskey involves
to produce an animal feed (Rae, 1967). three rather than two distillations (Figure 11).
Distillation of the low wines in the spirit is still The fermented beer is heated in the wash still,
more selective. The first fraction, which contains and the first distillation (‘strong low wines’) is
low boiling point compounds, is rejected as ‘fore- collected until the distillate reaches a
shot heads’. At a stage determined by continued predetermined specific gravity. The distillate,
hydrometric monitoring, which usually occurs then known as ‘weak low wines’, is switched
when the distillate has an alcohol content of into a separate vessel. The weak low wines are
approximately 70-73 o GL, the distillate is pumped into the low wines still and are re-
switched from the fore shots tank to the whisky distilled to produce two fractions termed ‘strong
receiver tank. This switch has traditionally been feints’ and ‘weak feints’. Strong feints are mixed
made at the discretion of the distiller; and he with the strong low wines in a spirit still.
has been aided by the disappearance of a bluish Distillates from this are collected in the same
tinge when water is added to the distillate. The fashion as in production of Scotch. The whiskey
monitoring process takes place in a spirit safe collected usually is about 89-92o GL.
secured with a Government Excise lock.
Collection of the distillate is terminated when Continuous distillation
the alcohol content has fallen to a specified value,
although distillation of the feints or tails is No fundamental changes have been introduced
continued until all of the alcohol has been into the design of the patent or Coffey still over
removed from the low wines. The residue the past century. Automation, particularly of the
remaining in the spirit still is known as ‘spent beer feed, is now commonplace, as is continuous
lees’, and like pot ale is either run to waste or monitoring of other stages in the distillation
evaporated to manufacture animal feed. The process. Nevertheless, many Scotch and Irish
Fermented beer
(about 8°GL)
Recycle
Figure 10. Flow diagram showing the stages in distillation of Scotch malt whisky.
Production of Scotch and Irish whiskies: their history and evolution 211
Fermented beer
(about 8°GL)
Wash Still
Figure 11. Flow diagram showing the stages in distillation of Irish whiskey.
producers of grain whiskies continue to use a about by storing the whisky in oak barrels for
still which, like the original Coffey still, has just periods of time that depend on traditional
two columns: a beer stripper (or analyzer) and a practice and legal requirements. In general,
rectifier. whiskies are matured for far longer than the
A description of the operation of two column legally-required period of time. The raw spirit is
continuous stills in the manufacture of Scotch taken by pipeline from the distillation plant to
grain whisky has come from Pyke (1965). In the tank house where it is diluted with water to
order to obtain whisky of high quality from these the required strength and then transferred into
stills, they must be operated such that the alcohol barrels.
concentration of the spirit at the spirit draw tray Maturation in barrels is accompanied by a loss
in the rectifier is not less than 94.17o GL. The of liquid by evaporation, and the relative rates
manner in which the precise control of still of loss of water and of alcohol determine
operation can affect the composition of the whether the aged whisky has a higher or lower
whisky is shown in Figure 12. As illustrated, if alcoholic strength than that at filling. In Scotland,
conditions are changed in either direction on the where the barrels of whisky are stored in cool,
abscissa, the concentration of congeners will unheated, but humid warehouses, the alcoholic
alter with a possible adverse effect on final strength decreases (Valaer, 1940). In contrast
product quality. Valaer and Frazier (1936) reported that in the
US storage conditions cause an increase in
alcoholic strength. Maturation in barrels is also
MATURATION AND AGING accompanied by changes in the chemical
composition of the whisky. These changes are
Freshly distilled whisky of any type is very attributable to extraction of wood constituents
different from the spirit that is later bottled, either from the barrel, oxidation of components present
singly or blended. The transformation is brought in the original whisky as well as those extracted
212 T.P. Lyons
102
10 90
Ethanol 80
1 60
50
Composition (% w/w)
n-Butanol 40
10-1 30
n-Propanol 20
Total esters
10
Total aldehydes
0
10-2
Total acids
10-3 Furfural
10-4
0 5 10 15 20 25 30
Tray number
Figure 12. Changes in composition of the vapor at different trays in a Coffey still rectifier used in the manufacture
of Scotch grain whisky (Pyke, 1965).
Production of Scotch and Irish whiskies: their history and evolution 213
from the wood, reactions between components desired value by adding caramel. Some
in the whisky and removal and oxidation of brownish pigment is extracted from the casks,
highly volatile sulfur components by the carbon but this may not be sufficient to provide the
char on the inner surface of the barrel. desired color. Finally, the whisky is clarified for
Some of the earlier investigators reported on bottling by filtration through sheets of cellulose
changes in the composition of the major classes (Simpson, 1968). Chill filtration may also be
of organoleptically-important compounds practiced, since it removes tannin material from
during maturation in barrels. Thus, Schidrowitz the whisky and prevents subsequent appearance
and Kaye (1905) found increased concentrations of haze.
of volatile acids in aged whiskies, a trend also
described in the report of the Royal Commission
on Whisky and Other Potable Spirits (1909). EFFLUENT DISPOSAL AND SPENT GRAINS
Several reports followed. Liebmann and Scherl RECOVERY
(1949), for example, reported increased
concentrations of acids, furfural, tannins and Traditionally effluents from whisky distilleries
color with maturation in barrels. The arrival of were disposed of in the most convenient manner.
the gas liquid chromatograph and HPLC greatly Spent grains were retailed, often quite cheaply,
accelerated research on this topic; and more to local farmers as animal fodder while pot ale
recent data on chemical changes that take place and spent lees were simply discharged into the
during maturation are described later in this local sewer, stream or river. This is no longer
chapter. the case, mainly because of the distiller’s
Maturation of whisky in oak barrels is an awareness of the nutritional value (largely the
expensive process; and it is hardly surprising protein content) of some of these effluents, and
that consideration has been given to methods public awareness of environmental problems
for acceleration. Jacobs (1947) described several arising from uncontrolled disposal of effluents
of these methods including pretreatment of the into waterways. Simpson has described the
beer with activated carbon, chemical treatment production of ‘distillers dark grains’. These
of the whisky to convert aldehydes into esters processes are widely used to dispose of effluents
and use of oxidation treatments. Such techniques from whisky distilleries, especially where the
are not used in the industry, which adheres to traditional methods of disposal are forbidden or
the traditional nature of the whisky-producing uneconomic.
process.
alcohols in Scotch whiskies imported into sensitivity and discrimination, a major problem
Australia. in analyzing whisky by GLC is the
overwhelming preponderance of ethanol and
water. Only one volatile compound, namely
CONCENTRATIONS OF ORGANOLEPTICALLY isoamyl alcohol, is likely to be present in a
IMPORTANT COMPOUNDS concentration exceeding 0.01%; while most of
the others are present in concentrations that
Since the introduction of GLC into distillery rarely exceed 50 ppm. Indeed many compounds
laboratories, several reviews have been now understood to have an important impact on
published on the composition of the major whisky flavor are present at ppb levels.
flavor-producing compounds in whiskies Carter-Tijmstra (1986) described a technique for
(namely higher alcohols, esters, carbonyl measuring dimethyl trisulfide, a compound with
compounds, organic acids, aromatic a threshold of only 0.1 ppb present in whisky at
compounds) and on the identification of concentrations below 50 ppb. Analyses are most
individual compounds that make up these conveniently conducted on extracted fractions
fractions. Higher alcohols, which are still of the different classes of compounds. When
routinely determined with GLC using a polar direct analysis of whiskies has been employed,
stationary phase (Aylott et al., 1994), are only a limited number of components have been
quantitatively the most important. Scotch malt determined (Morrison, 1962; Bober and
whiskies are richest in higher alcohols, with Haddaway, 1963; Singer and Stiles, 1965).
contents often well over 2 g/L. Free fatty acids Recent developments in headspace analysis
are relatively volatile and make a major using trapping and thermal desorption
contribution to the organoleptic qualities of techniques (Pert and Woolfendon, 1986) have
whiskies. Concentrations of acids in some Scotch enabled analysis of the more volatile
malt whiskies can be as high as 0.4-1.0 g/L components of whisky flavor whilst avoiding
absolute alcohol (Duncan and Philip, 1966). interference from high concentrations of other
Roughly comparable concentrations of esters are congeners. Headspace analysis using trap and
found in whiskies, although those produced with purge techniques would now be the method of
pot stills generally have higher concentrations choice for measuring highly volatile sulfur
than those from continuous stills. Ester compounds such as dimethyl trisulfide.
concentrations in Scotch and Irish pot still Some idea of the variety of compounds
whiskies have been reported in the range of 0.27- detected in whiskies came from a compilation
0.87 g/L absolute alcohol (Valaer, 1940). Lighter of both published and unpublished sources
whiskies contain lower concentrations of (Kahn 1969, Kahn et al., 1969). Of the some
carbonyl compounds, although the 200 compounds listed there are 25 higher
concentration varies with the brand. Grain alcohols, 32 acids, 69 esters and 22 phenolic
whisky may have as little as 20 mg/l of compounds. Undoubtedly, this list could now
aldehydes, while in a mature Scotch malt whisky be extended quite considerably. Of the higher
the concentration may be as high as 80 mg/L alcohols, isoamyl alcohol and optically-active
(Duncan and Philip, 1966). amyl alcohol predominate, accompanied by
lower concentrations of isobutanol and n-
propanol. Characteristically, there are usually
CHEMICAL NATURE OF only low concentrations of n-butanol and
ORGANOLEPTICALLY-IMPORTANT secbutanol. The principal organic acid in
COMPOUNDS whiskies is acetic acid, which can account for
Duncan and Philip (1966) reviewed chromat- between 50 and 95% of the total content of
ographic and other methods used to separate the volatile acids determined by titration. Of the
various organoleptically important compounds remaining acids, caprylic, capric and lauric are
from whiskies. Even though analytical methods quantitatively the most important (Suomalainen
such as capillary column gas chromatography and Nykänen 1970a). Some of the characteristic
linked to a mass selective detector (GC-MS) flavor and aroma of Irish whiskies may be
(Aylott et al., 1994) have developed in terms of attributed to somewhat higher concentrations of
Production of Scotch and Irish whiskies: their history and evolution 215
compounds. Interestingly, the relative used to ferment the wort and the charred oak
contributions made by the different classes of barrels in which the whisky is matured.
compounds are not very different from the Suomalainen and Nykänen (1966) fermented a
contributions Harrison (1970) reported that they nitrogen-free medium containing sucrose with
make towards the taste of beers. a strain of Saccharomyces cerevisiae and distilled
Threshold values can be assessed not only for the fermented medium either after removing the
individual components in whisky, but for the yeast by centrifugation or with the yeast
total aroma of the beverage. Salo (1975) remaining in the medium. Gas chromatographic
examined this by diluting different whiskies with analyses of these distillates are reproduced in
water until the characteristic whisky aroma could Figure 14, which also shows a chromatogram
only just be recognized. Values for the threshold of Scotch whisky for comparison. There is
dilution of several commercial whiskies shown clearly a similarity among the three analyses;
in Table 6 reflect the differences in aroma although differences, such as the higher
strength for several different commercial proportion of isoamyl alcohol in the distillate
whiskies. from the fermented medium, can be detected.
An excellent review of how these Also worth noting is the greater concentration
characteristics can be evaluated from an of ethyl caprate in the distillate from the spent
organoleptic point of view comes from Jack
(2003). Sensory data are only as good as the Pungent
2.5
sample preparation and guidelines were set as Stale Phenolic
applicable to the Scotch whisky industry. It is 2
The two main sources of the organoleptically Figure 13. Impact of temperature on flavor profile of Scotch
important compounds in whisky are the yeast whisky (adapted from Jack, 2003).
*Salo, 1975.
Production of Scotch and Irish whiskies: their history and evolution 217
medium containing yeast as compared with that Extensive research in Scotland revealed the
obtained by distilling the medium from which mechanism of ethyl carbamate production and
yeast had been removed. It would be interesting facilitated the introduction of very effective
to learn of the importance of yeast strain in control measures. These measures maintain
production of organoleptically-important levels close to zero and always less than 30 ppb
compounds in whisky. Unfortunately, there is a in distilled whisky spirit.
lack of published data on this matter. Cook (1990) reviewed the outcome of this
research and the resulting control procedures.
Scotch whisky When barley sprouts during malting, a
Isoamyl Alcohol
Ethyl Caprylate
Ethyl Caprate
Ethyl Laurate
Ethyl Palmitate
glycoside, epiheterodendrin (EPH) is present in
the acrospire. This glycoside, which survives
kilning and mashing, is extracted into the wort
and converted by yeast enzymes to glucose and
isobutyraldehyde cyanohydrin (IBAC). The
IBAC is stable during fermentation, but when
Sugar fermentation heated above 50oC at distillation it breaks down
Isoamyl Alcohol
Ethyl Caprylate
Ethyl Palmitate
Yeast separated
before distillation of potential routes nitriles can follow including
reaction with other beer components or complex
reactions often mediated indirectly by copper.
Some of the volatile nitriles may escape these
reactions and pass into the distillate. A number
Sugar fermentation of reactions can remove the nitriles from the
Isoamyl Alcohol
Ethyl Caprylate
Ethyl Caprate
Ethyl Laurate
Ethyl Palmitate
Distilled with yeast spirit, one being the reaction with ethanol to form
ethyl carbamate.
Two control strategies may be used to prevent
ethyl carbamate reaching the final spirit. Firstly,
some varieties of barley produce low levels of
the glycoside EPH (Cook, 1990), and plant
65
80 120 160 200 210
breeders are now concentrating on incorporating
Temperature (°C) this character in all varieties for the distilling
industry. The second strategy relies on the low
volatility of ethyl carbamate. Control of
Figure 14. Gas liquid chromatograms of aroma compounds distillation conditions eliminates all the volatile
produced by yeast in a nitrogen-free sugar fermentation, precursor formed by copper mediated reactions.
with a trace for comparison of the aroma compounds Non-volatile substances are formed prior to the
detected in a sample of Scotch whisky (Suomalainen final distillation in malt distillation or before
and Nykänen, 1966). rectification in grain distillation. Thus, ethyl
carbamate can be minimized in the final spirit.
The complexity of the processes for determining A considerable number of studies have been
the presence and concentration of compounds published on those organoleptically important
in whisky may be illustrated by looking at one compounds arising either directly or indirectly
compound for which this process has been from the oak barrels in which whisky is matured.
elucidated in detail. In the 1980s ethyl carbamate, The increase in coloring, tannin, dissolved solids
a naturally-occurring component of many and acid concentrations are not observed when
alcoholic drinks, was identified as being whisky is stored in glass, which is proof of the
undesirable. Canada specified by regulation a importance of the oak barrels in the maturation
maximum limit for ethyl carbamate in whisky process. An analysis of heartwood of the
spirit of 150 ppb and the US set guidelines of American oak (Quercus albus) gave cellulose
120 ppb. Control of this compound was only (49-52%), lignin (31-33%), pentosans (or
possible after the processes resulting in its hemicelluloses, 22%) and compounds extracted
presence in spirit were fully understood. with hot water and ether (7-11%; Ritter and Fleck,
218 T.P. Lyons
1923). However, when charred oak sawdust was evaporation and adsorption on the charred
directly extracted with water or 96o GL ethanol, surface of the barrel (Perry, 1986). Lastly there
the extracts obtained differed markedly in odor are reaction processes including establishment
from aged whisky. Moreover, none of the of equilibria among acetaldehyde, ethanol and
various fractions of ethanol-soluble oak acetal (Perry, 1986), polymerization reactions
extractives contained flavors that resemble (Nishimura and Matsuyama, 1989) and
mature whisky (Baldwin et al., 1967). As a oxidation-reduction reactions (Perry, 1986;
result, it is now generally held that the Connor et al., 1990). Many of the reactions
maturation process involves not only extraction involve, and are indeed dependent on, the
of compounds from the oak but also chemical components extracted from the wood.
modifications of at least some of the compounds Looking at several of these reactions in more
extracted from the wood. detail will serve to illustrate the complexity of
For a long time, most of the work reported on the maturation process. The work at the Seagram
this aspect of maturation of whisky came from laboratories, whilst focused on bourbon, is
the laboratories of Joseph E. Seagram and Sons directly relevant to Scotch and Irish whiskies for
in the US. More recently, accounts of the which once-used bourbon barrels are extensively
mechanisms of Scotch whisky maturation have utilized for maturation. Changes in the
been given by Philip (1989) and Perry (1986) concentrations of organoleptically-important
and on the maturation of whisky generally, by compounds during a 12 year storage of a 109o
Nishimura and Matsuyama (1989). A theme from US proof (54.5o GL) bourbon, on a 100o proof
all this work has been the identification of a (50 o GL) basis, are shown in Figure 15. The
number of mechanisms of maturation common nature and origin of some of these compounds
to all whiskies. These divide into addition, have been examined in some detail. Among the
subtraction and modification by reaction. There aldehydes, scopoletin and the aromatic
is addition of components from the oak wood, aldehydes syringaldehyde, sinapaldehyde,
including those derived from lignin, tannins and coniferaldehyde and vanillin are important.
oak lactones. There is the subtraction of volatile According to Baldwin et al. (1967), these
compounds from the maturing whisky by compounds could be formed by ethanol reacting
260
240
Concentration (g/100L) at 100° proof
220
Total solids
200
180 1.8
160 1.6
140 1.4
Color
120 1.2
100 1.0
80 0.8
ids
Total ac
60 0.6
40 Tannins 0.4
Esters Aldehydes
20 0.2
Furfural
0 0
1 2 3 4 5 6 7 8 9 10 11 12
Age (years)
Figure 15. Changes in composition of the vapor at different trays in a Coffey still rectifier used in the manufacture
of Scotch grain whisky (from Pyke, 1965).
Production of Scotch and Irish whiskies: their history and evolution 219
with lignin in the oak wood to produce coniferyl compounds extracted from charred oak wood
alcohol and sinapic alcohol. Under mildly depend on the ethanol concentration of the
oxidizing conditions, these alcohols could be whisky. It is to some extent an advantage to
converted into coniferaldehyde and mature whisky at a high proof, since this requires
sinapaldehyde, respectively. Vanillin could then fewer barrels and saves on storage space. Until
arise from coniferaldehyde, and syringaldehyde 1962 the US Treasury Department limited the
from sinapaldehyde. The increase in aldehyde barrelling proof of whisky to a maximum of 110o
content during maturation is also attributable in US proof (55o GL). In anticipation of this limit
part to formation of acetaldehyde by oxidation being raised to 125 o US proof (62.5 o GL),
of ethanol. Formation of ethyl acetate probably Baldwin and Andreasen initiated a series of
accounts for the steady rise in the ester content experiments in 1962 to establish the importance
of whisky during maturation. of barrelling proof on changes in color and
Several other groups of compounds not concentrations of organoleptically-important
described in Figure 14 are also important in the compounds during maturation of bourbon
maturation process. Monosaccharide sugars are whiskies. Their report in 1973 indicated that
found in mature whisky, and probably arise from color intensity and congener concentration of
the pentosans and other polysaccharides in the whiskies matured for 12 years decreased as the
oak wood. Otsuka et al. (1963) reported that a barrelling proof was raised from 109o US proof
mature Japanese whisky contained xylose, (54.5o GL) to 155 o US proof (77.5 o GL). The
arabinose, glucose and fructose, while Black and one exception was the higher alcohol content,
Andreasen (1974) added rhamnose to this list which remained approximately constant.
when they analyzed a mature bourbon. The latter
workers found that the concentrations of
arabinose and glucose increased at a faster rate Is it really Scotch?
than those of xylose and rhamnose over a 12
year maturation period. Salo et al. (1976) also Analysis of the variety of compounds in spirit is
detected low concentrations of mannose and also of interest for reasons of identifying or
galactose in a matured Scotch malt whisky in authenticating product origins or types. Adam
addition to the sugars already noted. The et al. (2002) investigated whether malt, blended
concentrations of sugars in mature whiskies (of and grain whiskies could be differentiated based
the order of 100 mg/L) are too low to suggest on content of various metals. While it was not
any sweetening effect on the beverage. Phenols possible to define a ‘metal fingerprint’ that would
are also detectable in mature whisky, although identify a whisky as to origin, malt whiskies had
some of these probably arise during mashing markedly higher concentrations of copper than
(Steinke and Paulson, 1964) or from malt blended or pure grain whiskies. Differences were
produced using peat-fired kilns (MacFarlane, significant with malt whiskies containing 385-
1968). However, Salo et al. (1976) reported an 480 ng Cu/mL and grain and blended whiskies
increase during a one year maturation of a contained 131-242 ng/mL. Since malt Scotch is
Scotch malt whisky in the concentration of produced in traditional copper pot stills while
eugenol, which is a major phenol extracted from grain whisky used in blending is made in
oak chips by ethanol (Suomalainen and continuous patent stills, copper content was
Lehtonen, 1976). Also present in mature suggested as a means of distinguishing a malt
whiskies are sterols, which may precipitate in Scotch from a blended product.
bottled whisky stored at room temperature. Another approach to distinguishing between
Black and Andreasen (1973) found campesterol, malt and blended products is based on the
stigmasterol and sitosterol in mature bourbon, differences between barley, which is used to
in addition to sitosterol-D-glucoside, although produce malt Scotch and the corn (maize) that
the possibility that some of these were formed is typically used in production of grain whisky.
during mashing cannot be excluded. Finally, Parker et al. (1998) found that the ratio of 12C
reference has already been made to the whisky and 13 C isotopes in volatiles such as
lactone, ß-methyl-octalactone, and its origin. acetaldehyde, ethyl acetate, n-propanol and
Not surprisingly, the nature and amounts of others differed among whiskies. This was a
220 T.P. Lyons
Harrison, G.A.F. 1970. Journal of the Institute Merritt, N.R. 1967. Journal of the Institute of
of Brewing 76:486. Brewing 73:484.
Harrison J.S. and J.C.J. Graham. 1970. In: The Morrison, R.L. 1962. American Journal of
yeasts. (A.H. Rose and J.S. Harrison, eds.). Vol. Enology and Viniculture 13:159.
3:283-348. Academic Press, London. Murtagh, J.E. 1970. M.Sc. Thesis: University
Heukelekian, H. and A. Heller. 1940. Journal of College of North Wales, Bangor.
Bacteriology 40:547. Nettleton, J.A. 1913. The manufacture of whisky
Hopkins, R.H. and S. Wiener. 1955. Journal of and plain spirit. Cornwall and Sons, Aberdeen.
the Institute of Brewing 61:493. p. 616.
Hough, J.S., D.F. Briggs and R. Stevens. 1971. Nishimura, K. and M. Masuda. 1971. Journal
Malting and brewing science. Chapman and of Food Science 36:819.
Hall Ltd., London. p. 678. Nishimura, K and R. Matsuyama. 1989. The
Institute of Brewing Analysis Committee, Science and Technology of Whiskies. (J.R.
Recommended methods of analysis. 1975. Piggott, R. Sharp and R.E.B. Duncan, eds.).
Journal of the Institute of Brewing 81:368. Longman Scientific and Technical, Harlow. pp.
Jacobs, M.B. 1947. American Perfumery and 235-263.
Essential Oil Review. p 157. Otsuka, K. and K. Morinaga and S. Imai. 1963.
Jack, F. 2003. Development of guidelines for Nippon Jozo Kyokai Zasshi 59:448.
the preparation and handling of sensory Parker, I.G., S.D. Kelly, M. Sharman, M.J.
samples in the Scotch whisky industry. J. Inst. Dennis and D. Howie. 1998. Investigations into
Brewing 109:114-119. the use of carbon isotope ratios (13C/12C) of
Kahn, J.H. 1969. Journal of the Association of Scotch whisky congeners to establish brand
Official Analytical Chemists 52: 1166. authenticity using gas chromatography -
Kahn, J.H., P.A. Shipley, E.G. Laroe and H.A. combustion isotope ratio mass spectrometry.
Conner. 1969. Journal of Food Science Food Chem. 63:423-428.
34:587. Perry, D.R 1986 Proceedings of Second Avie-
Kleber, von W. and N. Hurns. 1972. Brauwiss- more Conference on Malting, Brewing and
enschaft 25: 98. Distilling. pp. 409-412.
Kulka, D. 1953. Journal of the Institute of Pert, D. and E. Woolfendon. 1986. Proceedings
Brewing 59: 285. of Second Aviemore Conference on Malting,
Liebmann, A.J. and B. Scherl. 1949. Industrial Brewing and Distilling. pp. 417-424.
and Engineering Chemist 41:534. Philip, J.M. 1989. The Science and Technology
Lloyd Hind, H. 1948. Brewing Science and of Whiskies. (J.R. Piggott, R. Sharp and R.E.B.
Practice, Vol. 1:505. Chapman and Hall, Duncan, eds). Longman Scientific and
London. Technical, Harlow. pp. 264-294.
Lyons, T.P. 1974. The Brewer. p 634. Preece, I.A. 1947. Journal of the Institute of
McDowell, R.J.S. 1975. The Whiskies of Brewing 53:154.
Scotland. 3rd edit. John and Murray, London. Preece, I.A. 1948. Journal of the Institute of
p. 166. Brewing 54:141.
MacFarlane, C. 1968. Journal of the Institute of Preece, I.A. and M. Shadaksharaswamy. 1949a.
Brewing 74: 272. Journal of the Institute of Brewing 55:298.
MacFarlane, C., J.B. Lec and M.B Evans. 1973. Preece, I.A. and M. Shadaksharaswamy. 1949b.
Journal of the Institute of Brewing 79:202. Biochemical Journal 44:270.
McGuire, E.B. 1973. Irish Whiskey. Gill and Preece, I.A. and M. Shadaksharaswamy. 1949c.
MacMillan, Dublin. p. 462. Journal of the Institute of Brewing 55:373.
MacKenzie, K.G. and M.C. Kenny. 1965. Pyke, M. 1965. Journal of the Institute of
Journal of the Institute of Brewing 71:160. Brewing 71:209.
Mann, E.A. 1911. Government Analyst for West- Rae, I.L. 1967. Process Biochemistry I, 8:407.
ern Australia. Perth, W. Australia. pp. 1-12. Ritter, G.J. and L.C. Fleck. 1923. Industrial and
Engineering Chemistry 15:1055.
222 T.P. Lyons
Robb, J.M. 1950. Scotch Whisky. W. and R. Suomalainen, H. and M. Lehtonen 1976.
Cham-bers, London. p 197. Kemia-Kemi 3, 2:69.
Ross, J. 1970. Whisky. Routledge and Kegan Suomalainen, H. and L. Nykänen. 1966. Journal
Paul, London. p. 158. of the Institute of Brewing 72:469.
Royal Commission on Whisky and Other Potable Suomalainen, H. and L. Nykänen. 1970a.
Spirits. 1909. Appendix Q, p. 229. Naeringsmidelidustrien 23:15.
Salo, P. 1970. Journal of Food Science 35:95. Suomalainen, H. and L. Nykänen. 1970b.
Salo, P. 1975. Proceedings of the International Process Biochemistry. July 1.
Symposium on Aroma Research. Central The Scotch Whisky Order. 1990. Statutory
Institute for Nutrition and Food Research, Instruments No 998. HM Stationery Office,
TNO, Zeist, Netherlands. p 121. London.
Salo, P., L. Nykänen and H. Suomalainen. 1972. Thorne, C.B., R.L. Emerson, W.J. Olsen and
Journal of Food Science 37:394. W.H. Paterson. 1945. Industrial and
Salo, P., M. Lehtonen and H. Suomalainen. 1976. Engineering Chemistry 37:1142.
Proceedings of the Fourth Symposium on Valaer, P. 1940. Industrial and Engineering Che-
Sensory Properties of Foods. Skövde, Sweden. mistry 32:935.
Schidrowitz, P. 1902. Journal of the Chemical Valaer, P. and W.H. Frazier. 1936. Industrial and
Society. p 814. Engineering Chemistry 28:92.
Schidrowitz, P. and F. Kaye. 1905. Journal of Walker, J.W., D.M. Vaughan and H.L. Riffkin.
the Chemical Society. p. 585. 1990. Proceedings of Third Aveimore Conf-
Scotch Whisky Order. 1990. Stationery Office, erence on Malting, Brewing and Distilling. pp.
London. 421-426.
Sillett, S.W. 1965. Illicit Scotch. Beaver Books, Warricker, L.A. 1960. Journal of the Science of
Aberdeen. p. 121. Food and Agriculture 11:709.
Simpson, A.C. 1968. Process Biochemistry 3, Wilson, J. 1973. Scotland’s malt whiskies, a
1:9. dram by dram guide. Famedram Ltd.,
Simpson, B., A. Troon, S.T. Grant, H. Mac- Gartochan, Dunbartonshire, Scotland. p. 109.
Diarmid, D. MacKinlay, J. House and R. Wilson, J. 1975. Scotland’s distilleries. a visitor’s
Fitzgibbon. 1974. Scotch Whisky. Mac-Millan, guide. Famedram Ltd., Gartochan, Dunbart-
London. p. 120. onshire, Scotland. p. 104.
Singer, D.D. and J.W. Stiles. 1965. Analyst Wilson, Ross. 1959. Scotch made easy. Hutchin-
90:290. son, Ltd., London. p 336.
Stark, W.H. 1954. Industrial Fermentations (L.A. Wilson, Ross. 1970. Scotch: the formative years.
Underkofler and R.J. Hickey, eds). Chemical Constable, London. p 502.
Publishing Company, New York. Vol. 1, pp. Wilson, Ross. 1973. Scotch, its history and
17-72. romance. David and Charles, Newton Abbot,
Steinke, R.D. and M.C. Paulson. 1964. Journal Devon, England. p 184.
of Agricultural and Food Chemistry 12: 381. Wobben, H. J., R. Timmer, T. ter Heide and P.
Suomalainen, H. 1971. Journal of the Institute de Valois, P. 1971. Journal of Food Science
of Brewing 77:164. 35:464.
Zobell, C.E. 1943. Journal of Bacteriology
46:39.
Tequila production from agave 223
Chapter 15
and bacteria that produce ethanol, a diversity of codex, which notes that certain tribes had learned
chemical compounds, and some polymers that to cook agave plants and used them as food and
give a sticky consistency to the final product to compensate for the lack of water in desert
(Rzedowski, 1978; Sánchez Marroquin and lands. Also, these tribes discovered that cooked
Hope, 1953). Pulque is sometimes mixed with agave when soaked in water would ferment,
fruits or vegetables, but has poor stability as it is producing a much appreciated beverage. In fact,
neither distilled nor pasteurized. this primitive and rudimentary method was used
Agave plants still serve as food in some states for centuries to produce beverages from agave,
of Mexico; and other fermented regional considered a sacred plant possessing divine
beverages are produced (e.g., ‘Sotol’ in the state properties. In other codices, such as Nutall,
of Chihuahua and ‘Bacanora’ in the state of Laud, Borgia and Florentine, there are many
Sonora), but only tequila and more recently references to uses of the agave plant for soap
mezcal have reached international recognition. manufacture, a source of fiber, footwear,
Another difference between tequila and mezcal medicine and sewing needles as well as thread,
or other regional drinks is that both are subject paper and rope. In fact, Indians distinguished
to an official standard that for tequila is NOM- the different species of agave by color, size, stem,
006-SCFI-1994 (Secofi, 1994), and production leaf width and the different uses given each plant
is supervised by the Mexican government. Table (Muria, 1990). The great religious importance
1 shows the main differences among the regional of agave was apparent in those codices, as only
beverages. warriors and priests used fermented drinks in
ritual ceremonies.
In pre-hispanic Mexico the general name for
Origin and history of tequila all species of agave (or mezcal as it is also
known) was Metl, which is a representation of
The word ‘tequila’ is believed to originate from the goddess Mayahuel (Figure 1). The alcoholic
the tribe of ticuilas who long ago inhabited the drink produced was called Iztac octli (white
hillside of a volcano bearing the same name wine). The first Spaniards to arrive in Mexico
located near the city of Tequila. Another possible referred to the plant as ‘maguey’, a name used
origin is the Nahuatl word tequitl, which means for an identical plant they had seen in the
work or employment, and the word tlan, which Caribbean Islands, where they first encountered
means place. Therefore ‘tequila’ would mean new world plants and animal life (Bottorff,
place in which labor or work is done. 1971). It was not until the arrival of the
The oldest reference to the existence of agave Spaniards, who brought knowledge of
and its different uses is from the era before the distillation techniques, that tequila took its
Spaniards in several codices preserved to the present form (Goncalves, 1956). Tequila is
present time (a codex, from the Latin codex considered the first distilled beverage in North
meaning board or writing tablet, is a manuscript America and has been given several different
volume, especially of a classic work or scripture). names including agave wine, mezcal wine and
The most important is the Tonalmatlnahuatl finally tequila.
Tamaulipas (11)
Guanajuato (7)
Nayarit (8)
Jalisco (all)
Michoacán (30)
Figure 2. States and municipalities for agave cultivation and tequila production in Mexico.
226 M. Cedeño Cruz
immediately before the rainy season, from June theobromae and Colletotrichum agavae may
to September, so that the plants do not suffer cause serious damage to agave leaves (Halffter,
from water stress during the first year of growth. 1975; Agricultural Research Service, 1972).
Propagation is accomplished by vegetative Agave plants, adapted to hot and very arid
means in agave. Sexual reproduction via seeds conditions, employ crassulacean acid
is not usual. Asexual bulbils develop in the metabolism (CAM). Crassulacean acid
inflorescence at the base of the flowers, metabolism is similar to photosynthetic processes
producing small plants that after some time in other hot temperature-adapted plants in that
detach themselves from the floral peduncle and the enzyme phosphoenolpyruvate (PEP)
fall to the soil where they root. Another mode carboxylase is used to fix CO2 into a 4-carbon
of asexual propagation is by suckers, which are compound (C 4 metabolism). However in most
a characteristic type of lateral bud or branch hot season plants the light-dependent reactions
developing at the base of the main stem. Plants and CO2 fixation are separated in different cells
developing near each mother plant are separated of the leaf. In contrast, in plants with crassulacean
at the age of 3-4 years. These baby plants are acid metabolism such as cacti and agave these
called ‘first-class seed’ because they are better reactions occur at different times of the day.
and healthier (Sánchez, 1991). In practice, people Carbon dioxide (CO 2) enters the leaf through
use the word ‘seed’ to refer to such young stomata with the simultaneous loss of water. By
plants, but from a botanical point of view these only requiring the stomata to be open during the
are rhizome shoots or suckers (Valenzuela, cooler night for CO 2 uptake and keeping them
1992). Plant cell culture is used by some tequila closed during the hotter day, the agave plants
companies in order to maintain plants of lose much less water (Ehrler, 1967). Malic acid
consistent quality, reduce the cost of labor for formed by CO2 fixation using PEP carboxylase
sowing the plants, eliminate use of fungicides accumulates during the night, and is broken down
and antibiotics during sowing, and have bigger during the day (Figure 4). The CO 2 that is
plants with higher inulin content at the end of released can be re-fixed by the normal
the cultivation period. photosynthetic Calvin cycle (C3 metabolism) to
Land for agave cultivation must be cleared and produce sugars and then inulin.
deep-ploughed, sometimes twice. Agave is The thick cuticle overlying the epidermis,
planted approximately 2-4 m apart in straight which is quite evident in tequila agave,
lines called ruts. Sowing is done by hand in holes apparently prevents damage to the leaf from high
15 cm in depth. Plant density is around 2000- temperatures (Casado and Heredia, 2001). This
4000 plants per hectare, depending on the waxy cuticle produces turbidity in tequila
plantation system used; and yields can be because it dissolves in the distillation step and
between 30,000 and 200,000 kg/ha, assuming produces a haze in the final product when it is
that the weight of a harvested plant varies from diluted or cooled. One way to avoid this is to
15 to 50 kg. This variation is caused by treat tequila with activated charcoal and to filter
differences in soil conditions, quality of plants it through pure cellulose filter pads. This,
sown, rainfall, pests and fertilization. Sometimes unfortunately, results in the loss of some aromas.
agave is interseeded with nitrogen-fixing crops Agave fertilization is determined by soil
such as peanuts, beans, chickpeas or soybeans. composition, plant age, and the type of chemical
After the agave has been in the soil for a year, compound used. The normal procedure is to use
visual inspection is conducted and sick or dead urea as a nitrogen source in amounts of 30-70 g
plants are replaced. This operation is called re- per plant incorporated directly into the soil. In
seeding. The percentage of dead plants depends some areas, phosphorus and potassium
on soil and plant characteristics, but is generally fertilization is also required. As some agave
from 8 to 15% (Pérez, 1990). regions are also involved in cattle, swine, and
Agave plants regularly host borer insects that chicken production, manure is sometimes
live in the stems, leaves and fruits during the employed for fertilization (GEA, 1992).
larval stage. These include butterflies of the The average maturity time for agave is seven
family Megathymidae and moths of the family years; and at this time the content of inulin is at
Prodoxidae. Also, the fungi Diplodia its highest level (Mendez, 1999). Every year
228 M. Cedeño Cruz
Day Night
Air
CO2
H2O
CO2
Calvin PEP
Cycle
Glycerol CO2 fixation
Growth
C4 acids
following planting, fields are cultivated to loosen are harvested; but on the 12th year, all plants
the soil and weed and pest control is carried out. remaining (the weakest plants) are cut. These
Each plant matures individually. Harvest begins are called ‘drag’ and are generally discarded.
in the seventh year. The leaves are cut from the Agave composition varies seasonally, but an
base and left in the field to recycle nutrients. average would be (wet basis) 27% (w/w)
The harvested plants free of leaves look like large reducing sugars, a juice content of 0.572 ml/g
pineapples and weigh from 20 to 90 kg. They and a pH of 5.2.
are transported to the distillery. Only the better In Figure 5 the amount of agave harvested for
plants, meaning those of good size and high tequila production is presented for the period
inulin content (measured as reducing sugars), 1995-2002. These figures show, first of all, a
800 Tequila100%
700 Tequila
Tons of agave ('000s)
Total
600
500
400
300
200
100
0
1995 1996 1997 1998 1999 2000 2001 2002
decrease in total agave production. This reflects advances that improve process efficiency and
the shortage caused by disease in agave consistency; and some have implemented a ISO-
plantations in the year 2000, which was mainly 9000 standard. Whatever the process used,
due to a bacteria Erwinia carotovora and a tequila manufacture comprises four main steps:
fungus Fusarium oxysporum, which destroyed cooking, milling, fermentation and distillation.
more than 25% of the total agave plants for An extra step, maturation, is required for aged
tequila. From this graph, it is evident that the products. Figure 6 shows a flow diagram for the
agave utilization for 100% tequila is growing, tequila production process, which is explained
since that beverage is being recognized as a in detail in the following paragraphs.
premium product not only in Mexico but also
overseas.
Processing harvested agave
Autoclave
Oven
Yeast
STILLAGE STILLAGE
DISSOLUTION OF
BOTTLING OTHER SUGARS
plants or those damaged by pests. At this point, hydrolyzed corn syrup may be used to formulate
a representative sample of agave is taken for the wort. All sugars used in wort formulation
laboratory analysis. Modified AOAC (1990) are routinely analyzed by measuring solids
procedures are used to determine reducing sugar content and reducing and fermentable sugar
content (after acid hydrolysis of inulin) along content.
with pH, moisture, dry weight, juice and ash
content. CH2OH
Agave heads usually weigh between 20 and H
O
H
60 kg, although some can reach 100 kg. They H Glucose
are cut to sizes that facilitate uniform cooking OH H
and handling. Different agave cutting systems OH
exist, but the use of axes and a specialized tool
H OH O
called ‘coa’ are the most popular. The heads are
cut in halves or quarters, depending on the CH2OH O
Fructose
weight, and the pieces are arranged manually in
an oven or autoclave. Band saws can also be H OH
used to cut the agave heads, since they are faster H
CH2
n
and less labor-intensive than the manual
procedure, but the belts break frequently because OH H
of the resinous consistency of agave. Some O O
CH2OH
factories tear uncooked agave first with a knife
and place the resulting pieces mixed with water H OH
into autoclaves to be cooked. H
CH2OH
OH H
OTHER RAW MATERIALS FOR TEQUILA
PRODUCTION Figure 7. Structure of inulin, in which chain length, n, varies
from 3 to 19, depending on the maturity of the agave plant
When producing 100% agave tequila, the only
(Mendez, 1999).
source of carbohydrate is the inulin hydrolyzed
from agave in the cooking step. Inulin, a polymer
of fructose with a terminal glucose, belongs to
the family of fructans (Figure 7). After starch,
fructans are the most abundant non-structural The cooking step: hydrolysis of inulin
polysaccharide found in nature.
Cooking the agave serves three purposes. First,
For other kinds of tequila the law permits use
the low pH (4.5) together with the high
of other sugars in amounts up to 49% by weight
temperature hydrolyzes inulin and other
in wort formulation. There are no legal
components of the plant. A recent study (Pinal,
specifications regarding the type of adjunct sugar
2001) showed that the concentrations of furfural,
sources to be used in tequila manufacture; and
5-hydroxymethyl furfural, 5-methylfurfural and
theoretically any kind of fermentable sugar is
2-acetylfuran increase with length of cooking
allowed in wort formulation. In practice and from
time. These components are important as they
an economic point of view, only cane sugar,
may influence fermentation rate as well as the
Piloncillo, cane molasses and acid- or enzyme-
organoleptic characteristics of the final tequila.
hydrolyzed corn syrup are employed. Cane
For this reason, time control during cooking is
sugar is received in 50 kg bags and stored in a
very important. In addition, cooked agave has a
dry, cool place for subsequent utilization.
soft consistency that facilitates the milling
Piloncillo consists of brown cones of crystallized
operation.
complete cane juice, sometimes individually
In the pre-Hispanic era, agave was cooked in
wrapped in corn or cane leaves and packed in
holes filled with stones heated using wood for
sacks. Cane molasses is also used but it is difficult
fuel. The stones retained the heat for the time
to handle and there is risk of spoilage over a
needed to cook the agave. Nowadays some
long storage period. Acid- or enzyme-
Tequila production from agave 231
distillers have replaced the heated stone holes additional steam, cooking slowly in the
with brick ovens and heating is accomplished remaining heat. This step produces a syrup with
by steam injection after the chopped raw agave a high sugar concentration (>10% by weight)
has been introduced into the oven. Oven cooking that is later used to formulate the initial wort. To
is slow, and steam injection lasts around 36-48 calculate the yield and efficiency of this step,
hrs to obtain temperatures of 100°C. After that the amount of cooking honey and its reducing
period, the steam is shut off and the agave is left sugar content as well as the cooked agave are
in the oven for a further two days to complete measured. Figure 8 illustrates the temperature
the cooking process. During this step, a sweet profile over time for cooking agave in an
liquid called ‘cooking honey’ is collected and autoclave compared with an oven.
used later as a source of free sugars, mainly The main difference between autoclaved and
fructose. Also during this step some of the sugars oven-baked agave is that careful control of
are caramelized; and some of the compounds cooking time, temperature and steam pressure
that contribute significantly to the aroma and must be maintained in autoclaves to prevent
flavor in wort formulation are due to its high overcooking or burning the agave. Overcooking
content of fermentable sugars (>10% w/v). gives a smoky taste to the tequila, increases the
Finally, the oven door is opened to allow the concentration of furfural in the final product and
cooked agave to cool. The agave is then ready reduces ethanol yield due to the caramelization
for milling. of some of the fermentable agave sugars. This
In most distilleries brick ovens have been is why some factories with both cooking systems
replaced by steel autoclaves. Autoclaves have reserve the ovens for their better-quality
superior efficiency and allow good pressure and products. Although it is easier to obtain well-
temperature control, enabling homogeneous and cooked agave in an oven than in an autoclave,
economic cooking. In a typical autoclave there is no difference in terms of flavor and
cooking operation, steam is injected for 1 hr so fermentability between agave cooked in
that steam washes the agave. This condensed autoclaves or in ovens if both are correctly
liquid is called ‘bitter honey’ and is discarded controlled.
because it contains waxes from the agave cuticle
and has a low sugar content (<1 % w/w). Steam
is injected for an additional 6 hrs to obtain a Extraction of agave juice: milling
pressure of 1.2 kg/cm 2 and a temperature of
121°C. At the end of that time the agave remains Milling has gone through three historical stages.
in the autoclave for another 6 hrs without In ancient days cooked agave was crushed with
140
Autoclave
120
Oven
Temperature (°C)
100
80
60
40
20
0
0 20 40 60 80
Time (hours)
wood or steel mallets to extract the juice. Later, board, or enzymes (cellulase and pectinase).
a rudimentary mill consisting of a large circular Many of these projects are at the laboratory stage,
stone 1.3 m in diameter and 50 cm thick was and there is not enough information to evaluate
used. Driven by animals, the stone turned in a feasibility.
circular pit containing cooked agave and In milling, as in all steps of the tequila process,
extracted the juice. The resulting juice was a sugar balance is computed to determine the
collected by hand in wooden basins and carried yield. If the yield decreases, the extraction
to fermentation tanks. By the 1950s modern pressure in the mill and the water/agave ratio are
systems were implemented in which cooked increased to improve efficiency.
agave was passed through a cutter to be
shredded (except in factories that did this
operation before cooking); and with a Fermentation
combination of milling and water extraction,
sugars were extracted. The mills used for agave WORT FORMULATION
are similar to those used in the sugarcane
industry but are smaller in size (normally 50 cm To produce 100% agave tequila, only agave may
wide). This system is still employed in most be used and the initial sugar concentration ranges
distilleries. from 4 to 10% w/v, depending on the amount of
In recent years the tequila industry has been water added in milling. When other sugars are
using a technology to extract fructose in cooked employed, they are previously dissolved and
agave or inulin in raw agave based on mixed with agave juice to obtain an initial sugar
countercurrent extraction by means of a piece concentration of 8-16%, depending on sugar
of equipment called a diffusion band. With the tolerance of the yeast strain. Wort formulation
use of this technology, used for many years in in most distilleries is based solely on previous
the extraction of sugar and alcohol in grapes in experience. A few distilleries base wort
Spain, the process has been improved and the formulation on composition of raw materials and
bagasse at the end of the extraction has nearly nutritional requirements for yeast growth and
no residual sugar. fermentation. In these distilleries response
Juice obtained in milling is mixed with the surface methodology is the preferred method to
syrup obtained in the cooking step and with a optimize nutrient concentrations, using
solution of sugars, normally from sugarcane (if fermentation efficiencies and taste of the
the tequila to be produced is not 100% agave), resulting tequila as responses (Montgomery,
and finally pumped into a fermentor. Although 1984). To correct nutritional deficiencies of
the amount of sugar employed as an adjunct is agave juice and sugars employed in the growth
regulated by law and must be less than 49% by and fermentation steps, urea, ammonium sulfate,
weight at the beginning of the fermentation, each ammonium phosphate or magnesium sulfate can
factory has its own formulation. be added. Because pH of the agave juice is
The milling step generates a by-product called around 4.5, there is no need for adjustment and
bagasse, which represents about 40% of the total the same wort composition is used for both
wet weight of the milled agave. Bagasse inoculum growth and fermentation.
composition (dry weight basis) is 43% cellulose,
19% hemicellulose, 15% lignin, 3% total
nitrogen, 1% pectin, 10% residual sugars and YEASTS
9% other compounds. The bagasse, mixed with Some companies do not inoculate a specific
clay, is used to make bricks; but it is also the strain of S. cerevisiae and instead allow natural
subject of research to find alternative uses. fermentation to proceed. Others inoculate the
Examples are use of bagasse as an animal feed wort with fresh packages of baker’s yeast or a
or as a substrate on which to grow edible fungi commercial dried yeast to obtain initial
(Iñiguez et al., 1990; 2001). Attempts are also populations of 20-50 x 106 cells/ml. The dried
underway to recover bagasse components yeasts were originally prepared for wine, beer,
(cellulose, hemicellulose and pectin) using high- whisky or bread production; and sometimes the
efficiency thermochemical reactors (Alonso et quality of the tequila obtained using these yeasts
al., 1993), to obtain furfurals, make particle is not satisfactory, with large variations in flavor
Tequila production from agave 233
and aroma. To achieve high yields and maintain a pleasant aroma to the final product. The
a constant quality in tequila, some companies difference in alcohol yield in tequila production
use yeast strains isolated from a natural using a selected commercial strain of S.
fermentation of cooked agave juice. Nutrients cerevisiae compared with wild yeast is presented
are added and special conditions such as a high in Figure 9. The higher alcohol concentration at
sugar concentration or temperature are the end of the fermentation process clearly
maintained. These isolated and selected yeast means better productivity and a lower
strains have been deposited in national production cost.
microbial culture collections, the most important Fermentation temperature is important in
being the Biotechnology and Bioengineering tequila production since most of the distilleries
Department Culture Collection of CINVESTAV- are located in regions where ambient
IPN, located In México City. temperatures are normally warm, especially
Toward the goal of improving fermentation during the summer, and may be as high as 37ºC.
productivity, one alternative is to use a strain of Not all strains of S. cerevisiae are resistant to
yeast capable of both efficiently converting wort such high temperatures; and the selection of a
sugars into alcohol and of producing the commercial strain is very important as it can
appropriate organoleptic compounds that impart represent an advantage in process yield. Figure
4. 00
Wild yeast 26°C
3. 50
Alltech Thermosacc 37°C
Alcoholvolume (%)
3. 00
2. 50
2. 00
1. 50
1. 00
0. 50
0. 00
0 20 40 60 80 10 0
Fermentation time (hours)
Figure 9. Alcohol concentration in a fermentation for tequila production using wild yeast compared with
a commercial strain of yeast.
5. 00
Alltech Thermosacc 37°C
4. 50
4. 00 Alltech Thermosacc 26°C
Alcoholvolume (%)
3. 50
3. 00
2. 50
2. 00
1. 50
1. 00
0. 50
0. 00
0 20 40 60 80 10 0
Fermentation time (hours)
Figure 10. Alcohol concentration in tequila fermentation using a thermotolerant commercial strain of yeast
at temperatures of 26 and 37ºC.
234 M. Cedeño Cruz
10 illustrates the fact that alcohol volume in the Otherwise, microorganisms present in the wort
fermentation process for a thermotolerant strain carry out the fermentation. If an inoculum is not
is better at 37ºC than at 26ºC. added, the fermentation could last as long as
seven days. With an inoculum the fermentation
time ranges from 20 hrs in the faster process to
INOCULUM GROWTH three days in the slower one.
Production of ethyl alcohol by yeast is
When an inoculum is used, it is grown in the associated with formation of many fermentation
laboratory from a pure culture of S. cerevisiae compounds that contribute to the final flavor of
maintained on agar slants in lyophilized form the tequila. These are organoleptic compounds
or frozen in liquid nitrogen. All laboratory or their precursors produced either in subsequent
propagation is carried out under aseptic maturation of the wort before it goes to the
conditions using a culture medium with the same distillation step, in the distillation process, or in
ingredients used in the normal process but the barrels if tequila is aged. The factors
enriched to promote cellular growth. The influencing formation of the organoleptic
inoculum is scaled-up with continuous aeration compounds in alcoholic beverages have been
to produce enough volume to inoculate reviewed by many authors (Engan, 1981; Berry,
fermentation tanks at 10% of the final volume. 1984; MacDonald et al., 1984; Ramsay and
Populations of 200-300 x 10 6 cells/ml are Berry, 1984(a); Geiger and Piendl, 1976).
normally achieved. Strict cleanliness is Experience in the tequila industry is that the
maintained in this step as bacterial contamination amount of organoleptic compounds produced
is highly undesirable. When contamination is is lower in fast fermentations than in slow
detected, antibiotics or ammonium bifluoride are fermentations. As a consequence, the flavor and
used as antimicrobial agents. Once an inoculum general quality of tequila obtained from worts
is grown, it is maintained by mixing 10% of the fermented slowly is best. The rate of
volume of an active culture with fresh agave fermentation depends mainly on the yeast strain
juice and nutrients. used, medium composition and operating
Although inoculation with commercial yeast conditions, as previously explained. The wort
greatly improves yield and turnover time, some sugar content decreases from an initial value of
companies prefer a more complex (in terms of 4-11% to 0.4% (w/v) reducing sugars if an
the microbial diversity) fermentation. While efficient yeast strain is employed. Otherwise, the
yields might be lower and turnover time higher, residual sugar content could be higher,
the range of microorganisms produces more increasing production costs.
compounds contributing to a more highly Fermentation vessels vary considerably in
flavored tequila. It is also important to recognize volume, depending on the distillery. Capacity
that a change in taste and flavor could negatively ranges from 12,000 liters for small tanks to
affect the market for a particular brand of tequila. 150,000 liters for the largest ones; and they are
For this reason it is very important that when a constructed of stainless steel in order to resist
commercial strain of yeast is used in the the acidity of the wort. Ethanol production can
fermentation process, the final aroma profile in be detected almost from the onset; and a pH drop
tequila must be evaluated using a sensory from 4.5 to 3.9 is characteristic of the
evaluation panel together with gas chromat- fermentation. The alcohol content at the end of
ography analysis. fermentation is between 4 and 9% v/v,
depending on the initial sugar concentration. In
order to increase the fermentation yield, in
FERMENTATION OF AGAVE WORT
addition to selection of a good yeast strain,
another option is the use of enzymes or enzyme
Once a wort is formulated with the required complexes, to convert residual polymers from
nutrients and the temperature is around 30°C, it agave into fermentable sugars, which are
may be inoculated with 5 to 10% (volume) of a converted mainly into alcohol improving the
previously grown S. cerevisiae culture with a productivity of tequila production. Figure 11
population of 100-200 million cells/ml. shows the results of an experiment using
Tequila production from agave 235
Alcoholvolume (%)
3
Control 35°
0
0 20 40 60 80 10 0 12 0 14 0
Time (hours)
Figure 11. Alcohol volume in the wort at different concentrations of an enzyme complex (RhizozymeTM)
for tequila production at 35ºC.
different doses of a commercial enzyme complex compounds used in the propagation step may
in the alcohol concentration of the wort. be used here to decrease common bacterial
Alcohol losses may be significant because contaminants found in tequila worts.
many fermentation tanks are open, allowing Lactobacillus, Streptococcus, Leuconostoc, and
evaporation of alcohol with carbon dioxide. Pediococcus are the most common
Some of the largest distilleries have a cooling contaminants, but Acetobacter may be found in
system that keeps fermentation temperature fermented worts that are left inactive for a long
within a tolerable range for yeast, but small time prior to distillation.
producers do not have these systems. The In contrast to other distilled beverages, the
fermentation temperature can exceed 40°C, organoleptic characteristics of tequila come from
causing the fermentation to stop with an the raw material (cooked agave) as well as from
accompanying loss of ethanol and flavors that the fermentation process. In most of the processes
consequently decreases yields and affects the used in the tequila industry, fermentation is
quality of the tequila. Fermentations carried out spontaneous with the participation of
with pure agave juice tend to foam, sometimes microorganisms from the environment, mostly
requiring the addition of silica. In worts with yeasts and a few bacteria. This peculiarity brings
added sugars, foaming is usually not a problem. about a wide variety of compositions and
Non-aseptic conditions are employed in organoleptic properties. However, the special
fermentation, and in consequence bacterial characteristics of the wort make it a selective
activity may increase. The size of the bacterial medium for the growth of certain yeasts such as
flora depends on a number of factors including Saccharomyces and to a lesser extent, acetic and
the extent to which bacteria grow during yeast lactic bacteria. In experiments isolating microbial
propagation (if used), the abundance of bacteria flora from musts of various origins, different
on the raw materials and hygiene standards in microorganisms were isolated. In most cases,
the distillery. There is no doubt that the activity this difference in flora is responsible for the wide
of these bacteria contributes to the organoleptic variety of organoleptic characteristics of tequila
characteristics of the final product. Occasionally, brands (Pinal, 1999). Where a single purified
the size of the bacterial population in fermenting strain is used, the final flavor and aroma are
wort may become too large (>20 x 10 6 cells/ more neutral since the bouquet created by the
ml), in which case the bacteria use the sugars, contribution of several stains is richer than that
decreasing ethanol yields and sometimes obtained from only one type of yeast. Moreover,
excreting undesirable compounds. The same when yeast produced for bakery applications is
236 M. Cedeño Cruz
used, the final product is also more neutral. It is whisky (Ramsay and Berry, 1984(b)) and beer
also recognized that when non-100% agave (García et al., 1994).
tequila is made, a poorer bouquet is obtained The carbon:nitrogen ratio also has a significant
because a more defined medium yields a more influence on higher alcohol production. In
defined product. tequila musts, which contain mainly fructose
(≈95%) as a carbon source and an inorganic
nitrogen source (ammonium sulfate), it was
ORGANOLEPTIC COMPOUNDS GENERATED found for both native and bakery yeast strains
DURING FERMENTATION that low carbon:nitrogen ratios result in low
amounts of isoamyl alcohol: 19 mg/L in bakery
Fusel oil strains and 30 mg/L for native strains vs 27 and
As in many other alcoholic fermentation 64 mg/L, respectively, for high carbon:nitrogen
processes, higher alcohols are the most abundant ratios. A similar relationship exists for isobutyl
compounds produced along with ethanol. We alcohol production (Figures 12 and 13).
have found (in decreasing order of abundance) Temperature is a third factor affecting isobutyl
isoamyl alcohol, isobutanol, active isoamyl and isoamyl alcohol production with higher
alcohol and phenylethanol. It has been temperatures (e.g. 38 vs 32°C) yielding higher
established that production of isoamyl and concentrations of those alcohols. On statistical
isobutyl alcohols begins after the sugar level is analysis it was found that in addition to the direct
lowered substantially and continues for several effects of yeast strain, carbon:nitrogen ratio and
hours after the alcoholic fermentation ends. In temperature, the interaction of these three factors
contrast, ethanol production begins in the first also had an impact on higher alcohol
hours of the fermentation and ends with concentration in tequila (Pinal et al., 1997).
logarithmic yeast growth (Pinal et al., 1997). These results are consistent with the fact that with
The most important factor influencing the high carbon:nitrogen ratios there is a tendency
amount of isoamyl alcohol and isobutanol is the to use amino acids as a nitrogen source, which
yeast strain. It was found that a native strain implies the production of fusel oil as a by-
isolated from tequila must produces a higher product (by the Erlich pathway). On the other
amount of such compounds when compared hand, variables such as the type of nitrogen
with a strain usually employed in bakeries. These source (urea or ammonium sulfate) or the
results agree with those reported for Scotch amount of inoculum used for fermentation had
8.4 140
120
Log cells/ml
8.0 100
Reducing sugars, g/L
Ethanol, g/L 80
60
40
20
Isoamyl + isobutyl alcohols, mg/L
7.2 0
0 5 10 15 20 25 30 35
Time (hrs)
Figure 12. Higher alcohol production in tequila wort by a baker’s yeast strain.
Tequila production from agave 237
8.4 140
120
Log cells/ml
8.0 100
Reducing sugars, g/L
80
60
40
Ethanol, g/L
Isoamyl + isobutyl 20
alcohols, mg/L
7.2 0
0 5 10 15 20 25 30 35
Time (hrs)
Figure 13. Higher alcohol production in tequila wort by a native yeast strain.
little or no effect on the production of higher amounts of methanol in musts with the same
alcohols. Pareto diagrams involving all the composition but fermented with different strains
variables tested are depicted in Figures 14 and (Téllez, 1999).
15.
Aldehydes
Methanol
Along with the production of ethyl acetate, the
Another characteristic compound present in oxidation of ethanol also generates acetal-
tequila is methanol. It is generally thought that dehyde, an intermediate in the production of
methanol is generated through hydrolysis of acetic acid. It is well known in commercial
methylated pectins present in the agave plant. practice that an oxidation process instead of
Nevertheless, it is also believed that some yeast fermentation begins after the sugar concentration
strains, natural or inoculated, have pectin methyl declines, thus provoking the increase in
esterases. Some authors describe various acetaldehyde levels. However, there is no formal
A -26.68
21.61
AD -15.56
10.32
BD 8.34
-7.38
ABD -6.71
Factors
-6.41
ABE 5.61 A: Strain
-5.10
B: Temperature
BE -4.65
3.70 C: Nitrogen source
CDE -3.09 D: Carbon:nitrogen ratio
-2.53 E: Inoculum amount
ACE 2.50
-1.93
E 1.00
0 10 20 30 40 50
Effect level
A -41.85
16.33
AD -13.41
12.94
BD -12.73
-8.47
ABD 8.13
Factors
7.35
ABE 6.99 A: Strain
-6.56
B: Temperature
BE 6.56
C: Nitrogen source
-6.12
CDE -6.03 D: Carbon:nitrogen ratio
-6.01 E: Inoculum amount
ACE 5.66
3.84
E 1.55
0 10 20 30 40 50
Effect level
report of such phenomena in tequila, as is ethanol. Acetic acid can also be produced by
described in beer (Hammond, 1991). the oxidation of ethanol when the fermentation
has ceased and an oxidative process starts on
Organic acids the surface of the fermentation tank by
Saccharomyces and many other yeasts such as
Small organic acids (up to six carbons) and larger Brettanomyces. Therefore, long fermentation
molecules (fatty acids) are produced during periods (a current practice in the tequila
fermentation. The smaller molecules can be industry) yield high ethanol oxidation. In
products of intermediate metabolism of the addition, in open fermentation tanks with worts
normal microbial flora; and their production at low pH containing alcohol, ethanol is also
depends on the presence of oxygen. The larger transformed to acetic acid (itself a precursor of
fatty acids are synthesized for membrane ethyl acetate) by bacteria of the genera
structures during cell growth and can also Acetobacter. Besides ethyl acetate, the presence
appear at the end of the fermentation when lysis of several other esters has been described
takes place. The presence of octanoic and including ethyl and isoamyl esters. Some of the
decanoic acids in the final product has been most important esters found in silver tequila are
described particularly for tequila. listed in Table 2.
Distilling
Distillation involves the separation and of trays. Steam is injected from the bottom in a
concentration of the alcohol from the fermented coil and strips the wort of its volatile
wort. In addition to ethanol and other desirable components. Vapors condense higher in the
secondary products, wort contains solid agave column, depending on component volatility,
particles consisting mainly of cellulose, pectin, allowing liquids to be drawn off or recycled at
and yeast cells in addition to proteins, mineral the various plates as appropriate. Sometimes
salts and some organic acids. Although many tequila obtained in this way is mixed with tequila
types and degrees of distillation are possible, the from pot stills to balance the amount of
most common systems used in the tequila organoleptic compounds because in general,
industry are pot stills and rectification columns. tequila obtained through continuous columns
The pot still is considered the earliest form of has less aroma and taste than tequila obtained
distilling equipment. It is of the simplest design, from pot stills.
consisting of a kettle to hold the fermented wort, The presence of methanol in tequila is still a
a steam coil, and a condenser or a plate heat subject of discussion because whether methanol
exchanger. Pot stills are often made of copper, is produced only by a chemical reaction or in
which ‘fixes’, according to Thorne et al. (1971), combination with a microbial hydrolysis has not
malodorous volatile sulfur compounds produced been satisfactorily demonstrated. The chemical
during fermentation. Batch distillation using pot reaction is demethylation of agave pectin by the
stills is carried out in two steps. First the high pH during cooking and the first distillation
fermented wort is distilled to increase the alcohol steps. The microbial reaction could be the
concentration to 20-30% by volume, separating hydrolysis of agave pectin by the enzyme pectin
out the first fraction called ‘heads’, and the last methyl esterase produced by some
fraction, called ‘tails’. Composition of these microorganisms during the fermentation step, but
fractions varies depending on many factors this has not yet been demonstrated. Preliminary
including the yeast strain employed, wort results favor the first theory, but research is still
nutrient composition, fermentation time and needed on this matter because it is important to
distillation technique; but in general, heads are maintain the methanol concentration within the
rich in low boiling point compounds such as limits established by the official standard, which
acetaldehyde, ethyl acetate, methanol, 1- is 300 mg per 100 ml of anhydrous ethyl alcohol.
propanol, 2-propanol, 1-butanol, and 2-methyl Table 3 shows the specifications for tequila
propanol, which give a very pleasant flavor and according to the standard of quality from the
taste to tequila. Heads are normally mixed with Mexican Official Norm (Secofi, 1994).
the wort being distilled. The tails contain high
boiling point components such as isoamyl
alcohol, amyl alcohol, 2-furaldehyde, acetic acid Effluent disposal
and ethylactate, giving a strong taste and flavor
to the tequila; and when the concentration is The discharge from pot stills or distillation
above 0.5 mg/ml, the final product becomes columns is known as stillage, slops or vinasse;
unpleasant. This fraction is not used. and in a typical tequila distillery 7 to 10 liters of
In the second step, the liquid obtained from effluent are produced per liter of tequila at 110º
the first stage is re-distilled in a similar pot still proof. Tequila stillage has a biological oxygen
in order to obtain a final product that is 110° demand (BOD) of 25 to 60 g/L. In addition to
proof if it is sold in bulk (reducing transport the dissolved salts (mainly potassium, calcium,
costs) or 80° proof if it is to be bottled. Some and sulfate ions) and the low pH (<3.9) of the
companies obtain high proof tequila and dilute stillage, there are significant disposal or treatment
it with demineralized water or water purified by problems. A general solution to the disposal
reverse osmosis. problem does not exist because every factory
In continuous distillation systems, the has its own production process and is located
fermented wort enters the feed plate of the either in a city or near agave fields. As a result
column and flows downward, crossing a series of the difficulties of treating vinasses and due to
240 M. Cedeño Cruz
Alcohol, % bv at 20ºC 38.0 55.0 38.0 55.0 38.0 55.0 38.0 55.0
Dry extract, g/L 0 0.20 0 5.0 0 5.0 0 5.0
1
Using a Gas Chromatograph the value could be up to 500 mg/100 ml.
2
Minimum could be less than this value if the distillery could demonstrate the use of a technology to reduce methanol
concentration.
3
Using chemical analysis the maximum value could be up to 4 mg/100 ml.
their high concentrations of dissolved matter, a and Marchant, 1980). Stillage may be used as a
host of utilization schemes have been proposed. food supplement for cattle, but it has an
Some of the methods indicated below are under undesirable laxative effect on animals.
investigation and others are in use. Biological, aerobic, or anaerobic treatment offers
Recycling reduces the volume of waste to be a real means of disposal, but the cost is likely to
treated. Stillage can be recirculated, mixing 5 to be as high as the fermentation costs themselves
10% of the total volume of the waste obtained (Speece, 1983; Maiorella, 1983). Ultimately,
with clean water to substitute for the dilution tequila vinasses should be viewed as a raw
water used to prepare the initial wort. This can material rather than a waste, and a strategy
be carried out for a number of cycles, usually should be devised that maximizes economic and
no more than five, because the concentration of social benefits and reduces recovery costs.
dissolved salts increases and could affect the
fermentation process. Also, great care must be
taken with the final taste and flavor of the tequila Maturation
because some components present in stillage
could affect the organoleptic characteristics of Distillation is the final stage of tequila production
the final product. Currently, only one tequila if silver or white tequila is the desired product.
company uses this system. For rested (reposado) or aged (añejo) tequila,
Direct land application as irrigation water and maturation is carried out in 200 liter white oak
fertilizer in agave fields is under careful casks or in larger wood tanks. The time legally
evaluation to determine the optimum loading required is two months for rested tequila and 12
rates and the effects on the agave over the long months for aged tequila. Tequila is generally
time needed to reach maturity. Evaporation or matured for longer periods, depending on the
combustion of stillage could provide fertilizer characteristics each company desires for its
or potash, but the high cost of such a process is particular brand.
a serious limitation (Sheenan and Greenfield, As tequila ages in barrels, it is subject to
1980). The production of biomass and changes that will determine its final quality.
biochemicals including fodder yeast is a Thickness and quality of the stave, depth of the
possibility (Iñiguez et al., 1996), but the char, temperature and humidity in the barrelling
remaining liquor still has a high BOD (Quinn area, entry proof (40-110° proof), length of
Tequila production from agave 241
storage and number of cycles for the barrel (in million liters of tequila (at 40% alcohol by
Mexico barrels may be re-used several times) volume) in 1999 (Figure 16). After that, a
all affect the final taste and aroma of the tequila. shortage of agave occurred due to a combination
Fusel oil content decreases during maturation of environmental and market factors. A bacterial
owing to the adsorbant nature of the char, infection (Erwinia carotovora) and a fungus
smoothing the final product. Complex wood (Fusarium oxysporum), affected agave harvests,
constituents are extracted by the tequila, while the growing number of distilleries,
providing color and the particular taste. development of many new brands in the market,
Reactions among certain tequila compounds and an increase in tequila consumption in the
yield new components; and oxidation reactions domestic and export market affected demand.
change some of the original components in The drop in production began in 2000 with a
tequila and those extracted from the wood. As a reduction to 181.6 million liters and in 2002
result of all of these changes, the concentration declined to 141 million liters.
of acids, esters and aldehydes is increased, while Production of 100% agave tequila had the most
the concentration of fusel oils decreases as pronounced decrease due to the agave shortage,
tequila reposes in barrels. because 6.0 kg of agave are required to produce
After aging and dilution with demineralized 1 liter of tequila (at 55% abv) whereas other
water (if necessary) the color of the tequila may tequilas require only 3.0 kg of agave per liter.
be adjusted to the desired value by the addition Most tequila producers moved into production
of caramel. Alternatively, some companies blend of other tequila types and reduced amounts of
different batches of tequila to obtain a 100% agave tequila.
standardized final product. The outlook for the future appears promising
Government inspectors supervise the entire for tequila production, and a return to previous
aging process. Prior to bottling, tequila is filtered industry growth rates is expected by the end of
through cellulose filter pads or polypropylene 2003.
cartridges. Sometimes afterwards a pretreatment In the export market the scenario was very
with charcoal is used to eliminate turbidity. similar but the effect of the agave shortage lasted
through 2002 since the main export companies
sell tequila in bulk (Figure 17). Total export
Production statistics volume was at its lowest in 2001 (75.6 million
liters of tequila at 40% abv) with the beginning
Tequila production has grown since 1995 at an of recovery noted in 2002. Exports of 100%
average rate of more than 10% annually and agave tequila became almost stable at a volume
reached its maximum production of 190.6
200 Tequila100%
Tequila
Total
150
100
50
0
1995 1996 1997 1998 1999 2000 2001 2002
Figure 16. Tequila production in Mexico in millions of liters at 40% abv (CRT, 2002).
242 M. Cedeño Cruz
Tequila100%
10 0
Tequila
Total
80
60
40
20
0
1995 1996 1997 1998 1999 2000 2001 2002
Figure 17. Exports of tequila by type in million liters at 40% abv (CRT, 2002).
near 8 million liters. Again, it should be noted million liters of tequila at 40% abv), this has been
that this type of tequila can only be exported growing since 1995 when 6.5 million liters were
bottled and not in bulk. state-bottled (Figure 19).
Although the European and other markets are
growing, nearly 80% of all tequila exported
continues to go to the US market with volumes Future developments
that in 2002 represented over 69 million liters
of tequila at 40% abv, which is more than the 53 Research and future developments in the
million liters consumed in the domestic market production of tequila and agave cultivation can
(Figure 18). Finally, even though only 25% of take many directions, but the implementation of
the tequila is exported bottled during 2002 (22 any change must allow quality of the final
80
60
40
20
0
1995 1996 1997 1998 1999 2000 2001 2002
Figure 18. Exports of tequila by destination in million of liters at 40% alcohol in volume (CRT 2002).
Tequila production from agave 243
10 0
Bulk
Bottled
80
Total
60
40
20
0
1995 1996 1997 1998 1999 2000 2001 2002
Figure 19. Exports of tequila in millions of liters at 40% abv (CRT, 2002).
product to be maintained and provide substantial knowledge of distillation techniques, that tequila
improvement in the process. There are several acquired its present form. The agave plant, which
key areas where important developments could is often confused with cacti, is propagated
take place. Development of new varieties of through traditional methods or by means of plant
agave endowed with resistance to pests or cell culture. It is grown for 7 to 8 years before it
extremely dry environments, use of can be harvested and sent to the distillery.
micropropagation techniques, satellite remote The tequila production process comprises five
inspection to evaluate health status of agave stages. In the first step, agave is cooked to
plantations, higher inulin content, low wax hydrolyze the polymers present in the plant,
content in the leaf cuticle and superior growth mainly inulin, into fermentable sugars. In some
rates would all be of benefit. Mechanization factories this step is accomplished using stone
would improve aspects of cultivation and harvest ovens and in others it is carried out in autoclaves.
of agave. In addition, optimization of the The second stage is sugar extraction from cooked
cooking and fermentation steps would improve agave through milling; and the agave juice
yields and reduce the amount of waste, while obtained in this step can be mixed with sugars
yeast strain selection could improve ability to from other sources, normally sugar cane, if 100%
ferment musts with high concentrations of sugars. agave tequila is not desired. The third and most
Another area of opportunity is the use of isotope important stage is fermentation in which sugars
(H2 and O18) measurement in tequila to authenticate are transformed into ethanol and other
the beverage. Finally, low cost alternatives for compounds such as esters and organic acids.
waste (bagasse and stillage) treatment are These, along with other substances derived from
needed. the cooked agave, give the characteristic flavor
and taste to tequila. It is of great importance to
have a good yeast strain and nutritionally
Summary balanced wort for tequila production, as losses
can be as high as 35% of the total production if
Tequila is a beverage obtained from the inefficient yeast is used or nutrients are not
distillation of fermented juice from the agave present in the right proportions. In the fourth
plant (Agave tequilana. Weber var. azul). The stage, fermented wort is distilled, normally using
use of agave to produce drinks in Mexico dates pot stills or in some cases rectification columns,
from long ago when certain tribes used them to obtain the final product. At the end of the
for religious ceremonies. It was not until the distillation process white tequila is obtained.
arrival of the Spaniards, who brought Maturation, the last stage, in white oak barrels
244 M. Cedeño Cruz
is required for rested or aged tequila. The Diguet. L. 1902. Estudio sobre el maguey de
minimum maturation times are 2 and 12 months, tequila. Generalidades e historia. El Prog. Mex.
respectively, for rested and aged tequila as 9:424.
required by government regulations. At every Ehrler, W.H. 1967. Agave plant, efficient water
step of the production process, most companies user. USDA Agr. Res. 16(4):11.
employ several quality control analyses in order Engan, S. 1981. Beer composition: volatile
to ensure the quality of the product and the substances. In: Brewing Sciences, Vol. 2
efficiency of the process. Some major producers (J.R.A. Pollock, ed.) Academic Press, London,
of tequila are certified through an ISO-9000 p. 98.
standard. Tequila production is governed by the Estarrón, M., T. Martín del Campo and R. Cosío.
official norm NOM-006-SCFI-1994, which must 1999. Identificación de los componentes
be followed by all tequila producers to guarantee
volátiles que caracterizan la huella
a good quality final product.
cromatográfica distintiva de tequilas. Technical
Report for Tequila Herradura S.A.
García, A.I., L.A. García and M. Díaz. 1994.
Acknowledgement
Fusel alcohol production in beer fermentation
The author would like to express his gratitude processes. Process Biochem. 29:303-309.
to Tequila Herradura, S.A. de C.V., for its support GEA. 1992. La fertilización, La Jima. Boletin
in writing this chapter. Informativo, Gerencia de extensión agrícola.
Tequila Sauza, 7(50).
Geiger, E. and A. Piendl. 1976. Technological
References factors in the formation of acetaldehyde during
fermentation, MBAA Tech. Q. 13:51.
Agricultural Research Service. 1972. The Agave Goncalves de L. O. 1956. El Maguey y el
family in Sonora. Agriculture Handbook Pulque, Fondo dc Cultura Económica, México.
No.399, US Department of Agriculture, p. 195. Granados, S.D. 1985. Etnobotanica de los
Alonso, G.S., L. Rigal and A. Gaset. 1993. agaves dc las zonas áridas y semiáridas. In:
Valoración química de bagazo de agave de la Biología y Aprovechamiento Integral del
industria tequilera. Rev. Soc. Quim. Mex. Henequén y Otros Agaves, CICY, A.C.,
3(6):19. México, p. 127.
AOAC 1990. Distilled liquors. Official Methods Halffter, G. 1975. Plagas que afectan a las
of Analysis of the Association of Official distintas especies de agave cultivado en
Analytical Chemists (H. Kenneth, ed) Assoc- México. Secretaría de Agricultura y Ganadería,
iation of Official Analytical Chemists, México, D.F.
Arlington, Virginia, p. 690. Hammond, J.R.M. 1991. Brewer’s Yeast. In: The
Backman, G.E. 1944. A karyosystematic study Yeasts, Vol. 5. Yeast Technology (A.H. Rose
of the genus Agave. Am. J. Bot. 35:283. and J.S. Harrison, eds.) 2nd. ed. Academic
Berry, D.R. 1984. Physiology and microbiology Press, Redding, U.K.
of the malt whisky fermentation. In: Progress Iñiguez, C.G., I.J.A. Cuaron, G.P. Perez, M.M.
in Industrial Microbiology. (M.E. Bushell, ed.) De la Torre and P.I. Magaña. 1990.
Elsevier, Amsterdam, p. 189. Fermentation characteristics, digestibility and
Bottorff de B., V. 1971. A guide to tequila, performance of ensiled swine waste, wheat
mezcal and pulque. Minutae Mexicana, straw and cane molasses fed to sheep. Biol.
Mexico, D.F. Wastes 34(4):281-299.
Casado, C.G. and A. Heredia. 2001. Self- Iñiguez, C.G., G.Ma. de J. Franco and O.G.
association of plant wax components: a Lopez. 1996. Utilization of recovered solids
thermodynamic analysis. Biomacromolecules from tequila industry vinasse as fodder feed.
2:407-409. Bioresource Technology 55(2):151-155.
CRT. 2002. Consejo Regulador del Tequila. Iñiguez, C.G., S.E. Lange and R.M. Rowell.
Informe de Actividades. 2001. Utilization of byproducts from the
Tequila production from agave 245
Chapter 16
ROBERT PIGGOT
Alltech Inc., Nicholasville, Kentucky, USA
Rum defined
Rum differs from Scotch, bourbon or cognac in The two key points in both these definitions is
that it is produced in many countries. In fact that rum must be made with products derived
many countries import bulk rum, mature and from sugar cane (beet molasses products cannot
blend it then market it either locally or be called rum and have a very different flavor
internationally. The country where it is sold sets profile) and it must have taste and aroma
the definition of rum, which means that it is not associated with rum. This part of the definition
the producing country that defines the ingredients has been stretched in recent years with the very
and processes involved in making rum. light white rums. In some blended drinks the
In the US the BATF definition of rum is: spirit is referred to as ‘molasses spirit’ instead of
rum for this reason. These definitions may
“An alcoholic distillate from the fermented appear vague, which is due to the wide range of
juice of sugar cane, sugar cane syrup, sugar styles and production methods that go into
cane molasses, or other sugar cane by- producing quality rum. For this reason this
products, produced at less than 190° proof in chapter is intended to give an overall guide to
such manner that the distillate possesses the rum production and should be read in
taste, aroma and characteristics generally conjunction with the chapters on molasses as a
attributed to rum, and bottled at not less than feedstock and beverage alcohol distillation also
80° proof; and also includes mixtures solely in this volume.
of such distillates.”
alcohol is made from the cane grass. Rum can Dark rum was the only type of rum available
be loosely divided into three groups: dark, amber for many years. It was generally built on the
and white. White rum is produced by continuous image of British Navy Rum. As tastes changed
distillation. It is not aged, except where required it declined in popularity. The Bacardi revolution
by law. When aged, it is aged in old well-used of the 1960s, with its amber and light rums,
barrels and charcoal treated before bottling. It quality control and marketing changed the
has a light and delicate nose and flavor. Amber image of rum and helped make it the popular
rum is aged in wood for a short period, from six beverage it is today. Even quality dark rum,
months to two years. Amber rum may contain which has all the nuances of a fine cognac, is
some pot still rum for flavor. Dark rum is more making a strong comeback. Today the rum
fully bodied. It is aged in wood for longer industry employs 50,000 people in the Caribbean
periods, up to 12 years. basin and generates $260,000,000 in export
earnings.
History
Equipment and feedstocks
The beginnings of rum go back to the mid
1600s, possibly originating in Barbados, where EQUIPMENT
it was known as a ‘hot and hellish’ drink. It Rum is made with a wide variety of equipment,
carried nicknames such as ‘kill devil’ and both old and modern. Until recently, an original
‘rumbullion’, the latter meaning clamor or noise. wood-sided Coffey still was still in use in
Today rum is associated with warm tropical Guyana. The art of fine rum making is not in
islands, images of swashbuckling pirates and the the equipment; it is in understanding and
British Navy. The link between rum and the operating the equipment to produce the desired
British Navy runs deep: characteristics.
• Vice Admiral William Penn first gave it in
1655 after the capture of Jamaica, which did SUBSTRATES
not have brewing facilities. It gradually
replaced beer as it could be stored for longer Molasses
periods. The use of molasses as a feedstock for alcohol
• It became the official drink of the production has been covered in an earlier chapter
BritishNavy in 1731; and each sailor in this volume; therefore only a few comments
received daily a half pint (tot) of undiluted are necessary with specific reference to rum.
rum. The previous ration was one gallon of The source of the molasses can have a strong
beer or wine a day. In their days, Hawkins influence on the aromatic quality of rum. This
and Frobisher said they could cruise as long has been demonstrated in trials on making rum
as the beer lasted. from beet molasses which showed that it was
not possible to obtain the same characteristic
• In order to reduce drunkenness among the aromas as obtained from cane molasses (Arroyo,
crew, on August 21, 1740 Vice Admiral 1948). Arroyo (1941, 1942) reported that fresh
Vernon ordered the daily rum ration diluted blackstrap molasses with low viscosity, high total
with water. The admiral’s nickname, Grog, sugars, nitrogen, phosphorus and a low ash and
was given to the diluted mixture. gum content was preferable in the production
• After the battle of Trafalgar in 1805 the body of rums with desirable odors and tastes. Good
of Lord Nelson was preserved in rum, giving molasses for rum production should be low in
it the nickname ‘Nelson’s Blood’. ash and sulfur and high in fermentable sugars
and free amino nitrogen.
• The tot was reduced to 1/8 pint on January 1,
1851.
Cane juice
• July 30, 1970 ‘Black Tot Day’ was the last
day the traditional tot was given in the British Most rum is produced from blackstrap molasses,
Navy.
Production of heavy and light rums: fermentation and maturation 249
although cane juice is used in the French grown under sanitary conditions with a small
Caribbean islands. This spirit, known as quantity of sterile air and yeast food to establish
‘agricultural rum’ has a very high congener value healthy yeast growth. The solution is usually a
(over 2,250 ppm) (EU regulations). Using cane molasses solution of about 16°Brix. The volume
juice for rum production has both advantages of the final propagator should be about 10% of
and disadvantages. The main advantage is that the fermentor volume and have a cell count of
no additional processing is required. The cane at least 200 million cells/ml (Murtagh, 1999). In
is simply crushed and the juice fed directly into practice the yeast count in the propagation phase
the fermentors. Another advantage is that cane is often lower than this. It is important that the
juice does not have a high content of dissolved yeast in the propagator is still in the exponential
salts and therefore does not cause as much growth phase when added to the fermentor. If
scaling and blocking of distillation columns as kept too long, the yeast will enter the stationary
may occur using molasses. phase and bacteria present in the fermentor will
One of the main disadvantages of using cane flourish on the sugar present. Some rum
juice is that it normally contains only about 12- producers rely on the fermentation to start
16% w/w sugar, so that the alcohol content of spontaneously from naturally occurring flora.
the fermented material is limited to about 6-8% This procedure results in a relatively inefficient
v/v as compared to 10% v/v obtainable from fermentation with low yields and varying quality.
molasses. Another significant disadvantage is The high level of bacterial contamination in these
that cane juice cannot be stored in bulk. Cane fermentations results in many desirable
juice normally becomes heavily infected with congeners, especially acids and esters. Jamaica
bacteria and yeast in the crushing process. This in particular is renowned for its production of
contamination may significantly reduce alcohol high ester rums. Most of these are used as
yields and may cause it to start to ferment blending stock to add flavor to dark and amber
spontaneously and very rapidly. This means that rum.
the distillery must be located in close proximity Arroyo (1945) has shown that the best rum
to the cane mill. It also means that the cane juice yield and aroma were achieved with a
is only available during the cane-crushing yeast:bacteria ratio of 5:1. In order to create a
season, which generally does not exceed six non-yeast congener profile in a more controlled
months per year. Cane juice may also have cost manner, a pure culture of bacteria such as
disadvantages in that it is a primary, or at least Clostridium saccharobutyricum (its presence has
an intermediate product in the production of been shown to accelerate fermentation (Arroyo,
sugar. Sugar production is usually heavily 1945)) may be added after 6-12 hrs of the yeast
subsidized and so it is to the producer’s benefit fermentation. Usually the bacterial inoculum
to produce as much as possible. In contrast, amounts to about 2% of the fermentor capacity;
molasses is generally a by-product of sugar and the pH of the fermenting mash is adjusted
production, considered to be of lower value. upwards to about pH 5.5 before addition to give
Cane juice can be blended with molasses to more suitable conditions for the bacterial
improve the FAN content and reduce the amount propagation. The bacteria produce a mixture of
of ash. acids, predominantly butyric, together with
others such as acetic, propionic, and caproic
acids. These acids in turn react with the alcohol
Fermentation and yeast to produce desirable esters (Murtagh, 1999).
HEAVY RUM
FERMENTATION FOR LIGHT RUMS
Flavors present in the final product are produced
during fermentation. Distillation refines the In light rum production, the emphasis is
flavor, but does not create the base components. generally on maintaining clean, rapid
In order to get the desired flavor profile many fermentations to minimize development of
companies use their own yeast culture, which undesirable congeners and to maximize
they grow up from a slant. This yeast must be fermentation efficiency. Many distilleries use
250 R. Piggot
commercial distillers yeast. Good yeast for the (Figure 1). This still can be operated either
production of light rum should be tolerant of slowly with very low steam input or quickly with
high temperatures (most rum is made in the a high steam input. The still is made of copper
tropics and the fast fermentations make cooling and should be inspected on a regular basis as
especially difficult), fast (to use the sugars before the copper takes part in the chemical reactions
bacteria metabolize them thereby reducing during the distillation process. Wash at about 8%
yield), tolerant of high osmotic pressure and abv is pumped into the pot. The steam vaporizes
resistant to acid stress. The yeast selected should this mixture; and the vapor is approximately 50%
also provide a light congener profile. Unlike abv. Some of this condenses in the first (low
whole grain fermentations, it is possible to wine) retort. The vapor passing out of the low
recycle the yeast from molasses fermentations wine retort is enriched further to about 75% abv.
either by flocculation or centrifugation of the The same process is repeated in the second (high
dead wash. Traditionally recycled yeast has been wine) retort. The vapor leaving the high wine
‘acid washed’, i.e., reduced to a pH of about retort is about 85% alcohol. After the final
2.3 using phosphoric acid and added to the next strength drops below a fixed value, the still is
fermentor; or it may be treated with boiled out to feints. These may be used to charge
antimicrobials, or with about 15 ppm Cl from the retorts for future distillations, and the process
chlorine dioxide at a pH of about 3.5 (Murtagh, repeated.
1999). The objective of yeast recycling is to take
advantage of sugar that would otherwise be
TWO COLUMN DISTILLATION
needed for yeast growth and to ensure a very
high cell count in the fermentor inoculum. A 2-column still is basically the Coffey still.
Recycling of yeast is generally not While it does not yield a neutral spirit, it is
recommended. The yeast is in the stationary ‘cleaner’ than a pot distillation. Its main
phase, and therefore 30 times the amount of yeast advantage however is that it can be ‘tuned’ to
must be added in order to get the same rate of give the desired congener profile. This is done
alcohol production as fresh budding yeast. Also by deliberately overloading the rectifying
it has been shown that bacteria can become column to varying degrees meaning a greater
resistant to acid washing, so bacteria are also proportion of the congeners come over in the
getting recycled. It is recommended that when product. For example a light rum may have a
recycled yeast is used it is dumped and renewed fusel oil draw of 1000 mL/min and a heads draw
on a regular basis. of 600 mL/min, a medium rum may have a fusel
oil draw of mL/min and a heads draw of 400
mL/min and a heavier rum may have a fusel oil
Distillation draw of 300 mL/min and a heads draw of 200
mL/min. This gives the blender a full pallet to
The method of distillation used has a produce the desired blend.
considerable effect on the nature of the rum
product. Heavy rums are usually produced by
batch distillation or in a 2 column continuous PRODUCTION OF LIGHT (WHITE) RUM
still, while light rums are normally produced by
continuous distillation with a hydrofining The production of light beverages and
column. hydrofining is discussed in the chapter on
beverage distillation. Molasses has some
important differences from other feedstocks in
POT DISTILLATION that it has very little pectin to break down into
methanol. As such, it is generally not necessary
There are a wide variety of pot stills in use in to install a demethylizing column. The problems
the rum industry, which vary from the straight with mud and scaling are covered in the chapter
pot still to the pot still with a distillation head as covering molasses as a feedstock. It is important
described in the chapter on beverage alcohol that the wash column is designed to handle mud
distillation. One type of pot still common in the and scale. Sulfur is often a problem in molasses
Caribbean rum industry is the double retort fermentations. The distillation system must have
Production of heavy and light rums: fermentation and maturation 251
Condenser
Vapor
Feints Room
Liquid
level
Low wine
retort
Steam
Condensate
plenty of copper contact. A heads concentrating As mentioned previously, white rum is generally
system such as a barbet head or a separate heads not wood-aged. Amber and dark rums are aged
column may be needed in some cases. in wood barrels, most commonly once-used
bourbon barrels of American white oak. Some
of the barrels are used charred, and others are
Maturation and flavor developement de-charred before use. Flavor components such
as syringic acid, vanillin, syringaldehyde and
Esterification is very important in developing gallic acid all have their origin in the lignin
the flavor profile of rum. It is a chemical reaction material of the oak barrels (Lehtonen, 1982).
and so is affected by time, temperature and the Figure 2 compares the barrel aging components
components present. It takes place at a relatively in Scotch, cognac and dark rum. These can be
steady rate during the maturation of rum. affected by many things including the number
(Reazin, 1983) Some esters are also formed of times the barrel has been used, degree of
during fermentation. It is a reversible reaction toasting, depth of char and maturation
between an alcohol and an organic acid. For temperature. In some Asian countries local
example: woods are used for the barrels, imparting a
unique flavor to the spirit produced in that region.
Ethanol + Acetic acid → Ethyl acetate Rum, in contrast to whisky, is generally aged in
a warm climate. This changes the maturation
process in that some reactions speed up while
MATURATION others progress at the same rate as the cooler-
maturing spirits. As can been seen from the graph
Maturation of rum is a complex subject; and the in Figure 3, the greatest increase in color, tannins,
following is intended only as a brief description.
252 R. Piggot
6 Scotch
5 Dark Rum
Cognac
4
ppm 3
de
n
ol
ol
ol
id
id
id
co
no
co
li
nil
hy
en
en
iac
ac
ac
ac
aia
aia
ge
Va
lde
Ph
ua
llic
lic
gic
lp
u
Gu
gu
E
nil
ga
lg
Ga
rin
hy
l)
Va
hy
rin
Et
Sy
py
Et
Sy
p-
ro
p-
-P
(n
p-
Figure 2. Comparison of aging components in cognac, dark rum and Scotch whisky.
1.4
200
1.2
1.0
0.8
100
0.6
0.4
0.2
0 0.0
0 1 2 3 4 5 6 7 8 9 10 11 12
Years
acids and solids occurs in the first year while Arroyo, R. 1942. The manufacture of rum.
esters and aldehydes form at a steady rate Sugar 37(1), (5) (7).
thought the maturation process (Reazin, 1983). Arroyo, R. 1945. The Production of heavy
bodied rum. International Sugar J. 11(8).
Arroyo, R. 1948. The flavour of rum - recent
References chromatographic research. International Sugar
J. 50:210.
Arroyo, R. 1941. The manufacture of rum. Lehtonen, M. 1982. The chromatic determination
Sugar 36(12). of some organic trace compounds in alcoholic
Production of heavy and light rums: fermentation and maturation 253
Chapter 17
ROBERT PIGGOT
Alltech Inc., Nicholasville, Kentucky, USA
Introduction
Since at least the early 8th century men have been produce fine beverages with this type of
increasing the strength of their fermentations. distillation. The pot still is made of copper.
The first attempts were freezing the mash and Originally copper may have been chosen
removing the ice. Later distillation was because the metal was durable and easy to work
developed. Distillation is the separation of and a good conductor of heat. It has proved to
components by taking advantage of their relative be an important part of the distillation process.
volatilities. When a liquid is boiled its vapor will It is more recently that another attribute, its
be richer in its more volatile components. When ability to influence the flavor of the distillate,
this vapor is condensed, the resulting liquid will has been fully appreciated (Nicol, 1989). Copper
have a higher concentration of the more volatile aids in the removal of undesirable components,
component. This process can then be repeated. especially sulfur, which imparts an off flavor to
the distillate.
There are basically two ways to run a pot still.
Pot distillation In the first the wash is added to the pot, and
then steam is applied to the coils to start the liquid
Until the advent of practical continuous boiling. The collection of the alcohol enhanced
distillation by Stein in 1827 and Aeneas Coffey condensate is continued until the alcohol
in 1830, all distillations were carried out using remaining in the pot is not economical to
pot stills. Although simple, pot stills continue to remove. To prevent solids burning on the coils,
be used to produce the finest Scotch malt it is important that the liquid is always kept above
whisky, cognac and Irish whiskey. They have the steam coils. The spirit derived from this type
evolved from simple small retorts to unique of distillation is normally very rough as it
complex pieces of technology (Nicol, 1989). contains most of the congeners from the
Development took place in Scotland during the fermentation and it is often redistilled to remove
nineteenth century as each distiller tried to some of these congeners. The other way to
differentiate himself from the distiller next door. operate a pot still is to take three fractions from
Once set, however, the still design, even to dents, the distillation. The first spirit off, called the
and production methods are not changed foreshot or heads cut, is removed and stored
(Bathgate, 2003). Pot stills are also used in separately. The mid-run is the next spirit off, and
production of extracted flavor drinks such as gin. it is kept for further distillation. The last spirit
While the apparatus is simple it takes skill to off, called the tails cut, contains the heavy fusel
256 R. Piggot
Temperature
Cooling water out
Head
Sight Condenser
glass Lyne arm
Swan neck
Vacuum
breaker Spirit out
Charge
Manway Spirit safe
Cooling water in
Vapors To
feints
Shoulder Temperature
Spirit receiver
Low pressure
steam in
oils and is removed at the end of the distillation. and free from off odors or tastes. Wine used in
Distillation is continued (steam is normally brandy production should be as low as possible
increased at this time to speed the process time) in sulfur compounds. While it is obvious that
until all practical alcohol has been removed (pot different sugar sources, molasses, fruit or grain
temperature approximately 98°C) the distillation will produce different spirits, more subtle
is complete and the spent wash is discharged influences such as fermentation temperature will
from the base of the still. The mid-cut from the also affect the final product.
first distillation, called low wines, (20-25%
alcohol) is distilled again in similar fashion. Irish
pot distilled whiskey is distilled three times, HEADS CUT
cognac twice. The result is spirit containing 65-
75% alcohol that has the characteristics that the The wash strength and the cutoff points for the
distiller deems right for aging. This sounds like heads and tails cuts are only changed after very
a very simple process, but there are many factors serious consideration as such changes will
that can influence the characteristics of pot probably alter the flavor profile of the resulting
distilled spirit. spirit. The amount of heads removed at the start
of distillation will affect the volatile top note of
the distillate and remove the ‘low boilers’
WASH (compounds with low boiling points) such as
methanol and acetone from the product. The first
The wash must be from a fermentation that is cut has a role often overlooked, which is to
complete, free from infection and in which the remove the heavy oils and fatty acids in the swan
yeast has not been stressed or produced excess neck, lyne arm and condenser left over from the
fusel oils. The ingredients used to make up the previous distillation. The historic way to change
wash will add their own individual characteristics from the heads cut to the mid-cut in Scotland is
to the final distillate. Water should be soft, clear to perform a haze test, when it no longer turns
From pot stills to continuous stills: flavor modification by distillation 257
Water cooled
Temperature condenser
Vent
5 - 8 trays
or packed
column Recycle
Product
Temperature
Steam
Condensate
cloudy when mixed with water. These days the DISTILLATION RATE
change is based on time or volume measurement.
The amount of steam applied to a pot distillation
will affect the final spirit quality. When the
TAILS CUT desired final result is obtained, the staging and
amount of the steam addition should be kept
While some of the heavier alcohols are needed constant. Too much steam can result in foaming,
to give the spirit some body, too much results in liquid carryover, and very little internal reflux.
an oily, unpleasant character and may cause haze A light steam load can give the product a
problems when diluted to bottling strength. ‘stewed’ character. The large internal reflux will
produce a spirit lighter in character, but will
result in more time for chemical combinations
STILL DESIGN to occur during distillation. The resulting energy
losses raise the distillation cost.
The design of a still will affect the characteristics
of the product. Length and configuration of the
swan neck will influence the amount of reflux. CONTROL AND AUTOMATION
Another variation on the pot still is to put a head,
either packed or with cross flow trays, in the Traditionally, the changes from heads to mid-
vapor path (Figure 2). This allows a cleaner, cut and then to tails have been accomplished by
more consistent product. It is popular in small measuring spirit density after the condenser as
distilleries that cannot afford or justify a it flows through the ‘safe’. The safe is a device
continuous still. There are also some very large designed to let the distiller control the process
pot stills used by major beverage alcohol without actually removing a sample. Use of a
producers for flavor production. spirit safe allowed distillation to proceed without
the presence of customs and revenue personnel
258 R. Piggot
100
95
90
Temperature (°C)
Vapor composition
85
80
Liquid composition
75
70
65
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Mole fraction alcohol
100
Volume Weight
80
60
Percent
40
20
0
0.05 0.20 0.40 0.60 0.80 1.00
Mole fraction
who are very sensitive about the removal of spirit At the start of distillation the flow is valved to a
on which the excise duty has not been paid. container. When a level probe is activated by
Today many pot stills are automated. As can the distillate, flow switches to the product
be seen in Figures 3 and 4, if we know the receiver. A valve opens, dumping the container
temperature of either the liquid or vapor phase into the feints tank. When the temperature
we can know the composition. This is generally indicates that the mid-cut is finished, the flow is
the way that pot distillations are automated. returned to the feints tank until practically all
Since the feedstock can vary in strength and flow alcohol has been removed. This temperature is
can be erratic at the start of a distillation, it may about 98°C. Above that temperature, the steam
be better to take the heads cut on a volume basis. cost is worth more than the alcohol.
From pot stills to continuous stills: flavor modification by distillation 259
Analyser Rectifier
Vapor feed
High strength alcohol
Heated wash
Perforated
trays
Wash feed
Wash being
stripped of alcohol Reflux as spirit condenses, heating wash
Steam
Stillage
to disposal
FLAVORED SPIRIT PRODUCTION ON A where it is heated with the overhead vapors from
CONTINUOUS STILL the column. It passes through a tank that vents
the liberated carbon dioxide (CO 2 ) to the
As with the pot distillation, it is important that atmosphere and then into the column.
the fermented beer or wash is sound. Off odors Steam enters at the base of the column, either
or taste from poor quality feedstock or through a reboiler or by direct injection. The
contaminated fermentations can carry over into column below the feed tray (stripper section)
the final product. In the production of Canadian removes the alcohol. The stillage passing out
whisky and bourbon, the beer is distilled in a the base should be less than 0.07% alcohol. Since
single column that is really a stripping and it is a whole mash, it is important that the
condensing column stacked together. Some stripping section uses trays that are not easily
plants use a doubler or thumper (a sort of fouled by the grain solids or mineral buildup.
continuous pot still) after this column. The alcohol vapors pass through a solids
separation tray into the concentrating section.
Operation of a Canadian whisky flavor still This section concentrates the alcohol from 55
A Canadian whisky flavor still is illustrated in to 80% by volume. The draw strength has a great
Figure 6. The beer (normally 8 to 10% alcohol effect on the flavor profile of the spirit. Spirit
by volume, but it can be as high as 15%) is strength is controlled by a temperature reading
pumped from fermentation though a preheater at the top of the column. This controls either the
From pot stills to continuous stills: flavor modification by distillation 261
CO2 removal
Cooler
Copper demister pad
H2O
Vent
Vent Con.
Recycle to column
Temperature
control
H20
Product:
55-80% 20 -35
alcohol Stripping trays
Low pressure
steam
Reboiler
Condensate
Stillage to dryhouse
<50 ppm alcohol
amount of reflux or product draw. If temperature flavor in the final product, while the specifications
is below set point, then spirit is too strong and for good quality vodka demand no odor or taste.
recycle is reduced. If too high, then spirit is too Specifications will also indicate the amount of
weak and recycle is increased. congeners allowed to be present. These can be
Spirit straight from the still is raw. It is aged less than 1 ppm. Depending on the degree of
for a minimum of three years in small oak barrels. neutrality required, the rectifying section in a
beverage plant can consist of two, three or four
columns.
Production of light and neutral spirit
The product of neutral spirit in a continuous still SPIRIT PREPARATION FOR THE RECTIFYING
requires the removal of all components that have SYSTEM
odor or taste components over a set threshold.
Some products such as light rum require some Many whisky plants in North America also
262 R. Piggot
produce a neutral spirit. These plants generally difficult component to remove and is therefore
use the flavor column, or one of similar design, a good indication of how well the column is
as the base alcohol to feed the rectifying system. working. This cleaner low strength spirit is sent
The strength of this product is usually 70-80%. to the rectifying column.
Due to this low strength, the fusel oils have not
been removed. Some heads will be vented or
sent to a heads column. Rectification
In plants designed to just produce neutral spirit,
the spirit is often taken to a higher strength (over The rectifying column takes the low strength
95%). This allows the removal of some fusel oils spirit, removes the last traces of impurities, fusel
and heads, similar to the rectification column, oils and heads, and takes the strength up to the
before it is sent to the rectifying system. Another azeotropic point. One difference in the rectifying
possibility is the use of a Barbet head on the column is the large number of trays: over 60
column to concentrate and remove the heads trays is common. This enables a large amount
(Figure 7). of internal reflux inside the column, which helps
concentrate the congeners for better removal
(Figure 9). Fusel oils concentrate on the tray
Extractive distillation or hydrofining where alcohol concentration is 65%; propanol
concentrates on the tray with 80% alcohol. It is
The most common way to remove fusel oils is important that the column is kept stable and that
known as extractive distillation, or hydrofining. the draws are sufficient to prevent the column
The operating principle behind this column is becoming loaded with impurities. The cleaner
addition of water to the spirit. Ethanol is very product is removed about five trays from the
soluble in water while the fusel oils are not top of the column. While they concentrate at the
(Figure 8). The alcohol is added to the column top of the column where they are removed, it
about 2/3 of the way up. The water soluble should be noted that the heads must pass the
components (ethanol and methanol) will product draw tray to reach the top of the column.
combine with the water, forming a solution with It is therefore not possible to remove all the heads
a higher vapor pressure and hence move down from the product in the rectifying column. To
the column. The components that are less water improve alcohol recovery and to increase purity
soluble will rise to the top of the column where levels, other columns are usually added to the
they are removed. This column requires three system for demethylization, heads and fusel oil
components (alcohol feed, steam and water) to separation. In a well-designed and run extractive
be properly balanced. Generally at least eight distillation system, components other than
parts of water for each part of alcohol must be ethanol will be less than 1 ppm.
used. In practice much higher ratios (10 to 15)
are normally used for good separation.
In order to balance the steam, one must first FUSEL OIL AND HEADS COLUMNS
understand what is happening in the column.
The water coming from the top of the column These columns take the fusel oil and heads draws
will push the ethanol down the column, while from various parts of the process, concentrate
steam coming from the base of the column will them for removal and return cleaned spirit to
try to push it up the column. The result is that the process. They are generally small columns
the alcohol has a peak in concentration, known with low feed rates that have trays or are packed.
as the pinch point, about 16 trays from the base
of the column. Keeping this pinch in the correct
position and strength is a key to the proper DEMETHYLIZING COLUMN
balancing of this column. The concentrated fusel Certain feedstocks such as fruit and potatoes can
oil removed from the top of the column can be end up with very high concentrations of
sent to a fusel oil washer. Strength at the top of methanol that need to be reduced, or perhaps a
the column will be in the 60 to 68% range. As product very low in methanol is required. One
can be seen in Figure 9, propanol is the most way to do this, especially if methanol cannot be
From pot stills to continuous stills: flavor modification by distillation 263
Vent Vent
Condenser
Condenser Condenser
concentrator
36
Heads
30 Heads
Pre-stripper
60
Beer
feed
Heads
55
24
95° GL
Intermediate product
Surge
tank
Propanol draw
Beer stripper
Concentrator
28
Liquid
Vapor
Fusel oil
21
stripper
Bottom
Steam Steam
Figure 7. Concentrating neutral spirit: Barbet head on the column to concentrate and remove the heads.
10
Acetaldehyde
Ethyl acetate
Diacetyl
Acrolein
Propanol Methanol
1 Ethanol % (v)
0 20 40 60 80 100
Isobutanol
Isoamyl
0.1
Figure 8. Relative volatility of congeners and ethanol.
264 R. Piggot
Heads removed
from recycle
Heads
Product
Propanol
The dotted lines indicate the spread
of the peaks at higher congener Ethanol
concentrations.
Fusel oil
Increasing concentration
removed at the beginning of the process, is to column size increases they become increasingly
remove it after the rectifying column. Steam to complex and not cost effective.
this column is via a reboiler so as not to add
water to the system. The alcohol is added to the
column about 1/3 to 1/2 of the way up. The Energy efficiency
lighter methanol and any other heads present
are driven up the column and removed. The Distillation uses a lot of energy, however with
purified ethanol is removed at the base (Figure good design the energy requirements can be
10). reduced or recovered for other operations.
PACKED COLUMNS
REBOILERS
Packed columns are known as continuous
contact columns. They have an advantage of Reboilers are simply heat exchangers that take
continuous vapor-liquid contact. Because of this, a hot stream and use it to generate steam for
they may be an advantage when a small column another part of the process. This is illustrated in
needs an increase in capacity. However, they are the flavor column (Figure 10). The advantage
only used on clear, non-fouling tasks and as the of this is that the condensate can be returned to
From pot stills to continuous stills: flavor modification by distillation 265
Vent Vent
Condenser
Condenser
Vent
40 75 50
70
95°GL
ED Heads Heads
Intermediate product
heads 30
30
Demethylizer
Extractive distillation column
Rectifier
Surge
tank
Hot water
23 Propanol draw
20
15
Stripping
section
Steam Steam
Reboiler
Final product
Excess 96° GL
water Steam condensate
Figure 10. Three column rectification system used in neutral spirit production.
the boiler instead of creating extra liquid that molasses as a feedstock. The lower temperatures
must be disposed of in the stillage. in the stripping section reduce calcium fouling
of the trays.
PRESSURE DISTILLATION
References
Quality assurance Bathgate, G. 2003. History of the development
In order to be sure the correct quality product of whisky distillation. In: Whiskey Technology,
reaches the receiver, a system of predictive Production and Marketing (I. Russell, ed).
quality control must be implemented. This can Academic Press.
include such things as a check on the fermentor Murtagh, J. Feedstocks, fermentation and
for odor and taste before it is pumped to the still. distillation for production of heavy and light
Samples of the fusel oil and propanol draws on rums. In: Alcohol Textbook, 3rd ed. (K.A.
the rectifying column can be obtained to see if Jacques, T.P. Lyons and D.R. Kelsall, eds),
there is a buildup before it gets into the final Nottingham University Press, UK.
product. Whatever works in a particular system, Nicol, D. 1989. Batch distillation. In: The Science
the trick is to detect potential problems before and Technology of Whiskies, Longman
they become full-blown problems that affect Scientific and Technical.
final spirit quality.
While many countries including the EU have
specifications for neutral spirit (the US does not)
the companies using the spirit usually have their
own much tighter specifications that must be met
by the producer.
From liqueurs to ‘malternatives’: the art of flavoring and compounding alcohol 267
Chapter 18
Introduction
Using flavor in alcoholic beverages expands the sweetness, sourness, saltiness and bitterness.
variety of consistent, high quality products Additionally our mouths detect textures referred
available to the consumer. From high proof to as ‘mouthfeel’. In beverages these include
spirits that have a twist of fruit flavor or spice, viscosity and slickness often described as
to medium strength cordials, to low alcohol oiliness. Other sensations are drying and
premixed cocktails and ‘malternatives’, the trigeminal sensations or salivation stimulation.
inclusion of flavors gives the beverage producer Some tastes, especially bitterness, tend to linger
limitless flexibility in creating products to satisfy or are perceived less quickly.
the constantly changing beverage alcohol Most of what we describe as flavor perception
market. is actually aroma. Aroma perception occurs in
Historically, flavors were added to alcohol to the nose and can occur without tasting.
cover up poor quality and bad taste. Gin was Thousands of aromas are distinguishable. In
first used for medicinal purposes in the 17 th order for a chemical to be smelled, it must
century. Juniper berries were added prior to become airborne so that it can reach the air in
distillation to improve the taste. It was common the nose and be detected. Holding a beverage
practice to use poor quality corn and barley and in the mouth warms it and more of the aromatic
to add a flavor to make the product more compounds are released. Most aroma chemicals
palatable. In the 18 th century, production of are not soluble in water. Alcoholic beverages
flavored vodkas became popular. This was done have an advantage in that rich, intense flavors
to improve marketability of poor quality spirits. are produced because alcohol is a natural solvent
Today, flavors are often added to give a that is exceptionally volatile. The aroma
marketing edge to a competitive industry. compounds release more readily from an
According to Adams Handbook Advance, “The alcoholic beverage than from a soft drink.
US distilled spirits industry was up for the 5 th
consecutive year in 2002”. The number of Aesthetics of flavored beverages
flavored alternatives is credited with this increase.
Because people ‘drink with their eyes’, the
Flavor perception appearance of these lower proof products is
important. Approved artificial colors and
Our tongues sense the four basic tastes of caramel can produce virtually any shade of
268 A. Head and B. Timmons
color. Adding a homogenized clouding flavor beverages, are an extremely important segment
can simulate the inclusion of juice. The of the alcoholic beverage market. Coffee,
complexity of the interaction of these taste, smell chocolate, vanilla, almond, cherry, peach,
and visual sensations is what provides nuance orange, cinnamon, hazelnut, anise and even
and character to a beverage. gold leaf are just the beginnings of a list for
popular flavors used in this broad category.
Sweetened to between 12 and 40% sugar, these
Types of flavored beverages
products can be consumed by themselves, but
HIGH PROOF SPIRITS are frequently used as ingredients in cocktails,
in cooking or even as toppings for deserts.
Although originally considered an inferior
substitute for the distiller’s art or a mask for PREMIX COCKTAILS
product deficiencies, flavor-added spirits are
now considered to be the equal of unflavored Shelf-stable cream-type beverages can be
spirit. As previously mentioned, one of the produced using non-dairy cream substitutes. This
earliest examples of added flavor in high proof allows for a whole array of milkshake type
spirits was improvement of inferior spirit flavor products, with names such as mudslides,
by adding juniper berries during distillation. creamsicles and grasshoppers. Margaritas and
These were the earliest gins. Now, many gin similar premixed cocktails give the consumer
products are made by adding flavor from the the convenience of a complete spirits drink
refined oil of juniper berries, along with other without the inconvenience of buying multiple
essential oils, directly to high proof neutral ingredients and the hassle of complex and time
spirits. Refined essential oils are much more consuming preparation.
consistent than raw botanicals, which can be
affected by varying growing conditions. MALTERNATIVES
Blended American and Canadian whiskies use
flavors to provide a fuller, more balanced flavor ‘Malternatives’ and ‘Alcopops’ are beverages
profile while reducing the ingredient costs of that contain alcohol at the same level as beer.
more expensive spirits. American whisky blends Malternatives, which are produced in the US,
are made by combining a straight bourbon use as a base a malt product with an intentionally
whisky with neutral alcohol. A well-designed low flavor. Flavor and sweetener are added to
flavor can mimic the flavor compounds lost produce a drink similar to a cocktail from a
from the neutral spirit. Vanilla can be added to distilled spirit. Many are marketed using the
replace that which would have been extracted brand name of a spirit product, e.g. SMIRNOFF
from the oak barrel. Higher alcohols and other ICE®, BACARDI SILVER™, etc. This category
congeners can be added back in amounts and has been the most successful of recently
proportions that are more desirable than the introduced product lines. Among the marketing
original product. Oak flavor from oak wood is advantages these products have is unlimited
not permitted for flavoring American blended flexibility for product design. Virtually any type
whiskies. of cocktail can be developed in this base product.
Producers in the crowded vodka and rum These products can be sold at the same retail
market have found that flavor and a little sugar outlets as standard beer products, which is a
can add an ‘extra’ that sets a product apart. Citrus tremendous advantage over spirit-based
extracts and pepper or spice flavors are quite beverages. Crossover advertising can benefit the
popular in flavored vodkas, which have found spirit product, as well. Distilled spirits have
a new niche in trendy martini cocktails. Spiced traditionally not been advertised on television
rums, which are high in vanilla flavor, and in the US, however malternatives with the same
tropical rums, reminiscent of piña coladas, have brand name can be advertised (e.g. Schmirnoff
also become the base of many popular drinks. Ice). The production costs are much lower
because of the differential between beer ($18.00
CORDIALS AND LIQUEURS per 31 gallon barrel regardless of alcohol level),
wine and flavoring ($2.00 per gallon of alcohol)
Cordials, and other medium strength sweetened
From liqueurs to ‘malternatives’: the art of flavoring and compounding alcohol 269
and distilled spirits excise taxes ($27.00 per In addition to these classifications the US
gallon). Bureau of Alcohol, Tobacco and Firearms
Alcopops are similar products that are actually (BATF) has placed restrictions on the inclusion
made with spirits set to a comparable alcohol and labeling of certain ingredients. Artificial
level. Smirnoff Ice sold outside the US will flavor substances can be included in an alcoholic
contain vodka as the alcohol source in place of beverage product without being labeled
a malt base. artificially flavored if the BATF-approved flavor
contains 0.1% or less of an artificial ingredient.
There are also restrictions on the level of certain
Types of flavors ingredients, whether or not they are artificial.
ARTIFICIAL VS VATURAL
KOSHER
For the purposes of the Codex Alimentarius
“Natural flavors and ‘natural flavoring Foods, beverages, raw materials and processing
substances’ are preparations and single aids are kosher if they are produced in
substances, respectively, acceptable for human accordance with the Jewish dietary laws. There
consumption, obtained exclusively by physical are several rabbinical services that will, for a fee,
processes from vegetable and sometimes animal certify processes, equipment and ingredients as
raw materials either in their natural state or as kosher and allow their seal to be placed on
traditionally processed for human consumption.” product packaging. Although these certifications
‘Natural’ does not mean coming from the plant are useful for marketing purposes, Kosher does
that characteristically produces the flavor. The not mean more healthful, safer or superior in
classic example is benzaldehyde derived from quality to non-Kosher.
peach pits and used in almond or cherry flavors.
‘Natural’ does not mean more healthful, safer
or higher in quality than artificial. It is merely a BOTANICALS
description of the source.
‘Nature–identical flavoring substances’ are These are the starting point for most natural
“substances chemically isolated from aromatic flavors. Botanicals are parts of a plant that contain
raw materials or obtained synthetically. They aromatic chemicals that are desirable as flavoring
are chemically identical to substances present agents. All parts of plants have been used for
in natural products intended for human flavor. A few examples are juniper berries, hop
consumption, either processed or not, as defined flowers, leaves and stems, rose petals and hips,
above.” cinnamon leaf and bark, whole peppers and
Another group of ingredients called ‘nature- pepper corns, clove buds, angelica root and citrus
identical’ is comprised of “substances that are rinds. Non-plant derived natural flavoring agents
chemically identical to flavoring substances include beeswax, castoreum and fermentation
naturally present in vegetable or animal raw congeners.
materials that are not normally considered as
human food.”
The same ingredient can be available in both EXTRACTS
natural and nature-identical forms (i.e. natural
It is frequently useful to remove the aromatic
vanilla versus vanilla that is nature-identical).
compounds from the whole plant material prior
The natural version is typically considerably
to use. Essential oils, absolutes and concretes
more expensive and less concentrated. However,
are the fragrant chemicals found in the plant
there can be valuable reasons for selecting the
material. There are various extraction techniques.
natural form for a beverage flavor.
Expressed oils are squeezed out in presses.
‘Artificial flavoring substances’ are “those
Carbon dioxide, propylene glycol, steam and
substances which have not yet been identified
ethanol are commonly used for extraction of
in natural products intended for human
essential oils. Further purification can be
consumption, either processed or not.”
achieved by distillation into various fractions
based on boiling points.
270 A. Head and B. Timmons
the intended effect, can take time. Any changes provide tartness and simulate juice content. The
in the other parameters can put you back to the correct acid/sugar ratio is important for giving
drawing board with the flavor. the right taste and mouth feel to a drink. Coffee
Sweetness level is based on the beverage type and vanilla beverages require little or no acid
and flavor. As little as 1% sugar can smooth the addition. Cream drinks are incompatible with
roughness of a high proof spirit. Premixes acid, so the flavor selected for these products
blended with ice need the sugar level should not need tartness to enhance flavor. Malt,
sufficiently high to prevent the product which is fermented in an acidic pH range, is
becoming watered down after dilution. Fruit also not suitable for cream drinks.
flavored beverages may have different After the flavor profile is completed the
sweetness levels depending on the fruit flavor. aesthetic issues of color and cloud are
Sugar also adds viscosity to a beverage. Some addressed. Many citrus flavored drinks are
products such as eggnogs can be made more clouded to simulate juice content. Others such
viscous by adding gums such as xanthan. as a melon cordial would be shiny with only
Acids (usually citric and sometimes with some color added. Deep color suggests richness
sodium citrate) are used in most cocktails to and enhances the flavor experience.
272 A. Head and B. Timmons
1
Absolute alcohol by volume
2
Alltech Inc.
All $13.50 less any credit for wine and flavor content. $2.14 (at 80 proof)
Conclusions References
The beverage alcohol industry is increasingly Code of regulations- http://www.ttb.gov/
competitive. Producers are in a continuous regulations/27cfr5.html
struggle to increase profits and market share and TTB. 2003. Tobacco Tax and Trade Bureau,
consumers are looking for the latest trend and BATF, Dept. of the Treasury. www.ttb.gov.
more convenience. The art and practice of 2003 Adams Handbook Advance, Adams
flavoring beverage alcohol provides short Beverage Group, Norwalk, CT
product development times for a wide variety
of products, improves consistency and quality
of products and can reduce tax liability and
improve profits.
Production of American whiskies: bourbon, corn, rye and Tennessee 275
Chapter 19
RON RALPH
Ron Ralph & Associates Inc., Louisville, Kentucky, USA
1787 indictment of James Garrad (later a sold ‘from the barrel’, and quality standards were
Kentucky governor) and two others by a almost nonexistent. It was not until the end of
Bourbon County grand jury for retailing liquor the century that consumers could purchase
without a license. The only certainty in any of whisky sold in a corked and sealed bottle. Old
the lore is that Kentucky has a county named Forester was the first product with ‘guaranteed
Bourbon and produces a whisky by the same quality’ put on the label. Old Heritage was the
name (Connelley and Coulter, 1922). first bourbon with a strip stamp over the cork,
thereby becoming the first ‘bottled-in-bond’
bourbon.
ESSENTIAL TRADITIONS
ground, with the hammer mill being the most atmospheric batch cooking is used. Bourbon,
common type of processing; however some rye, wheat, Tennessee and corn whisky are
roller and attrition mills are still in use. The mashed using batch cookers. Only the ‘blend’
milling (Figure 1) is checked for grind by a sieve or ‘light’ whisky producers use continuous
analysis. A typical sieve analysis for grains in a cookers. Pressure cooking is usually done at
bourbon mash bill is shown in Table 2. 124 o C while atmospheric cooks are done at
100oC. Cooking time varies from 15 minutes to
Table 2. Typical sieve analysis for grains in a bourbon 1 hr. Conversion time and temperature are very
mash bill. consistent among distilleries. Malt is never
subjected to temperatures greater than 64oC; and
US Sieve # Corn Rye Malt Wheat conversion time is usually less than 25 minutes
16 15 22 2 20
to minimize contamination. All distillers use
20 21 25 8 26 backset (centrifuged or screened stillage from
30 17 13 14 12 the base of the still), but the quantity of backset
40 13 9 16 8 will vary based upon the beer gallonage (gallons
50 10 7 13 6 of water per 56 lb distillers bushel of grain) to
60 3 2 8 2 be used. American whiskies have beer
Through 60 21 22 34 22 gallonages in the 30-40 gallon range. High
energy costs for by-product recovery have
encouraged some distillers to use lower beer
Mashing gallonage ratios for spirits. However, bourbon
Mashing techniques vary considerably, but the and Tennessee whisky producers continue to use
major difference is whether pressure or 30-40 gallon beers. The cooling of cooked mash
280 R. Ralph
Cyclone
Grain
cleaner
Grain Meal
storage holding
bins bins
Meal
scale bin
Hammer mill
To cooker
Grain
receiving pit
As required
2 x 15 ml
test tubes
2 x 400 ml
Erlenmeyer
flasks
4 x 3 liter
Erlenmeyer
flasks
2000 gallon
yeast tub
200 gallon
dona tank
To fermentor
and let the fermentors work up to 31-32oC. All from the fermentor to the beer well. The beer
of this is controlled by mash cooker cooling, well is a holding tank for the fermented beer,
addition of cold water and weather factors. such that a continuous feed to the beer still can
Contamination is controlled by cleaning be maintained. Beer wells are usually 1-1.5
fermentors, ensuring no mash pockets in pipes times the size of a fermentor. They also have
and regular steaming of fermentors. continuous agitation to prevent solid grain
Both traditional and modern distillers ferment particles from settling to the bottom of the vessel.
their beers for at least 72 hrs, and some for as All American whisky producers use a continuous
long as 96-120 hrs. Three and five day still (Figure 3), though some have a second
fermentations are the norm. During these distillation ‘doubler’ or ‘thumper’ (Figure 4).
periods the Balling will drop to 0.0 and the pH The basic difference between a doubler and a
from 5.0 to 3.8 while the alcohol concentration thumper is whether the unit is operated with a
rises to 8-10%. All the changes that happen liquid level (doubler) or essentially dry
during fermentation are checked daily by (thumper). Both the doubler and the thumper
performing ‘beer chemistry’. Balling, pH, acids, provide a second distillation.
and fermentor temperature are monitored and The beer is pumped into the upper section of
recorded daily. The pH is the main indicator of the first continuous column, the beer still, six to
contamination and potential fermentation ten plates from the top. Live steam is introduced
problems and is regularly measured by at the bottom. Beer stripping plates 1-18 have
traditional and modern distillers (Table 3). perforations, a downcomer from above and a
dish on the plate below to hold the beer liquid at
a set level so the plates are never dry, as the
DISTILLATION beer moves back and forth across each plate.
The steam passing up and through the
Upon completion of the fermentation process, perforations, controlled by pressure or flow rate,
the beer with 8.0-10.0% alcohol is transferred strips the lighter, more volatile alcohol from the
Table 3. Typical analysis of beers from bourbon, rye and corn whisky production.
Bourbon Rye Corn
Set sample
Balling 13.4 13.4 12.3
Titratable acidity 4.5 3.6 2.8
pH 4.5 5.0 5.2
Temperature, °C 27.0 24.0 27.0
24 hr sample
Balling 2.6 4.0 3.6
Titratable acidity 5.1 4.6 4.2
pH 4.2 4.4 4.3
Temperature, °C 30.0 31.0 29.0
48 hr sample
Balling 2.4 3.6 1.0
Titratable acidity 7.8 7.5 6.1
pH 3.8 3.9 3.8
Temperature, (°C) 30.0 30.0 30.0
Drop sample
Balling 0.4 1.5 -0.4
Titratable acidity 8.2 7.9 7.1
pH 3.8 3.9 3.8
Temperature, °C 30.0 30.0 30.0
Alcohol, % by volume 6.73 5.8 6.8
Residual carbohydrates, % 8.0 8.4 4.2
Residual carbohydrates, % maltose 0.73 0.6 0.46
Production of American whiskies: bourbon, corn, rye and Tennessee 283
condenser
Vent
Beer Dephlegmator
pre-heater
Concentrating
Section
plates
Whisky
cooler
Rotameter
Beer feed
Low
Whisky
18 sieve plates
wine
Beer stripping
tank tank
section
Steam
To Maturation
Vent condenser
High wine
Whisky
(4-8 bubble cap
Concentrating
tank
section
plates)
Whisky
cooler
(Cistern room)
Rotameter
Low
Beer feed
wine
Whisky
cooler
tank
tank
(18 sieve plates)
Beer stripping
First Second
section
distillation distillation
Doubler
Steam
New charred
white oak
barrels
Stillage
Steam
Figure 4. Bourbon whisky distillation system, including doubler.
284 R. Ralph
water/grain mixture on the plates. When this centrifuged to prevent solids accumulation in the
alcohol gets to the top 4-8 bubble cap plates it is cooker and fermentors. Nearly all distillers have
concentrated to about 100o proof (50o GL). As a dryer house, though a couple of traditional
the vapors continue up into the beer preheater distillers continue to sell their ‘slop’ to nearby
(a heat exchanger to heat beer going into the farmers. For a while some, like the Jack Daniel
column), some alcohol is condensed and Distillery, fed the stillage to cattle in wet-feeding
refluxed to the top of the beer column. The rest operations; but environmental restrictions have
flows to the doubler or thumper as vapor. From generally eliminated such practices. Also,
either of these pot still-type chambers, the vapor modern dryer house operations have become
goes to the condenser where the product is very profitable and the profit helps lower the
drawn off at 130-140 o proof (65-70 o GL). cost of making whisky.
Bourbon cannot be distilled above 160 o proof
(80o GL). If it comes off at more than 160o proof,
it must be called ‘light’ whisky. The stillage from COOPERAGE AND MATURATION
the base of the beer still is pumped to the dryer
house to be processed. The ‘high wine’ from Cooperage and maturation are the processing
the stills is pumped to the cistern room where it factors that distinguish bourbon and American
is held, tested and reduced to barrelling proof whisky from the other whiskies of the world.
(Table 4). Only bourbon whisky regulations require that it
be matured in a new charred, white oak barrel.
Other whiskies of the world may require the use
BY-PRODUCT RECOVERY of small wooden barrels, but no other whisky
but bourbon goes through the care and expense
The by-product recovery does not usually of requiring small white oak barrels to be freshly
receive the same care that the other processes charred.
demand. The only essential for this process is The making of the bourbon whisky barrel is a
ensuring that the backset stays hot, around 99oC, very traditional, but exact science and craft.
so that it is absolutely sterile when used in the Staves and heading are quarter-sawed from
mash tubs, yeast tubs and fermentors. (Most mature, white oak timber. Actually, some
distillers use 20-30% backset in their total physical variance exists within a single tree, but
process.) Furthermore, the hotter the backset no more than between trees. After being quarter-
remains, the greater the savings in energy costs sawed with the medullary rays not less than 45
during the cooking process. The stillage from degrees to the stave surface, the staves and
the base of the beer still has about 7-10% total heading are air-dried in a stave yard. The more
solids. When used for backset it is screened or traditional distilleries require wood to be air-
Bourbon Rye
dried at least one year. The modern distillery from the alcohol via oxidation. The components
usually has no air-drying specifications and coming from the wood are tannins, sugars,
allows its barrels to be produced from wood that glycerol and fructose. The hemicellulose in the
has been air-dried for six months or less. All of wood appears to be the source of the sugars
the wood is kiln-dried at the cooperage, with found in aged American whiskies.
staves and heading dried to 12 and 10% The depth of charring and ‘toast level’ in the
moisture, respectively. The kiln-drying is barrel determines the color of the whisky. Color
essential to prepare the wood for the planing, formation is almost instantaneous when the
milling, edging and joining operations that distillate is put into the charred barrel, with 25-
cannot be done on wet wood. More important, 30% of the color formed in the first six months.
however, is that proper drying (air-drying a year Some color development occurs each year of
followed by kiln-drying), makes the wood maturation until the whisky is dumped from the
chemistry satisfactory for flavoring the whisky barrel. The final product is then filtered, its proof
during maturation. reduced with demineralized water and bottled.
As the whisky goes through maturation (3-4 Compared to the sweet rums, ‘breathless vodka’,
years for the modern distillers and 4-8 years for fruity gins, light Canadians and Scotches, the
the traditional distillers), two distinct types of American whiskies have flavor and bravado with
reactions occur: reactions between the distillate a balance that is pleasant to the taste. American
components (regardless of the barrel); and whisky produced and matured as described has
reactions that occur when the distillate extracts a big, pungent aroma that leaves no doubt that
chemical compounds from the wood. A major it is bourbon or Tennessee whisky.
factor differing among the distillers is the Though the modern and traditional distillers
warehouse environment. Most modern distillers have different levels of technology, they both
heat their warehouses in winter. One particular use the same basic processes for making
distiller even controls the heat cycles to ensure bourbon or other American whiskies. Individual
constant aging. The traditional distillery seldom plant nuances produce different flavors, but they
has heated warehouses and depends strictly on all have an American whisky bouquet and taste.
Mother Nature to determine the number and
range of its heating and cooling cycles.
Whether the heat cycle is natural or forced, References
the greatest rate of change or formation of
congeners occurs in the first 12-16 months. Only
ester formation occurs at a fairly constant rate. Carson, G. 1963. The social history of bourbon.
Proof increases at a fairly constant rate of 4-5% Dodd, Mead & Company, New York.
each year of aging. Other specific changes Connelly, W.E., and E.M. Coulter. 1922. History
occuring when the distillate reacts with the of Kentucky, Vol. II. The American History
charred wood are: a) aldehyde formation, Society, Chicago and New York.
specifically acetaldehyde, which comes from the Lyons, T.P. 1981. Gasohol, A Step to Energy
alcohol via oxidation, b) acetic acid formation Independence. Alltech Technical Publications,
with greatest activity in the first year of Nicholasville, Kentucky.
maturation, and c) ester formation (ethyl acetate)
Contamination and hygiene
Bacterial contamination and control in ethanol production 287
Chapter 20
N.V. NARENDRANATH
North American Biosciences Center, Alltech Inc., Nicholasville, Kentucky, USA
Diagnostic No acetic Acetic acid produced Elongated yeast, Cocci Cocci, Cocci, Rods, often in
characteristics acid production from ethanol multiple shapes some tetrads chains chains, no spores
Appearance No distinguishing Yellow zone, Yellow zone No distinguishing Yellow zone Yellow zone Yellow zone
on LMDA characteristics (on green underside characteristics
sulfite media colonies are
black on underside)
Distinctive Requires Does not Slowly Produces Does not grow Produces CO2
physiology simple sugars further oxidize oxidizes acetic acid in fermentor
acetic acid acetic acid
Present name Zymomonas Gluconobacter Acetobacter Dekkera Micrococcus Pediococcus Leuconostoc Lactobacillus
acetic acid from sugar substrates and does not from 2 to 53°C, with an optimum temperature
further oxidize the acetic acid produced. generally between 30 and 40°C. Optimal pH for
Acetic acid bacteria are rod-shaped cells and growth is 5.5-6.0. Growth generally occurs at
are unable to survive and grow in the absence pH 5.0 or less, but growth rate is significantly
of oxygen. Therefore, these organisms are not reduced. Under optimal growth conditions, the
a threat in the fermentor where the conditions doubling time for lactic bacteria is much quicker
are anaerobic. However, acetic acid bacteria can than that of yeast.
occur in the yeast propagator or yeast Metabolically, LAB possess efficient
conditioning tank, which is subjected to high carbohydrate fermentation pathways. The main
agitation and aeration to promote yeast growth. fermentation pathways for glucose are the
Once the inoculum is pumped to the fermentor Embden-Meyerhof pathway, converting 1 mole
and an anaerobic system is established, acetic of glucose to 2 moles of lactic acid (homolactic
acid bacteria will usually die; however sufficient fermentation) and the 6-phosphogluconate
acetic acid to slow or inhibit yeast growth may pathway, resulting in 1 mole of CO 2, 1 mole
already be present. It is easier to control these ethanol (or acetic acid) and 1 mole lactic acid
bacteria in batch propagation systems where the (heterolactic fermentation). The homoferment-
tank is emptied, cleaned and sanitized between ative LAB produce virtually a single
batches. fermentation product, lactic acid, whereas the
other LAB, called heterofermentative, produce
other products, mainly ethanol (or acetic acid)
LACTIC ACID BACTERIA and CO2 as well as lactic acid. The abbreviated
pathways for the fermentation of glucose by
Lactic acid bacteria (LAB) are Gram-positive, homo- and heterofermentative organisms are
catalase negative, microaerophilic or shown in Figure 2. The differences observed in
aerotolerant anaerobes. They are either rod- the fermentation products are determined by the
shaped or cocci-shaped cells that produce lactic presence or absence of the enzyme aldolase, one
acid as a major end product of carbohydrate of the key enzymes in glycolysis.
metabolism. LAB have complex nutritional Lacking the aldolase enzyme, hetero-
requirements (Kandler and Weiss, 1986). These fermenters instead oxidize glucose-6-phosphate
bacteria can grow in a wide range of temperatures, to 6-phosphogluconate and then decarboxylate
Triose-3-phosphate 6-phosphogluconate
Heterofermentation
Homofermentation
Lactic acid
Figure 2. The fermentation of glucose in homofermentative and heterofermentative lactic acid bacteria.
290 N.V. Narendranath
this to pentose phosphate, which is then broken increases) of the bacteria occurred. Other
down to triose phosphate and acetylphosphate scientists, however, have reported that when
by means of the enzyme phosphoketolase. bacterial numbers exceeded 108 CFU/ml at 30
Whereas the hetero-fermentative lactobacilli hrs of fermentation, alcohol loss was
possess ketolase but no aldolase, homo- approximately 5% (Barbour and Priest, 1988;
fermentative species possess aldolase but no Dolan, 1979). According to Makanjuola et al.
phosphoketolase. Obligate homofermenters are (1992), reduced ethanol yields, lower yeast
thus unable to ferment pentoses, which are numbers, reduced carbohydrate utilization, and
broken down by the heterofermenters via an increase in acidity were all caused by the
phosphoketolase, yielding equimolar amounts build up of lactic acid produced by lactobacilli.
of lactic acid and acetic acid. However, one They found that a bacterial count of 4.5 × 108
group of homofermentative lactobacilli possesses CFU/ml at 30 hrs resulted in a 17% reduction in
an inducible phospho-ketolase with pentoses ethanol yield due to a stuck fermentation.
acting as inducers. They are thus able to ferment Initial bacterial contamination of mash with
pentoses upon adaptation to lactic acid and acetic approximately 107 CFU/ml led to as much as
acid, while hexoses are homofermentatively 0.6-1% v/v reduction in ethanol depending on
metabolized. Therefore, these lactic bacteria are the strain of bacteria (Narendranath et al., 1997).
called facultative heterofermenters. Table 1 These authors were the first to report that both
shows some examples of the different classes final lactic acid concentrations and decreases in
of lactic acid bacteria. Isolates of LAB from ethanol yields at the end of fermentation were
distilleries are well-adapted to the conditions directly correlated with initial numbers of viable
existing in such fermentations (Bryan-Jones, bacteria in mash (Figure 3). Apart from the
1975). In fact, enumeration of bacteria in many diversion of glucose for growth, the production
distilleries is often limited to the detection of of end products such as lactic and acetic acid,
lactic acid bacteria because aerobes and and a suspected competition with yeast cells for
facultative anaerobes with little pH tolerance are essential growth factors in the fermenting
not considered serious threats to product quality medium are the major reasons for the reductions
or production efficiency. in yeast growth and final ethanol yield when
lactic bacteria are present.
Type Examples
1.8 0.8
0.7
1.2
0.5
0.9 0.4
0.3
0.6
0.2
0.3
0.1
0.0 0.0
4 6 8 9 10 8 10
10 10 5 10 10 7 10 10 10 10
4
10 5 10
6
10 7 10 10
9
10
Initial bacteria in mash (CFU/ml) Initial bacteria in mash (CFU/ml)
Figure 3. Effect of the initial numbers of lactobacilli on the final lactic acid concentration and on the reduction in final ethanol
concentration compared to the control with no bacterial inoculation. (Adapted from Narendranath et al., 1997).
80% reduction in yeast cell mass occurred at generally more toxic to microorganisms at low
concentrations of 7.5 g and 38 g/L, respectively pH, it is assumed that the antimicrobial activity
(Maiorella et al., 1983). The effects of these of these acids is the result of an increased
acids on the specific growth rate of yeast are proportion of undissociated molecules (Salmond
detailed in the chapter on continuous et al., 1984). Therefore, acetic acid (pKa = 4.74)
fermentation in this volume. The minimum with a higher pKa value is more toxic to yeast
inhibitory concentration (MIC) of acetic acid for than lactic acid (pKa = 3.86) at any given pH of
yeast growth is 0.6% w/v (100 mM) and that of the medium. Acetic acid has between two and
lactic acid is 2.5% w/v (278 mM). However, four times more molecules in the undissociated
acetic acid at concentrations as low as 0.05-0.1% form over a pH range between 4.0 and 4.6
w/v and lactic acid at concentrations of 0.2-0.8% compared to lactic acid (Lindgren and
w/v begin to stress yeast as seen by decreased Dobrogosz, 1990). Undissociated weak acids
growth rates and decreased rates of glucose diffuse passively into the microbial cell until
consumption (Narendranath et al., 2001a). equilibrium is established across the membrane.
The inhibitory effect of weak organic acids As molecules enter the cytoplasm, they dissociate
such as acetic acid and lactic acid on yeast at the higher intracellular pH. The protons
growth depends on the pH of the medium, the liberated are either pumped out of the cell in
dissociation constant of the acid, and its molar exchange for cations or are neutralized by the
concentration. In solution, a weak acid exists in buffering capacity of the cytoplasm (Booth and
a pH-dependent equilibrium between dissociated Kroll, 1989). Figure 4 illustrates how the anions
and undissociated states, as described by the (A-) and undissociated acids (HA) are distributed
Henderson-Hasselbach equation (pH = pK a + inside and outside the yeast cell for a given
log[A - ]/[HA] where A - and HA are the concentration of acetic acid and lactic acid at a
dissociated and undissociated species, particular pH of the medium. However,
respectively). This equation indicates that at a Narendranath et al. (2001b) found that acetic and
pH above its pKa value, more than 50% of the lactic acids inhibit yeast growth by different
acid is dissociated and that the concentration of mechanisms.
undissociated acid increases logarithmically as Therefore, the two major reasons for reduction
the pH declines. Since organic acids are in ethanol yield due to lactobacilli contamination
292 N.V. Narendranath
ATP
pH outside
cell: 2.5 CH3COOH CH3COO- + H+
Inside the cell 82.05% is dissociated!
pH in Equals 181.8 mM or 1.1% acetic acid!
cell: 5.4
CH3CHOHCOOH CH3CHOHCOO- + H+
Plasma
Inside the cell 97.2% is dissociated! ATP
membrane
Equals 1.33 M or 12% lactic acid! H+
Only undissociated acid
can enter the cell
Figure 4. Illustration of the concentrations of anions and undissociated acids that would be present in the medium and inside the
cell (based on the pHout and pHin) when 40 mM of either acetic acid or lactic acid is present. The concentrations were calculated
based on the Henderson-Hasselbach equation. The concentrations differ based on the molar concentration of the acids in the
medium and the pHout and pHin values. The concentrations indicated do not remain steady because there is constant pumping out
of the excess protons (by H+-ATPase) to raise the intracellular pH, resulting in further penetration of weak acid molecules into the
cell that re-acidify the cytoplasm. In most situations, the pH of the medium also falls, affecting results. The concentrations of
anions could reach 181.8 mM for acetate and 1.33 M for lactate under the stated conditions when 40 mM acid is added to the
medium (adapted from Narendranath et al., 2001b).
are: 1) Glucose that could be available for yeast processes. These bacteria must be removed as
to produce ethanol is taken up by these bacteria, quickly as possible. The methods used in the
and 2) the end products of metabolism, acetic ethanol industry to control contaminant bacteria
acid and lactic acid, inhibit yeast growth and include stringent cleaning and sanitation, acid
concomitantly ethanol production. washing of yeast destined for reuse (in case of
As little as 1% decrease in ethanol yield is breweries), adjustment of mash pH, and the use
highly significant to distillers of fuel alcohol of antibiotics during fermentation. The
(Makanjuola et al., 1992). In large plants with method(s) used depends to a large degree on
outputs of 400 million to 1100 million liters of the end use of the alcohol.
ethanol per year, such a decrease would reduce
income by 1 to $3 million USD annually.
USE OF ANTIBIOTICS
There are different kinds of antibiotics and they strains is a global issue, and one that ethanol
can be grouped according to their mechanism producers must understand because of DDGS
of action against bacteria. use in food animal diets. Penicillin G is
Penicillin G has been widely studied (Day et inactivated during the alcohol fermentation
al., 1954) and used in the alcohol production (Islam et al., 1999). Inactivation of penicillin G
industry during fermentation since the 1950s. is significant at 35°C. At pH 4.8, the biological
Penicillin G is primarily active against Gram- half-life of penicillin G is 24 hrs at 25°C and 4
positive bacteria. The extensive use of this hrs at 35°C, whereas at pH 3.8, it is <4 hrs for
antibiotic, which acts by inhibiting bacterial cell both temperatures. Moreover, any penicillin G
wall synthesis, has led to the emergence of still active at the end of fermentation is definitely
resistant microflora. For this reason and due to destroyed in the still since penicillin G
the instability of penicillin G pH levels below 5, decomposes rapidly at temperatures over 52°C
various other antimicrobials have been (Kheirolomoom et al., 1999). In contrast,
introduced or investigated for application virginiamycin is not significantly altered during
including virginiamycin (Hynes et al., 1997), alcohol fermentation after 72 hrs at 35°C (Islam
streptomycin, tetracycline (Aquarone, 1960; Day et al., 1999). Research has shown that using
et al., 1954) and monensin (Stroppa et al., 2000). over 2 ppm virginiamycin to control bacteria
Tetracycline is a broad spectrum antibiotic, suppresses the rate of fermentation. Upon
where as penicillin V, monensin, and distillation (100°C for 30 min) virginiamycin
virginiamycin are more active against Gram- activity is reduced to 2.6% of the original
positive bacteria. The other antibiotics that have (Hamdy et al., 1996). Residues can therefore be
been tried such as streptomycin and polymixin present in DDGS; which is a potential problem
are more active against Gram-negative bacteria. in markets where this antibiotic has been banned
The typical usage rate for these antibiotics in in animal feeds.
the fermentation is 2-5 ppm. The mechanisms
of action and the spectrum of these antibiotics
are summarized in Table 2. SYNERGY: ANTIBIOTIC COMBINATIONS
Although many antibiotics have been tested
at a laboratory scale, the most widely used in Using single antibiotics such as penicillin G or
the ethanol industry are penicillin G and virginiamycin repeatedly can lead to
virginiamycin. Antibiotic residues and development of resistance in microorganisms.
establishment of antibiotic-resistant bacterial Figure 5 shows the development of resistant
Table 2. List of some of the antibiotics used in the ethanol industry, their modes of action and activity spectrum.
Antibiotic Mechanism Bactericidal/static Spectrum
1a. Penicillin G Inhibits cell wall synthesis Bactericidal Gram(+) bacteria
1b. Penicillin V*
2. Bacitracin Affects cell wall Bactericidal Gram(+) bacteria
3. Tetracycline Protein synthesis inhibitor Bacteriostatic Gram(+) & Gram(-) bacteria
4. Streptomycin Protein synthesis inhibitor Bactericidal Gram(+) & aerobic
Gram(-) bacteria
5. Erythromycin Protein synthesis inhibitor Bacteriostatic, cidal at high doses Gram(+) & Gram(-) bacteria
6. Polymixin Affects cell membrane Bactericidal Gram(-) bacteria
7. Virginiamycin Protein synthesis inhibitor Bactericidal Gram(+) bacteria
8. Monensin Affects cell membrane Bactericidal Gram(+) bacteria
9. Chloramphenicol Protein synthesis inhibitor Bactericidal (or) static Gram(+) & Gram(-) bacteria.
Good against anaerobes.
Heat stable?
* Has the same properties as penicillin G but is more stable at acidic pH.
294 N.V. Narendranath
Figure 5. A typical disc assay performed to test the antimicrobial activity of penicillin G and virginiamycin against Lactobacillus
paracasei. Note the growth of resistant mutants in the zone of inhibition.
mutants to single antibiotics such as penicillin indicated that LactosideTM significantly reduced
or virginiamycin. One of the strategies to reduce the viable numbers of bacteria. Virginiamycin
the risk of resistance is to use a combination of was the least effective antibiotic at the dosage
antimicrobials, including non-antibiotic used in controlling L. paracasei (Figure 6).
substances, thereby taking advantage of In experiments performed to determine the
synergism in activity as well. Lactoside TM, a minimum inhibitory concentrations (MIC) for
proprietary blend of antimicrobials, reduced risk various antibiotics against lactobacilli at various
of resistance while increasing spectrum of stages of their growth, Lactoside TM at low
activity and pH range. Following five years of concentrations killed L. plantarum at early, mid
use, no build up of resistance has been reported. and late log phases of growth (Table 3).
The advantage of balanced mixtures of
antibiotics has been demonstrated in several Table 3. Minimum inhibitory concentrations for
practical situations. A recent experiment different antibiotics and LactosideTM (expressed in ppm)
examined corn mash (30% dry solids; pH 5.6) for L. plantarum at different stages of its growth.
contaminated with an aggressive Lactobacillus Antibiotic MICs when antibiotic added at
paracasei strain. Yeast was inoculated at 30 various stages of growth
million cells/ml. Just prior to yeast inoculation, 0 hrs Early log Mid log Late log
test fermentors were contaminated (inoculated)
Penicillin G 0.2 12.8 >12.8 >12.8
with L. paracasei at ~10 7 CFU/ml (in the
Penicillin V 0.2 12.8 >12.8 12.8
treatments with bacteria). The antibiotics used, Virginiamycin 0.2 >12.8 >12.8 >12.8
penicillin G, virginiamycin, and Lactoside™, LactosideTM 0.2 0.2 0.2 0.8
were added from 10,000 ppm stock solutions. Oxytetracycline >6.4 >6.4 >6.4 >6.4
Virginiamycin was made up in HPLC-grade Erythromycin 12.8 >12.8 >12.8 >12.8
methanol. The temperature was maintained at Streptomycin >12.8 >12.8 >12.8 >12.8
31.1°C (88°F) throughout the fermentation.
Samples were withdrawn for analysis from each
fermentor at 8, 24, 32, 48, 56, and 72 hrs. Viable Similar results were obtained with four other
cell counts were monitored by the spread plate species of lactobacilli. Figure 7 shows the regions
technique. For enumeration of bacteria, the MRS of a typical growth curve for a microbe where
agar plates (with 10 ppm cycloheximide) were the various antibiotics when used at low
incubated in anaerobic jars at 30°C. The plating concentrations would be effective in killing the
was done in duplicate for each dilution. Results microbe. This information is important for a
Bacterial contamination and control in ethanol production 295
0
0 10 20 30 40 50 60 70 80
Time (h)
Figure 6. Survival of L. paracasei in the fermentation of corn mash (30% DS, pH 5.6) at 31.1°C (88°F) in the presence or
absence of various antibiotics. Values are means of duplicate fermentations that had a CV of <8%.
1.200
1.000
Absorbance (620 nm)
0.800
0.600
Lactoside™
0.400
0.200
0.000
0 12 24 36 48 60
Time (hrs)
Virginiamycin Penicillin G
Figure 7. A typical growth curve of a microorganism showing at what stages antibiotics are effective at low concentrations.
distillery since contamination can be detected The ideal antimicrobial should, however, possess
as early as 6-10 hrs from the start of filling a all the following characteristics:
fermentor. By this time, the contaminants are
approaching the mid-log phase of growth. 1. Broad-spectrum activity: effective against
Therefore, an antibiotic or antimicrobial effective both Gram-positive and Gram-negative
at a low concentration at this stage would be an bacteria.
ideal choice.
2. Effective over a wide pH range.
296 N.V. Narendranath
3. Control the bacteria at various growth stages. with correct proportions of each antibiotic.
Experimental designs are available to generate
4. Effective at low concentrations. such equations for studying the combination of
5. Bactericidal rather than bacteriostatic. more than two antibiotics.
It is becoming a common practice in some
6. Able to be combined with other antibiotics
industrial operations to either under- or overdose
or antimicrobials to avoid resistance
antibiotics for controlling contaminating
development in bacteria.
bacteria. Under-dosing antibiotics leads to the
development of resistance in microorganisms.
FORMULATION OF ANTIBIOTIC Overdosing antibiotics can affect fermentation
COMBINATIONS rate by yeast (Hamdy et al., 1996) and increase
the chances that all the antibiotic will not be
Factorial experiments are conducted to test inactivated by the distillation process. Therefore,
interaction effects (either synergistic or it is always advisable to use the antibiotics at
antagonistic) among antibiotics used in the dosage recommended by the supplier/
combination to control growth of a test lactic manufacturer.
acid bacteria. These experiments are conducted
in corn mash rather than laboratory media so
that the results can be translated to the field. From LOWERING MASH pH CAUSES ETHANOL LOSS
these experiments, the individual effects of the
antibiotics can also be determined. Lactic acid Since the growth rate of lactic acid bacteria is
is the parameter measured since lactic acid reduced significantly at pH levels <5.0 (Kandler
produced is directly proportional to the viable and Weiss, 1986), most ethanol plants tend to
cell numbers of bacteria present in mash reduce the pH of the mash with sulfuric acid to
(Narendranath et al., 1997). The results of the <4.5 at the start of fermentation. Some
interactions are plotted in 3-D graphs with the continuous fermentation plants even start
two test antibiotics on x- and y-axes, and lactic fermentations at pH of 4.0 or less. Lowering
acid on the z-axis. The graphs are created based mash pH may reduce the growth rate of
on the response surface regression analysis of contaminant bacteria but it significantly reduces
the data using the SAS program. Figure 8 shows ethanol yield. In an experiment using mashes
the model created by using the equation based of various dissolved solids concentrations
on the data obtained in an experiment. This adjusted to pH values from 4 to 5.5, significantly
equation can then be used to make formulations higher ethanol was obtained when the initial
1.0-1.2
0.8-1.0
0.6-0.8
1. 2
0.4-0.6
Lactic acid (%w/v)
1. 0 0.2-0.4
0. 8 0.0-0.2
0. 6
0. 4
0. 2 2
0. 0 1
0 0. 5
0. 5 Monensin
1 0
Penicillin 2
Figure 8. Lactic acid produced by L. plantarum in corn mash (pH 5.5, D.S. 24%) at 30°C as affected by various concentrations
(ppm) of penicillin and monensin in combination in a 4 × 4 factorial experiment. Model was significant (P<0.001) with R2 = 0.9993.
Bacterial contamination and control in ethanol production 297
16
pH=4.0 pH=4.5 pH=5.0 pH=5.5
15
14
Ethanol (%v/v)
13
12
11
10
9
8
7
20 25 30 35
Dissolved solids (% w/v)
Figure 9. Ethanol produced by S. cerevisiae in 48 hrs at 30°C at different concentrations of dissolved solids in the medium at four
pH levels. Coefficient of variation among the duplicates was <2%.
mash pH was 5.5 (Figure 9). Moreover, yeast whisky fermentations. J. Inst. Brew. 94:89-92.
can tolerate higher acetic acid (0.1-0.2%w/v) and Booth, I.R. and R.G. Kroll. 1989. The
lactic acid (up to 3%w/v) when the starting pH preservation of foods by low pH. In:
is 5.5, since at this pH these acids are primarily Mechanisms of action of food preservation
in the dissociated form. By the end of procedures (G.W. Gould, ed). Elsevier,
fermentation, however, the pH drops 1 unit to London, UK, pp. 119-160.
4.5 under normal ‘contaminant-free’ conditions. Bryan-Jones, G. 1975. Lactic acid bacteria in
Similar observations were also made with wheat distillery fermentation. In: Lactic acid bacteria
and milo mashes. in beverages and foods. Proceedings of the IV
The occurrence of bacterial contaminants in Long Ashton Symposium (J.G. Carr, C.V.
an industrial-scale ethanol production process Cutting and G.C. Whiting, ed). Academic
is unavoidable, especially when the pH of the
Press, London, UK, pp. 165-175.
mash at set (start of fermentation) is between
Chin, P.M. and W.M. Ingledew. 1994. Effect of
5.0 and 5.5. Therefore, a good cleaning and
sanitation regime combined with an effective lactic acid bacteria on wheat mash
antimicrobial program will reduce bacterial fermentations prepared with laboratory
numbers and significantly increase ethanol backset. Enzyme Microb. Technol. 16:311-
yield, which ultimately results in increased profit 317.
for ethanol producers. Day, W.H., W.C. Serjak, J.R. Stratton and L.
Stone. 1954. Antibiotics as contamination
control agents in grain alcohol fermentations.
References J. Agric. Food Chem. 2:252-258.
Dolan, T.C.S. 1979. Bacteria in whisky
Alexander, M. 1971. Microbial ecology. John production. Brewer 65:60-64.
Wiley & Sons, Inc., London, UK. Hamdy, M.K., R.T. Toledo, C.J. Shieh, M.A.
Aquarone, E. 1960. Penicillin and tetracycline Pfannenstiel and R. Wang. 1996. Effects of
as contamination control agents in alcoholic virginiamycin on fermentation rate by yeast.
fermentation of sugarcane molasses. Appl. Biomass Bioenergy 11:1-9.
Microbiol. 8:263-268. Hynes, S.H., D.M. Kjarsgaard, K.C. Thomas and
Barbour, E.A. and F.G. Priest. 1988. Some effects W.M. Ingledew. 1997. Use of virginiamycin
of Lactobacillus contamination in scotch to control the growth of lactic acid bacteria
298 N.V. Narendranath
during alcoholic fermentation. J. Ind. Narendranath, N.V., K.C. Thomas and W.M.
Microbiol. Biotechnol. 18:284-291. Ingledew. 2000. Urea hydrogen peroxide
Islam, M., R. Toledo and M.K. Hamdy. 1999. reduces the numbers of lactobacilli, nourishes
Stability of virginiamycin and penicillin during yeast and leaves no residues in the ethanol
alcohol fermentation. Biomass Bioenergy fermentation. Appl. Environ. Microbiol.
17:369-376. 66:4187-4192.
Kandler, O. and N. Weiss. 1986. Regular, non Narendranath, N.V., K.C. Thomas and W.M.
sporing gram-positive rods. In: Bergey’s Ingledew. 2001a. Effects of acetic acid and
manual of systematic bacteriology, Vol. 2 lactic acid on the growth of Saccharomyces
(P.H.A. Sneath, N.S. Mair, M.E. Sharpe and cerevisiae in a minimal medium. J. Ind.
J.G. Holt, ed). The Williams & Wilkins Co., Microbiol. Biotechnol. 26:171-177.
Baltimore, MD, pp. 1208-1234. Narendranath, N.V., K.C. Thomas and W.M.
Kheirolomoom, A., A. Kazemi-Vaysari, M. Ingledew. 2001b. Acetic acid and lactic acid
Ardjmand and A. Baradar-Khoshfetrat. 1999. inhibition of growth of Saccharomyces
The combined effects of pH and temperature cerevisiae by different mechanisms. J. Am.
on penicillin G decomposition and its stability Soc. Brew. Chem. 59:187-194.
modeling. Process Biochem. 35:205-211. Ralph, R.J. 1981. Practical aspects of operating
Lindgren, S.E. and W.J. Dobrogosz. 1990. an alcohol plant. In: A Step to energy
Antagonistic activities of lactic acid bacteria independence - a text book for fuel alcohol
in food and feed fermentations. FEMS production (T.P. Lyons, ed.). Alltech Technical
Microbiol. Rev. 87:149-164. Publications, Lexington, KY, pp. 255-265.
Maiorella, B., H.W. Blanch and C.R. Wilke. Salmond, C.V., R.G. Kroll and I.R. Booth. 1984.
1983. By-product inhibition effects on The effect of food preservatives on pH
ethanolic fermentation by Saccharomyces homeostasis in Escherichia coli. J. Gen.
cerevisiae. Biotechnol. Bioeng. 25:103-121. Microbiol. 130:2845-2850.
Makanjuola, D.B., A. Tymon and D.G. Springham. Stroppa, C.T., M.G.S. Andrietta, S.R. Andrietta,
1992. Some effects of lactic acid bacteria on C. Steckelberg and G.E. Serra. 2000. Use of
laboratory scale yeast fermentations. Enzyme penicillin and monensin to control bacterial
Microb. Technol. 14:351-357. contamination of Brazilian alcohol
Narendranath, N.V., S.H. Hynes, K.C. Thomas and fermentations. Int. Sugar J. 102:78-82.
W.M. Ingledew. 1997. Effects of lactobacilli on
yeast-catalyzed ethanol fermentations. Appl.
Environ. Microbiol. 63:4158-4163.
Managing the four Ts of cleaning and sanitizing 299
Chapter 21
Introduction
Cleaning and sanitizing is an integral part of the • Titrations
operation in many processing facilities including
distilleries, ethanol plants and breweries. The • Turbulence
cleaning and sanitizing effort represents a
significant investment of time, labor and Time refers to the frequency of running the
operating costs. At the same time there is usually cleaning operation, for example weekly, daily,
a positive payback for this effort in terms of or after each batch. In a given cleaning cycle it
yields, efficiencies, safety and product quality. also refers to the length of each step. For example
In a clean plant that is running efficiently with a 5 min rinse, a 1 hr caustic wash, etc.
no infection problems, there is always the
temptation to reduce operating costs by trimming Temperature refers to the temperature of the
back cleaning times and chemical usages. This cleaning solutions used in each step of the
type of ongoing evaluation is a worthwhile part cleaning cycle. The first rinse may be at ambient
of plant management. We should always strive temperature while the caustic wash may be
to clean and sanitize to the extent required by specified between 150 and 160oF.
the process and the equipment but overkill in
this area should be avoided. If a little bit is good, Titrations refers to the chemistry of the cleaning
a lot is not necessarily better. solutions. This includes selecting the right
To achieve the proper degree of cleaning and cleaning chemical for the job being done. It also
sanitizing at an acceptable cost, there are a includes the concentration of chemicals used in
myriad of approaches and tools that can be used. the cleaning solutions.
Sometimes called the ‘cleaner’s bag of tricks’,
these tools can be grouped in four categories Turbulence is the mechanical action in the
known as the ‘Four Ts’ of Cleaning and cleaning program that will physically scrub soil
Sanitizing. The Four Ts are: from dirty surfaces. Turbulence encompasses the
use of scrub brushes and high-pressure nozzles
as well as pumping cleaning solutions at high
• Time velocities through dirty pipelines.
• Temperature
300 J. Larson and J. Power
For any cleaning job, many combinations of the Four Ts (Figure 3). If the turbulence created
time, temperature, titrations and turbulence can by the velocity of cleaning solution flowing in a
be used to achieve effective cleaning at a pipeline is inadequate, it may be possible to
reasonable cost. There are also many compensate by cleaning for a longer time, by
combinations that will not work because of a increasing the concentration of the cleaning
deficiency in one or more of the Four Ts. For chemicals or by raising the cleaning solution
example, cleaning time may be too short or too temperature.
infrequent. The detergent may be too weak, it Our challenge is to achieve an acceptable
may be at too low a temperature or it may not degree of cleaning and sanitizing in specific
be the right choice for the soil type present processes. We sometimes go through elaborate
(Figures 1 and 2). procedures but in the end still have infections.
If cleaning is inadequate, it is usually possible Why?
to correct it by increasing the intensity of one of
Time Temperature
Temperature Time Temperature
Turbulence Titration Titration
Turbulence
Figure 1. Some combinations work. Figure 2. Some combinations do not work; and dirt remains.
Time Temperature
Temperature
Titration
Tu
Figure 3. Deficiencies in one of the Four Ts can be corrected or compensated by changing another.
Managing the four Ts of cleaning and sanitizing 301
This paper examines the tools at our disposal Therefore, cleaning is defined as the removal of
for effective cleaning and sanitizing – our soil while sanitizing is the reduction of viable
cleaning bag of tricks. Proper attention to the organisms remaining on the clean surface. An
Four Ts in the toolkit and how they are important concept in cleaning and sanitizing is
combined is needed to make cleaning and that first we remove the dirt, then we sanitize.
sanitizing effective. Sanitizing dirt is not in the program!
Definitions Time
Cleaning is reducing the amount of soil to an If you don’t get it clean the first time, wash it
acceptable level. Cleaning is usually done with again. Spend more time cleaning and you will
a combination of water and added detergents. get better results. This sounds like a simple
Undercleaning means that we have not done the solution, but a dilemma arises because time spent
job properly. Gross overcleaning means wasted cleaning could be time spent producing product.
resources – time, chemicals and manpower. An Figure 4 illustrates this in the operation of a
acceptable level of cleaning in a pharmaceutical batch ethanol fermentor. In the fermentation
plant would probably be overcleaning in an cycle, a period of time is set aside for cleaning
ethanol distillery. and sanitizing between batches. The ethanol
concentration curve, however, shows that we can
Sanitizing is reducing the population of viable produce more ethanol in the fermentor if we omit
organisms on a clean surface to an acceptable all or part of the cleaning time. But what will the
level. Again, ‘acceptable’ has different meanings longer-term results be if we take shortcuts in
in different operations. Within the ethanol cleaning? We must find a happy medium, which
industry it may even vary from plant to plant. is the optimum amount of time taken from
Heat and/or sanitizing chemicals are used for production to keep the process equipment clean.
this step. The other three Ts of cleaning and sanitizing can
help us do this.
Disinfection is the destruction of all vegetative Time refers to the frequency of cleaning as well
microorganisms, but not necessarily spores. as the length of the cleaning cycle. We learn from
experience that some parts of the process are
Sterilization is the complete destruction of all very sensitive to infection and require frequent
organisms including spores and viruses. cleaning. An example is the yeast propagation
system, which is thoroughly cleaned and
Soil is any unwanted material left on a surface sanitized after every use. Other parts such as the
that needs to be clean. Soils can be composed slurry tank or liquefaction tank are less critical
of sugars, salts, fats, proteins, microbes, scales and may be cleaned every 1-6 months.
and other mineral deposits. Knowing the nature Continuous fermentation systems present
of the soil is necessary when planning the special problems compared to batch fermentors.
cleaning procedure used to remove it. Soils left A continuous fermentation train may be shut
on surfaces harbor contaminating bacteria, down only one or two times a year for cleaning.
reduce efficiency in heat exchangers and plug This shutdown requires a total stoppage of
pipes and other passageways. production, so it has significant cost implications.
Because of the continuous flow through the
CIP is the acronym for cleaning-in-place. fermentation train, there is some degree of
Pipelines, tanks, process equipment and flushing of the infecting organism. Frequently
accessories are cleaned by pumping cleaning the culture yeast grows more slowly than the
solutions through them without disassembly or contaminant, and reduced ethanol yield requires
manual cleaning. CIP processes are usually some action. Antibiotics such as Lactoside ™
automated. and Allpen™ and antimicrobials may be added
to the mash to kill bacteria without requiring a
COP is cleaning-out-of-place. COP involves shutdown. For a wild yeast infection it is usually
manual disassembly and cleaning by hand.
302 J. Larson and J. Power
Money
%
Ethanol Time
Fermentor
use
Ferment Ferment
Time
Figure 4. The dilemma between cleaning time vs. production: time = money!
necessary to shut down and clean because Table 2. Effect of temperature on amount of soil present
agents that kill the wild yeast will also kill the before and after use of cold or warm cleaning solutions.
culture yeast.
Organizing a cleaning schedule for every piece Cold cleaning Warm cleaning
23°C (74°F) 57°C (135°F)
of equipment in the plant is an important part of
the CIP program. An example of such a schedule Before cleaning >500,000 >500,000
is shown in Table 1. After cleaning 18,242 59
DAYTIME Start DAYTIME Start DAYTIME Start DAYTIME Start DAYTIME Start DAYTIME Start DAYTIME Start
time time time time time time time
Exch:1302 A or B Exch: 1400 Exch: 1402 Exch: 1404 Exch: 1302 A or B Evaporator
Exch. 1303 Exch: 1401 Exch: 1403 Exch: 1301B Exch: 1302C
Exch 1302C Exch: 1406 Exch: 103
Managing the four Ts of cleaning and sanitizing 303
304 J. Larson and J. Power
warmer or hotter cleaning solution could be the present, they may react with proteins in a reaction
answer. similar to the tanning of leather, forming tough,
water-resistant deposits.
Titrations Oils
Titrations, the third T in the cleaning and Oils are liquid fats and are intrinsically insoluble.
sanitizing bag of tricks, refers to the chemistry Much of the oil present in grains is broken into
of cleaning and sanitizing chemicals and microscopic droplets or emulsions that are
solutions. It includes matching the type of suspended in the mash. As conditions change,
chemical cleaner to the type of soil present and the emulsions may coalesce forming larger
using that chemical at a strength that is both deposits of insoluble oil.
effective and affordable. It is therefore essential
to have as much information as possible about Inorganic compounds
the soil we are trying to remove (McCabe, 1999;
Various inorganic compounds may become
Kretsch, 1994).
insoluble during mashing or fermentation. Most
commonly calcium in water reacts with the
oxalate present in grains to form insoluble
THE NATURE OF SOILS
calcium oxalate, commonly called beerstone.
Any of the compounds present in raw materials Carbonates present in water may combine with
or in the water supply used for fermentation may calcium to form calcium carbonate scale.
become insoluble and be deposited on Magnesium hydroxide tends to co-precipitate as
equipment surfaces as an unwanted soil. part of the carbonate scale. Occasionally water
Methods used to remove these soils depend on with high silicate content may generate a silica
the specific nature of the material involved. scale as the pH drops. Soluble silicate is
converted by acid to insoluble silica (SiO2).
Carbohydrates
Biological deposits
The soluble sugars present during fermentation
may be left on surfaces, especially the non- Many microorganisms tend to grow better on
fermentable sugars. A much bigger problem is surfaces than when freely suspended in a liquid.
the higher molecular weight carbohydrates such The layer of microorganisms that forms together
as gums, which can form tough deposits on with other materials such as non-living soil and
surfaces. Retrograded starch can form when external polysaccharides produced by the
starch is not sufficiently digested by α-amylase microorganisms is called a biofilm (Czechowski
enzymes. Depending on the type of grains used, and Banner, 1992). A biofilm may be observed
gums such as ß-glucans or pentosans may as a slimy layer on tanks used to store water.
become soluble in the mash and then be Fortunately, most microorganisms found in
deposited later on. Dextrans produced by fermentation do not have a pronounced
Leuconostoc bacteria from sucrose can be tendency to form a biofilm, but films of yeast or
produced in plants fermenting sucrose- bacteria growing inside a surface scale are
containing materials. frequently observed. These films are a potent
source of contamination.
Proteins
Most of the protein present in grains is insoluble MATERIALS USED FOR CLEANING
and passes through mashing and fermentation
in the insoluble form. The soluble portion of the Water
protein may become denatured by conditions
Water is a good solvent for many compounds,
such as temperature changes or pH shifts. As
especially polar compounds. Materials such as
the protein becomes insoluble it may be
simple sugars, soluble proteins and the soluble
deposited onto various surfaces. If tannins are
Managing the four Ts of cleaning and sanitizing 305
products of fermentation are easily removed with than caustic or silicates. It is safe for use in hand
water. Most compounds are more soluble in hot cleaning applications. It helps to soften water
water than cold, so heating water helps make it by forming an easily removed precipitate with
a better cleaner. Exceptions are the water hardness compounds. It can be combined with
hardness compounds, particularly those of sodium hypochlorite to form a co-crystal, which
temporary hardness. These become less soluble is a convenient, stable source of chlorine.
at higher temperature leading to the deposition Sodium carbonate or washing soda is a mildly
of calcium carbonate-based water scales. An alkaline compound that helps soften water as
important consideration in the use of hot water well as provide alkalinity. Its main use is in a
for cleaning is to prevent formation of scales by mixture with solid caustic soda where its anti-
softening the water. Many cleaners contain caking activity helps prevent handling problems.
water-softening ingredients. When alkaline compounds raise the pH level,
Pure water has a relatively high surface they increase the amount of negative hydroxide
tension, as shown by the fact that water tends to ions present. This has two main effects. Bonds
form round drops when placed on a surface. between the amino acids of proteins can be
Water is described as not being good at wetting hydrolyzed or broken by reaction with
because the water forms droplets and does not hydroxide. This makes the proteins smaller and
wet the entire surface. Compounds that more soluble so that they can be removed. Acid
concentrate at the surface of water often decrease compounds can have their acidic hydrogen
the surface tension of water droplets, making it removed by neutralization with hydroxide. The
easier for the water to spread out and cover more compounds are converted to their negatively
of the surface. These compounds are called charged ionic form. The negative ions repel each
wetting agents. other, leading to a breakup of large aggregates
of soil to smaller ones. This breakdown is called
Alkali peptizing. The negative ions are also more
soluble in water, leading to their removal from
An alkali is something that raises the pH of the surface. Proteins, composed of amino acids,
water. The pH depends on the concentration of are very effectively cleaned by alkali. If the alkali
hydroxide ions: the more hydroxide the higher is strong enough, carbohydrates can also be
the pH. The most common alkali used for converted into negative ions leading to their
cleaning is caustic soda (sodium hydroxide), easier removal.
which is an excellent source of hydroxide ions Alkali will also break down fats and oils to
and is relatively inexpensive. Unfortunately, their component glycerol and free fatty acids.
caustic soda has poor rinsability; it does not This is the basis for making soap from fat and
rinse away very easily after it is used for cleaning. alkali, and the process is called saponification.
Potassium hydroxide is also a strong alkali and Provided that hardness ions are not present to
it rinses more easily than sodium hydroxide, but precipitate them and that the high pH keeps the
it is significantly more expensive. Addition of fatty acids in the ionized form, fats and oils are
surfactants to cleaning compounds with caustic effectively removed by alkali.
soda will improve rinsability. Most inorganic scales are not effectively
Other alkaline compounds include silicates, removed by alkali.
phosphates and carbonates. Sodium metasilicate
is a commonly used alkali. It provides a high Chelating agents and sequestrants
pH, although not as high as caustic. It tends to
keep removed soil in suspension, providing Chelation is the binding of ions, most
good removal from the area cleaned. Silicate importantly the hardness ions calcium and
does not tend to corrode soft metals such as magnesium, so that the hardness does not cause
aluminum and copper, while caustic attacks precipitation and scale formation. Common
these metals. Silicates are also less dangerous to chelating agents are EDTA (ethylene diamine
handle. tetraacetic acid) and NTA (nitrilo triacetic acid).
Alkaline phosphates, especially trisodium Sequestrants, especially the polyphosphates,
phosphate, provide significant alkalinity, but less have a similar action. Sodium gluconate acts as
306 J. Larson and J. Power
a chelator under strongly alkaline conditions and of soap. The sulfonic acid portion is more easily
is often combined with caustic soda. ionized, so the pH does not need to be so high.
A strong chelating agent can also help dissolve The acid also does not tend to form salts with
inorganic scale that has already formed on water hardness, so these detergents can be used
surfaces. Chelators such as EDTA are relatively in hard water. The drawback with these synthetic
expensive and relatively large amounts are compounds, called anionic detergents, is that
needed to remove scale where it has acumulated they are so surface active that they form a heavy
to significant depths. It is a good idea to measure layer of foam when sprayed onto surfaces in
the amount of free EDTA remaining in solution clean-in-place applications. Their use is therefore
when it is being used for scale removal to make limited in industrial cleaning.
sure that not all of the compound has formed Another class of synthetic detergents, non-
complexes with hardness ions. ionic detergents, does not have a strong
Manufacturers of EDTA do not recommend tendency to form foam. They therefore can be
using EDTA at high pH such as a mixture with used in clean-in-place applications. Most of
caustic soda. The reason is because it loses its these are polyoxyethylene compounds.
effectiveness at chelating magnesium at the high
pH, but its chelation of calcium is not affected. Acids
Since cleaning difficulties are more associated
with calcium than with magnesium, this caution Acids are used in cleaning especially where
can often be ignored. removal of inorganic scales is necessary.
Calcium compounds do not dissolve well in
Surfactants alkali, but are often quite soluble in acid. Most
commonly, phosphoric acid is used or a
Surfactants concentrate at the surface of water. combination of phosphoric and nitric acids.
They have both hydrophobic (water-fearing) Sulfamic acid is also occasionally used.
and hydrophilic (water-loving) properties. Both An acid cleaning can follow an initial cleaning
properties can be satisfied at the same time at with a caustic soda-based alkaline detergent,
the surface where the hydrophilic portion can which removes the organic components of the
remain in the water and the hydrophobic portion soil. The acid is then used to remove inorganic
can exit the water. The surface of the water is residue.
stabilized by surfactants. This means that water Broader range detergents based on acid with
can more easily spread out and wet larger areas additional ingredients to remove organic soils
of surface. By increasing the wetting ability of have been used in some applications, but their
cleaning solutions, surfactants allow the cleaner use is not widespread.
to spread out and clean more of the surface. By
a similar mechanism it also helps the cleaner Bleach
penetrate small crevices in the soil, increasing
its effectiveness. Oxidizing bleaches are used to partially oxidize
Surfactants may also congregate at the interface components of the soil. Bleach can oxidize
between water and small particles of soil. They subunits of long chain proteins or
can form a stabilizing sphere around the particle polysaccharides, producing smaller and more
that keeps it in suspension and allows it to be soluble fragments, which can be more easily
efficiently swept away from the surface. removed. Bleaches are usually added to other
Particles stabilized in this way are called micelles. components of a formulated detergent to
The oldest surfactant used for cleaning is soap. enhance overall cleaning.
Soap is free fatty acid. As a rather insoluble acid, Chlorine is the most commonly used bleach.
it only remains soluble if the pH is high Sodium hypochlorite, produced by adding
(alkaline), stabilizing the acid in water. Soaps chlorine to a caustic solution, can be added in
also form insoluble salts with the hardness ions, small amounts to alkaline cleaners, such as
calcium and magnesium (bathtub ring) so they caustic soda. Chlorinated trisodium phosphate,
work poorly in hard water. a stable solid, is also commonly used.
Newer synthetic detergents such as alkyl (from Typically about 200 ppm of chlorine
oil) benzene sulfonates overcome the problems equivalent is present in the cleaning solution.
Managing the four Ts of cleaning and sanitizing 307
Chlorine can be corrosive to stainless steel if left The lower the pH, the more hypochlorous acid
in contact for long periods of time, so the is formed and the more effective sanitation will
equipment should be thoroughly rinsed be. Unfortunately, if the pH drops too low, free
immediately after cleaning. Chlorinated alkali chlorine gas begins to be released, leading to a
cleaners are very effective at removing soils with very dangerous situation. Therefore, for effective
high protein content and also beerstone scales. use, the pH should be below 8.2 but not lower
Probably the beerstone is stabilized by organic than 7.0. There is a very narrow range where
compounds between the calcium oxalate hypochlorite should be used.
crystals, and the cleaner removes the organic Dual halogen sanitizers have been developed
part very efficiently, making it easier to remove that extend the effective range of chlorine-based
the exposed crystals. sanitizers. A source of bromine is added to
The chlorine in cleaners is diluted to a much hypochlorite, allowing hypobromite to be
lower concentration than in the original sodium formed. The dual halogens are effective at about
hypochlorite; and diluted hypochlorite is much one pH unit higher than plain hypochlorite.
more stable than concentrated. Diluted cleaners It may seem paradoxical that hypochlorite as
have been used at temperatures as high as 80ºC a bleach can be used in high pH alkaline cleaning
(176°F). formulations, but that it should be used at nearly
Hydrogen peroxide is also effective as a neutral pH as a sanitizer. The bleaching action
‘bleach’ for cleaning. Stabilized peroxides such is indiscriminate, while the antimicrobial action
as sodium percarbonate and sodium perborate is more specific, appearing to depend on the
are available in powdered form. These are added ability of the non-ionized form to penetrate and
to cleaner or blended with solid cleaning exert killing action inside living cells.
products. Chlorine in the form of hypochlorite is the least
expensive sanitizer available. It is effective
against a wide range of microorganisms,
SANITIZERS including spores. Its action is not diminished in
hard water. If residual soil is present, the
The use of a sanitizer after effective cleaning hypochlorite will bleach or oxidize the soil and
can provide nearly sterile equipment. It should be lost. It is important that equipment be
be emphasized that the cleaning must be thoroughly cleaned before a chlorine sanitizer
extremely thorough in order to achieve this level is used or it will be inactivated before it can exert
of sanitation. Cleaning removes most dirt from its sanitizing effect. Chlorine is also corrosive
surfaces and a part of this soil will be the to many metals. It may be used on stainless steel
microorganisms left behind after previous use. at low concentrations, but it must be thoroughly
Most of the removal of microorganisms will removed by rinsing after use.
occur during the cleaning step. Sanitizers
remove most of the last organisms remaining Chlorine dioxide
behind.
Chlorine dioxide (ClO 2), is a relatively new
Chlorine-releasing materials sanitizer based on chlorine. Chlorine dioxide will
kill organisms more quickly and at lower
The most commonly used ‘chlorine’ solution is concentrations than hypochlorite. Typical use
actually a solution of sodium hypochlorite. rates are 2 to 10 ppm, measured as available
Adding chlorine gas to a caustic solution chlorine, versus 25 to 200 ppm for hypochlorite.
produces sodium hypochlorite, and under the When residual soil is present, chlorine dioxide
right conditions the chlorine is slowly released reacts more slowly than does hypochlorite so
to do its sanitizing job. The sodium hypochlorite the need for thorough cleaning before use is
can exist in two forms, hypochlorite ion and reduced. It also can be used over a wide pH range
hypochlorous acid. It is the hypochlorous acid since it dissolves as a gas, not forming an acid
that attacks microorganisms and provides the or salt in water. It is less corrosive than
sanitizing effect. When the sodium hypochlorite hypochlorite. The biggest disadvantage of
is mixed with water, the pH of the solution chlorine dioxide is that it is relatively expensive.
decreases and hypochlorous acid begins to form.
308 J. Larson and J. Power
It is a gas and will not remain in water solution and are active at low pH, so they must be
for long periods of time; it must be generated combined with acid. While hypochlorites,
and used on the same day. Many methods of iodophores and chlorine dioxide all dissipate or
generation are available. Acid conversion of evaporate in a relatively short period of time,
chlorite is used on a small scale, but is inefficient. anionics are completely stable. They will last for
Larger scale generators may use dangerous weeks or months and are therefore the best
chemicals such as chlorine gas or hydrochloric choice for applications where equipment is to
acid. Electrolytic generators are a newer be stored in water or filled with water for longer
development that promise greater efficiency and than a few days. The practice of storing
safety. equipment soaking in plain water is extremely
unsanitary and is a very common but bad
Iodine-releasing materials practice. Bacteria will grow in the water,
especially if some residue is left on the
Iodine is a very efficient sanitizer when used at equipment. Slimy layers that often form on the
a concentration of about 25 ppm at colder equipment and storage containers are layers of
temperatures, and at an acid pH. Iodine tinctures billions of bacteria or a biofilm. Store the
in water are not very stable and actively stain equipment with mild acid and an anionic
materials with which they come into contact. surfactant and it will remain clean for months.
Industrial use of iodine is usually in the form of Being detergents, some of the anionic
a complex with detergents and acid called an surfactants are also fairly good cleaning
iodophor. When concentrated iodophor is diluted compounds and acid cleaning can also be
in water, the iodine is freed from its complex accomplished with some of these compounds.
with detergent. The acid in the iodophor reduces
the pH of the diluted solution to the acid range Peracetic acid
where iodine is most effective. The acid can also
help dissolve mineral scale left behind after Peracetic acid is a complex of acetic acid with
alkaline cleaning. Because of the detergents in hydrogen peroxide. The peroxide is the sanitizer.
the iodophor, the sanitizer tends to foam when It is effective at about 200 ppm of peroxide and
used in spray applications. The recent at colder temperatures. It is expensive and is used
development of iodine stabilized with chlorine, mainly in special situations where residue of
‘stabilized iodine’, has eliminated foaming as a chloride or iodide from less expensive sanitizers
problem and requires about half as much iodine or some of their reaction products cannot be
to do the same sanitizing job. Since stabilized tolerated, especially some food applications.
iodine is relatively new, it has not been legally
cleared for use as a food equipment sanitizer in Quarternary ammonium compounds
all areas.
Iodine must be used at an acid pH to be most Quarternary ammonium compounds, called
effective. If an iodophor is used at higher than ‘quats’, are stable, effective and long lasting,
recommended dilution, the pH of the mixed tending to leave a residue. They adsorb to
sanitizer should be checked to make sure that surfaces and have a residual killing effect for
the pH of the diluted solution is not too high. relatively long periods of time. Usually the
Since iodine tends to stain, be certain that staining residue is not desirable in vessels such as
can be tolerated before using an iodine based fermenting tanks since the residue will also kill
compound. There is little tendency to corrode some yeast. If a lot of yeast are present, the
metal equipment unless the metal is not resistant number killed may not be too significant, so
to the acid contained in iodophors. quats are occasionally used to sanitize these
vessels. A better application is to use quats on
Anionic surfactants areas where yeast should not be present. They
are used on the outside of fermenting tanks and
The anionic surfactants are not very commonly on floors where spills are likely to occur to keep
used as sanitizers, but they do have properties down the growth of undesirable organisms such
that make them the choice for some special as molds on these surfaces.
applications. They are surface-active compounds
Managing the four Ts of cleaning and sanitizing 309
Plate
heat
exchanger
Rule of thumb:
1.5 times the normal process flow rate
Plate heat
exchanger
Figure 6. Dead ends and turbulence: did the bypass get cleaned?
Fitted joints, that is homemade or fabricated-in- become trapped and prevent cleaning solutions
the-field tees, are not sanitary and should not be from touching those parts of the pipe.
accepted. For processing equipment the American Meat
Figure 8 shows an innocent looking tee with a Institute created a task force to improve
valve on the branch line. This section of pipe equipment design with the goal of eliminating
was opened for inspection and the branch pipe harborage areas where microorganisms can
was packed with soil. The soil buildup was not collect and grow. The result is a set of guidelines,
removed during CIP because the cleaning the Ten Principles of Sanitary Design (Stout,
velocity was insufficient or because the dead end 2003). These principles, for food processing
was more than one pipe diameter. equipment, are as follows:
During construction, piping should be welded
using sanitary welding procedures including (for 1. Cleanable to microbiological level. A piece
stainless steel) butt welding with an inert gas of equipment must be more than just visually
purge. Pipelines should be self-draining and cleanable – over the entire life of the
without high spots or domes where air can equipment.
Managing the four Ts of cleaning and sanitizing 311
45° Elbow
4" 2"
UNSANITARY TEE
ARRANGEMENT
Fitted joint
4" x 4" x 2" Sanitary Tee
Mash Mash
System System
CIP supply CIP supply
CIP CIP
System System
Return to CIP
Return to CIP
Coolant Coolant
Coolant Coolant
Product Product
Receiver Concrete pad Receiver Concrete pad
Figure 10a. The mash filling pipeline loop. Figure 10b. The fermentor CIP sprayhead loop.
Mash Mash
System System
CIP supply CIP supply
CIP CIP
System System
Return to CIP
Return to CIP
Coolant Coolant
Coolant Coolant
Product Product
Receiver Concrete pad Receiver Concrete pad
Figure 10c. The external heat exchanger loop. Figure 10d. The fermented product loop.
314 J. Larson and J. Power
Temperature, pressure and flow records can D. Acid wash and water rinse (optional). This
be used to document that the CIP programme additional cleaning is done to prevent scale
ran successfully. If automation does not provide formation, for example in beer kegs, or to
these reports, the parameters can be checked and remove scale deposits.
recorded by the operators. These records are
often part of a formal HACCP program. E. Sanitizing rinse. A sanitizing agent may be
Visual inspection may be possible from injected into the clean rinse water to reduce
manways, inspection ports or windows. This microorganisms on the clean surfaces to an
immediately provides an indication of gross acceptable level.
deficiencies in the cleaning process and
appropriate measures can be taken before the CLEANING PROCEDURE FOR CO2 SCRUBBER
next batch is started.
Microbiological swabs and plating, along with A. Stop flow of CO2 through the scrubber and
ATP bioluminescence swabs, are more specific remove CO 2 from inside the scrubber by
indicators of successful cleaning and sanitizing. venting or rinsing with fresh water. Residual
An ATP bioluminescence swab indicates the CO2 in the scrubber will neutralize the caustic
amount of soil left on a surface that may look solution.
‘clean’ (Ehrenfeld et al., 1996). A big advantage B. Make a caustic soda solution 2 to 3% by
is that the result is known in less than one weight at ambient temperature or preferably
minute. Again, appropriate action (repeat the CIP warmed to 100 to 120°F. Make enough to
process) can be taken. Microbiological swabs maintain a level in the bottom of the scrubber
and plating can tell us how much of what when recirculating.
organisms are present, but results are not known
for 2-5 days. Newer rapid methods are C. Circulate the caustic solution through the
shortening this time. scrubber sprayer for at least 1 hr. Several
hours may be required depending on the soil
load.
Examples of CIP cycles D. After 15 to 30 minutes take a sample of the
caustic and titrate to make sure it has not
A TYPICAL CIP CYCLE FOR TANKS AND
been neutralized by residual CO 2 in the
PIPELINES
scrubber. If less than 2% strength, add more
A. Prerinse to the sewer. Removes gross caustic to get 2 to 3%.
amounts of soil before the main wash. It is E. Drain caustic, then rinse using burst rinse
common to save the water rinse (item C) and (water for 30 sec, drain for 3 min.). Repeat
reuse it for this step. This practice conserves burst rinse cycle until rinse water shows no
water, and the rinse water usually has a small caustic when tested with phenylphthalein or
amount of detergent in it that enhances the litmus paper.
effectiveness of the prerinse.
F. Prepare acid wash, 1 to 2%, using ScalebiteTM
B. Alkaline (caustic) wash. For organic and oily at ambient temperature. Circulate for 1 hr.
soils, caustic soda is most commonly used.
It often has additives to improve wetting and G. Drain acid solution and rinse using burst
rinsing properties. Strengths from 1 to 3 % rinse, three cycles.
by weight are common and it is used at H. Prepare IotechTM solution, 25 ppm at ambient
temperatures from ambient up to 180°F. It is temperature, and recirculate for 30 mins.
a more effective cleaner at the warmer
temperatures. I. Drain IotechTM solution and rinse using burst
rinse, three cycles.
C. Water rinse. Rinses out the washing water. It
is often saved for reuse as the prerinse. The J. Scrubber is ready to bring back on stream.
source should be fresh, clean water so as not
to recontaminate.
Managing the four Ts of cleaning and sanitizing 315
D. Rinse with clean water. Inspect visually. Note: • The cleaning solution used was too dilute.
when tank is wet you may not see the scale. • The wind blows contaminated air or an
When it dries, the scale is more visible. aerosol into equipment when it is
E. Scale-removing acid wash. If scale remains temporarily opened after cleaning and before
after the ScalebanTM wash, an acid cleaner use.
may be required. ScalebiteTM is a formulated • Insects carry contamination into a piece of
acid cleaner that is used between 1 and 3% equipment temporarily left open.
in water. It is recirculated several hours. Note:
ScalebiteTM is safe with stainless steel but does
Dead ends are sources of contamination that are
attack mild steel, especially at warmer
always in contact with the process stream.
temperatures. Sulfamic acid is also effective
Usually these are the most serious causes of
against some scales. Like ScalebiteTM, it is
contamination. Some examples are:
used in concentrations between 1 and 3 %,
temperatures up to 140°F.
316 J. Larson and J. Power
• Dead ends in piping that cannot be can be found in Alltech’s Laboratory Manual
thoroughly cleaned. (Alltech, 2001).
• Scale built up inside fermenting tanks or
coolers. Scale cannot be cleaned because the 2 NaOH + CO2 Na2CO3 + H2O
cleaning compounds cannot penetrate the
scale once it is thick enough. Sodium hydroxide Sodium carbonate
• Carbon dioxide vent lines on fermentors into (an excellent cleaner) (a fairly good cleaner)
which fermenting liquid has flowed or been
sprayed as an aerosol. Figure 11. Dilution of caustic by carbon dioxide.
• Poor drainage of cleaning solution from a Cleaning the inside of a fermentor with a
tank while it is being cleaned. sprayhead and cleaning the tank outlet (Figure
12)
• Seats on valves that have worn and begun
to develop cracks The normal flow through a sprayhead is around
• Non-sanitary pipe connections, especially 100 gallons/min (about 380 liters/min). For a 6-
pipe threads. These will trap liquid in the inch fermentor outlet pipe, the minimum velocity
threads that cannot be removed by cleaning. for cleaning is not reached until the flow rate is
410 gallons/min. This is a serious cleaning
deficiency and it is common to see gross
Some of the likely ultimate sources of accumulations of soil on the top half of these
contamination include: outlet pipes because they do not even fill with
cleaning solution when the sprayhead is being
• Grains being brought into a plant carry large used.
numbers of a variety of microorganisms
• People and the food they bring into the plant
The cleaning and sanitizing program
• Insects
The most likely sources of contamination, A routine and effective microbiological sampling
though, are within the plant itself. Infection program should be in operation. Sampling
sources are often found in scale, dead ends, even should be planned so as to provide a complete,
in coolers, spilled material left on the ground, overall picture of the state of sanitation in the
material in sewers that can be splashed into the plant at any time. Special attention should be
air as an aerosol, dirty and contaminated carbon paid to an evaluation of the cleaning procedures
dioxide lines and aerosols created when opening in place. Infections usually develop in areas
contaminated equipment such as perhaps the where cleaning is not effective or is missed.
beer well. Swabs should be taken from cleaned equipment.
Hidden issues related to chemistry occur when An ATP detecting luminometer can provide more
cleaning tanks that have a CO2 atmosphere. The immediate evaluation on the effectiveness of
CO 2 will react with caustic to form sodium cleaning. Strengths of cleaning and sanitizing
carbonate, a cleaner that is much weaker than solutions should be measured routinely, as well
caustic soda (Figure 11). However, if this as temperatures of cleaning solutions.
reacted solution is analyzed by the normal single Sanitation Quality Control is like preventive
end-point titration, the effect is not detected. maintenance for the product. It is troubleshooting
Carbonate titrates just like the hydroxide. This designed to identify potential problems so that
problem can be overcome by replenishing the they can be corrected before damage is done to
reacted caustic or by venting the vessel the product. There are four basic components
completely before cleaning. The correct titration in an effective sanitation quality control
uses barium chloride first, then acid. The method program:
Managing the four Ts of cleaning and sanitizing 317
Vent
20 in. Manway/
sampling port
CIP spray head
100 psi
100 gpm
D A
5 - 10 ft/sec.
Coolant
Agitator C
(if required)
specified at different stages of the process Cleaning and sanitation is an area where the
and each test has standard results or control quality control principle of ‘continuous
limit. improvement’ can be effectively applied. No
plant is perfectly clean. There is always room
C. Records for improvement, and small improvements can
accumulate into major improvements in the
These should include: when was the cleaning overall efficiency and profitability of the plant.
really done (vs schedule), automatic or manual
recording of temperatures and strengths, when References
was maintenance done (UV lights renewed, eg),
problems encountered when the job was done. Alltech Inc. 2001. Laboratory Methods in
Brewing and Distilling. Alltech Technical
D. Results Publications, pg 3-16.
Immediate reporting of microbiological results Anon. 1992. 3-A Accepted Practices For
and charting trends for statistical quality control Permanently Installed Product And Solution
are of key importance. Pipelines And Cleaning Systems Used In Milk
and Milk Product Processing Plants, Number
605-04. Dairy, Food and Environmental
Summary Sanitation, Volume 12, No. 2.
For every cleaning and sanitizing problem there Czechowski, M.H. and M. Banner. 1992. Control
is a solution, in fact there can be many solutions. of biofilms in breweries through cleaning and
Unfortunately the solutions are not available as sanitizing.Tech. Q. Master Brewers Assoc. of
defined recipes in a cookbook. The Four Ts, our Am. 29(3):86-88.
cleaning and sanitizing bag of tricks, experience Ehrenfeld, E.E., J. Scheld, S.A. Miller and C.R.
and trial and error are the tools at our disposal. Carpenter. 1996. Use of ATP-bioluminescence
We know that the program is to first remove the in the brewing industry. Tech. Q. Master
dirt, and then sanitize. Sanitizing dirt may work Brewers Assoc. Am. 33(1):59-62.
for a short time but it is not a permanent solution. Kretsch, J. 1994. Practical considerations for
We must also determine the acceptable levels of brewery sanitation. Tech. Q. Master Brewers
cleanliness and sanitation and then avoid Assoc. Am. 31(4):124-128.
excesses that are hard on the equipment and McCabe, J.T. 1999. The Practical Brewer, 3rd
costly in terms of time, money and utilities. Edition. Master Brewers Association of the
We must use our knowledge of the Four Ts of Americas.
cleaning and sanitizing to find a combination of Peter, W.A. 1983. Designing for automation.
time, temperature, turbulence and titration that MBAA Technical Quarterly 20(1):22.
is both effective and economical. Finally, we Stout, J. 2002. Perspective on practices in food
must realize that while CIP is the ideal, COP plant sanitation and hygiene. Food Safety
combined with CIP is usually the reality. Magazine, April/May pg.20.
Stout, J. 2003. 10 Principles of equipment design
“Regardless of technology, the most
for ready to eat food processing operations.
important food plant sanitation approach
Food Safety Magazine, June/July, pg. 18.
is getting back to basics, which means
having motivated, passionate people who
know both the importance of and how to
clean effectively. It comes down to training
and motivation; simply put, how to use
mechanical action, with the right
detergent, at the right concentration for
the right time at the right temperature with
the right attitude, and giving people credit
for the good work they do.”
(Stout, 2002).
Recovery
Ethanol distillation: the fundamentals 319
Chapter 22
P.W. MADSON
KATZEN International, Inc., Cincinnati, Ohio, USA
Vapor
Reflux
Rectifying
Low boiling product
Feed mixture
Stripping
Steam (energy)
will be devoted to a description of the d. Vapor will rise up the tower and liquid will
modifications required of the simple distillation flow down the tower. The purpose of the
system in order to make it effective for the tower internals (trays) is to allow intimate
separation of a very pure ethanol product, contact between rising vapors and
essentially free of its water content. descending liquids correlated with separation
Figure 2 expands on Figure 1 by showing some of vapor and liquid.
additional features of a distillation tower. These
are:
Figure 3 shows a vapor-liquid equilibrium
diagram for the ethanol-water system at
a. The highest temperature in the tower will
atmospheric pressure. The diagram shows mole
occur at the base.
percent ethanol in the liquid (X axis) vs mole
b. The temperature in the tower will regularly percent ethanol in the vapor (Y axis). The plot
and progressively decrease from the bottom could also be made for volume percent in the
to the top of the tower. liquid vs volume percent in the vapor and the
equilibrium curve would only be slightly
c. The tower will have a number of similar,
displaced from that shown in Figure 3. Mole
individual, internal components referred to
percent is generally used by engineers to analyze
as ‘trays’ (these may also be described as
vapor/liquid separation systems because it
stages or contactors).
Ethanol distillation: the fundamentals 321
Vapor
Reflux liquid
Overhead product
Feed
Trays (contacting devices)
temperature
Higher
Vapor
Thermal energy
Boiling liquid
Bottoms product
relates directly to molecular interactions, which become quite complex and would not lend itself
more closely describe the process occurring in easily to graphic analysis.
a distillation system. Referring to Figure 3, a 45o line is drawn from
Analysis of the ethanol-water distillation the compositions of 0, 0 to 100% and 100%.
system is mathematically straightforward when This 45o line is useful for determining ranges of
using molar quantities rather than the more compositions that can be separated by
common measurements of volume or weight. distillation. Since the 45 o line represents the
This is because of an energy balance principle potential points at which the concentration in the
called ‘constant molal overflow’. Essentially, this vapor equals the concentration in the liquid, it
principle states that the heat (energy) required indicates those conditions under which
to vaporize or condense a mole of ethanol is distillation is impossible for performing the
approximately equal to the heat (energy) separation. If the equilibrium curve contacts the
required to vaporize or condense a mole of 45 o line, an infinitely large distillation tower
water; and is approximately equal to the heat would be required to distill to that composition
(energy) required to vaporize or condense any of vapor and liquid. Further, if the equilibrium
mixture of the two. This relationship allows the curve crosses the 45 o line, the mixture has
tower to be analyzed by graphic techniques formed an azeotrope. This means that even if
using straight lines. If constant molal overflow the tower were infinitely large with an infinite
did not occur, then the tower analysis would amount of energy, it would be impossible to
distill past that point by simple rectification.
322 P.W. Madson
Azeotrope
194° proof
100
90
Equilibrium Curve
80
70
Mole % ethanol in vapor
60
50
40
30
20
10
0
0 10 20 30 40 50 60 70 80 90 100
Consider a very simple system consisting of a collection of small amounts of vapor, analysis
pot filled with a mixture of ethanol in water (a would reveal that each successive portion of
beer) containing 10% by volume ethanol (3.3 vapor would become richer in ethanol.
mole %). This composition is identified in the Thus we have created a series of steps by
lower left portion of Figure 3. A fire could be which we keep increasing the ethanol content
kindled under the pot, which would add thermal of the analyzed sample, both liquid and vapor.
energy to the system. The pot would begin to Unfortunately, this oversimplified process is
boil and generate some vapor. If we gathered a idealized; and practically speaking, is
small portion of the vapor initially generated and impossible. However if we had supplied our
measured its ethanol content; we would find original pot with a continuous supply of ethanol-
about 24 mole % ethanol (53 volume %). If we water feed and vapor generated in the first pot
condense this vapor (note: there will be only a was continuously condensed and supplied to the
small amount of this vapor), boil it in a second second pot, etc. then the process becomes similar
pot and again collect a small amount of the initial to the industrial distillation tower operation
vapor generated, this second vapor would shown in Figure 2.
contain about 55 mole % (83 volume %) of How far can this process be extended? Could
ethanol (see Figure 3). If we should continue we produce pure ethanol by continuously
this simplified process to a third and fourth extending our process of boiling and reboiling?
Ethanol distillation: the fundamentals 323
The answer is, no! We would finally reach a solutions of ethanol and water are also known
point in one of the downstream pots, where the as constant boiling mixures (CBM) since the
vapor boiling from the liquid was of the same azeotropic liquid will have the same temperature
composition as the liquid from which it was as the azeotropic equilibrium vapor being boiled
being generated. This unfortunate consequence from itself. Without some sort of drastic process
limits our ability to produce anydrous ethanol intervention, further ethanol purification
from a dilute ethanol-water feed. What we finally becomes impossible. The question then
encounter in our simplified process is the becomes: What can we do to make it possible to
formation of an azeotrope. This is a concentrated produce anhydrous ethanol? Methods of doing
solution of ethanol and water that when boiled so will be covered later in this chapter.
produces a vapor with a composition identical Figure 4 depicts the structure of the distillation
to the composition of the liquid solution from process by dividing the vapor/liquid equilibrium
which it originated. information into three distinct zones of process
In summary then, we are limited in ethanol- and equipment requirements: stripping, rectifying
water purification in any single multistage and dehydration. This division is the basis for
distillation tower to the production of azeotropic the design of equipment and systems to perform
ethanol-water mixtures. These azeotropic the distillation tasks.
Anhydrous
product
Stillage
Stripping system
Spirit
Dehydration
Rectifying system system
100
90
Equilibrium curve
80 Azeotrope 194° proof
70
Mole % ethanol in vapor
60
50
40
30
20
10
0
0 10 20 30 40 50 60 70 80 90 100
Mole % ethanol in liquid
(10 volume %)
DISTILLATION CONTACTORS
Figure 6 depicts a perforated tray contactor with
certain accoutrements required to control the
Trays are the most common contactor in use. flow of liquid and vapor and to assure their
What are the functions expected of tray intimate contact. Another type of tray contacting
contactors in the tower? Figure 5 depicts a single device, the disc-and-donut or baffle tray is
tray contactor in a distillation tower and shows shown in Figure 7. The characteristics of this
the primary functions desired: type of contactor make it especially useful for
distilling materials such as dry-milled grain beer,
- mixing rising vapor with a falling fluid which would foul ordinary trays such as
perforated, valve, etc.
- allow for separation after mixing
- provide path for liquid to proceed down the ENERGY ANALYSIS
tower
In addition to the selection of the basic contacting
- provide path for vapor to proceed up the
device, the energy requirement must be
tower
Enriched Downcoming
Outlet weir vapor liquid
Inlet weir
V/L separation
V/L mixing
Downcomer
area
Perforated
area Vapor Stripped
liquid
Figure 6. Perforated trays.
Ethanol distillation: the fundamentals 325
Descending Descending
liquid liquid
Enriched
Donuts vapor
Disc
Vapor
Stripped stillage
Example 2. Calculate the steam required to strip 100 gpm of a 5% volume beer (95 gpm water/5 gpm ethanol).
L/V = 6.33 (typical for a 5% volume beer) or L = 6.33 • V
L = 95 gpm • 500 lbs/hr = 47,500 lbs/hr = 6.33 • V
gpm
Therefore, V = 7,500 lbs/hr (steam)
And: 7,500 lbs/hr (steam) • hr = 25 lbs steam/gallon of product
5 gpm (product) 60 min
*L and V are liquid and vapor flow rates, respectively, expressed in lb-mole per hr.
Note: at base of column use simplifying assumption of water (L) and steam (V). Therefore: L (lb-mole/hr) = L (lbs/hr)
V(lb-mole/hr) V(lbs/hr)
326 P.W. Madson
design chart like that shown in Figure 8 for the are based on the energy input, as calculated and
ethanol-water system. Such a graph is useful represented in Figure 8. Because of the principle
when calculations are needed to ascertain of constant molal overflow, the operating lines
technical and economic feasibility and can be represented as straight lines. If constant
preliminary conditions for the design. molal overflow was not valid for the ethanol/
Figure 9 demonstrates how the liquid:vapor water distillation, then these lines would be
flow ratio, in connection with the number of curved to represent the changing ratio of liquid
stages (theoretically ideal trays) required for a flow to vapor flow (in molar quantities)
specified separation between ethanol and water, throughout the tower. The slope of the operating
is graphically determined. Note that the stages line (the ratio of liquid flow to vapor flow) is
are constructed by drawing straight lines also called the internal reflux ratio. If the energy
vertically and horizontally between the input to a tower is increased while the beer flow
equilibrium curve (previously determined remains constant, the operating lines will move
experimentally) and the operating lines. For an toward the 45o line, thus requiring fewer stages
ethanol stripper/rectifier, there are two operating to conduct the distillation. Likewise if the energy
lines: one for the rectification section and one input is reduced (lowering the internal reflux
for the stripping section. The operating lines ratio), the operating lines will move toward the
represent the locus of concentrations within the equilibrium curve, reducing the degree of
distillation tower of the passing liquid and vapor separation achievable in each stage and therefore
streams. The operating lines for a given tower
30
Constraints:
1. 190° proof product
2. 0.02% (wt) bottoms
25 3. Saturated feed
4. Direct steam
Lbs steam/gallon (anhydrous ethanol basis)
Practical minimum
20
15
10
Theoretical minimum
0
0 5 10 15
90 8 9 10 11 12 13 14
Equilibrium curve
80
7
6
5 190° proof
70 4
Mole % ethanol in vapor
60
2
50
40
Operating line
rectification
30
1
Operating line
stripping
20
Typical L / V=5.0
10
0
0 10 20 30 40 50 60 70 80 90 100
Stillage
(0.02 wt %)
Mole % ethanol in liquid
Beer feed
(10 volume %)
requiring more stages to conduct the distillation. diameter required for the given distillation tower.
The calculations underlying the preparation of Since all of the distillation ‘work’ is done by the
Figure 9 go beyond the scope and intent of this trays, the tower is actually the ‘container’ to
text, but have been included for continuity. The surround the vapor and liquid activity that is
dashed lines represent the graphical solution to ‘managed’ by the trays. Tower diameter design
the design calculations for the number of is, therefore, actually the design of the necessary
theoretical stages required to accomplish a tray diameter for proper vapor/liquid interaction
desired degree of separation of the feed and movement.
components. Figure 9 is referred to as a McCabe- The f factor (vapor loading) is an empirically
Thiele diagram. For further pursuit of this determined factor that depends primarily upon
subject, refer to the classical distillation textbook tray type and spacing, fluid physical properties,
by Robinson and Gilliland (1950). froth stability and surface tension at the
operating conditions of the system. The proper
values for f are determined by field observations.
TOWER SIZING In summary then, the f factor can be described
as an adjusted velocity term (units are ft/sec) that
The goal of the design effort is to establish the when multiplied by the square root of the density
size of the distillation tower required. Table 2 ratio of liquid to vapor, will give the allowable
shows the basic procedure to determine the vapor velocity in the empty tower shell, such
328 P.W. Madson
Example: Calculate tower diameter required for a 10% volume beer at 100 gpm (15 lbs steam/gallon of product)
W Vapor flow rate = 9,000 lbs/hr (steam)
P Operating pressure at base = 1.34 ATM
MAVG Average MW of vapor = 18 lbs/lb-mole
ρL Liquid mixture density = 59.5 lbs/ft3 (227 °F)
T Absolute operating temperature = 687 °R
D Tower inside diameter in inches
f Vapor loading factor = 0.05-0.3
W 9000 16.45
D = 0.2085 • = 0.2085 • =
Sizing equation: P • M AVG • UL f•
1.34 • 18 • 59.5 f
f•
T 687
Assuming f = 0.16 (specific to tray design and spacing), the tower diameter is:
16.45
D= = 41.125 inches
0.16
Tower diameter calculation utilizes the following equation. Values for the terms are indicated above.
W
D = 0.2085 •
P • MAVG • UL
f•
T
The final design equation can be derived beginning with the fundamental equation:
-
u = f • ȡL ȡV
ȡV
(Note: For most cases ρL is much greater than ρV, so that ρL – ρV ≅ ρL. For example, water (steam) at 212oF and atmospheric
pressure: ρL = 59.8 lbs/ft3 and ρV = 0.0373 lbs/ft3. Then ρL – ρV = 59.7627 lbs/ft3 ≅ 59.8 lbs/ft3, which results in a negligible
0.06% error.)
ȡL
Consequently, u# f •
ȡV
Imposing the equation of continuity: W = A•ρV•u or u = W/A•ρV
Where A = tower cross-section area (ft2)
ρV = vapor density (lbs/ft3)
u = average vapor velocity in empty tower shell (ft/sec)
and W = vapor mass flow (lbs/sec)
Substituting for u in the equation above:
W W W
# f • ȡL then # f • ȡȡLL • ȡȡvV or A #
A • ȡȡV
V ȡV A f • ȡȡLL • ȡȡvV
Ethanol distillation: the fundamentals 329
Using the Universal Gas Constant R = 0.73 (ft3)(atm)/(lb-mole)(oR) the equation becomes:
W
A= ʌ • D2
P •• M U
AVG • ȡ L
MAVG Now A= = 0.7854 • D22
1.17 • f • 4
T
2 W 1.0879 • W
D = =
and PP ••M
MAVG U
AVG •• ȡ LL P •• MAVG UL
AVG • ȡ L
0.9192 • f • f•
T T
1.043 • 12 W W
Adjusting units: D = • Therefore, D (inches) = 0.2085 •
60 PP ••M
MAVG U
AVG •• ȡ LL PP••M U
AVG •• ȡ LL
MAVG
f• f•
T T
that liquid entrainment and/or vapor phase vapor/liquid mass transfer systems involves a
pressure drop in the tower will not be excessive. number of interrelated parameters. The positive/
Excessive vapor velocity will first manifest negative balance of a variety of contacting
itself by causing excessive liquid entrainment devices with different capacities and efficiencies
rising up the tower, causing loss of separation for promoting vapor/liquid mass transfer must
efficiency. Ultimately the excessive entrainment be taken into consideration. Along with the
and pressure drop will cause tower flooding. technical issues considered in such designs,
To achieve a well-balanced tower design, the economical operation is essential not only in the
foregoing analysis must be performed at each reduction of energy and other direct costs, but
stage of the tower, from bottom to top. also in relation to investment and return on
Composition changes, feed points, draws, etc., investment from the operation being considered.
each can cause a different requirement. The In this respect, distillation towers are not
tower must be examined to locate the limiting independent process-wise, as consideration must
point. also be given to other auxiliaries such as
Similar analyses, with empirically-observed reboilers, condensers, pumps, controls and
performance coefficients, are applied to vapor related equipment.
passing through the trays and through the liquid,
and to the movement and control of liquid
passing through downcomers and across the SIZING TOWERS
trays. These analytical procedures are beyond
the scope of this text. Reference should be made In determining optimum diameter and height of
to the aforementioned text by Robinson and towers for distillation, absorption, stripping and
Gilliland for further information. similar mass transfer operations, design factors
are affected by whether the installations will be
indoors or outdoors. With indoor installations,
Considerations in optimizing distillation building height limitations, as well as floor level
accessibility, are an important factor in the
system design design. Where there are height limitations, towers
Optimizing the technical and economic design must be increased in diameter to provide for
of distillation equipment and similar gas and reduced tray spacing, which in turn will require
330 P.W. Madson
lower vapor velocities. With outdoor Trays of various types are predominant in
installations, literally the ‘sky is the limit’, and vapor/liquid contacting operations, particularly
refinery and petrochemical towers of 200 feet on the very large scale encountered in the
in height are not uncommon. petroleum and petrochemical industries, in large
In either case, indoors or outdoors, the scale operations of the chemical process
interrelated tower diameter and tray spacing are industries and in the large scale plants of the
limited by allowable entrainment factors (f motor fuel grade ethanol industry.
factors) (Katzen, 1955). If outdoors, tower The venerable bubble cap tray, with a wide
heights and diameters must be related to variety of cap sizes, designs and arrangements
maximum wind loading factors in the specific to maximize contact efficiency, has fallen out
plant location and may be complicated by of favor during the past few decades because of
allowance for earthquake factors. the relatively high cost of manufacture and
assembly. Valve trays of several types have taken
over in operations requiring a relatively wide
TRAY AND PACKING SELECTION vapor handling capacity range (turndown). This
has been extended by use of different weights
Vapor/liquid contacting devices may be of two of valves on the same tray. Specialty trays such
distinct types, namely packed or tray (staged) as the Ripple, Turbogrid, tunnel cap and others
towers. In packed towers, the transfer of material designed to improve contact under certain
between phases occurs continuously and specific circumstances have been used to a
differentially between vapor and liquid through- limited extent.
out the packed section height. By contrast, in The long established perforated tray is a
tray towers, the vapor/liquid contact occurs on contacting tray into which a large number of
the individual trays by purposely interrupting regularly oriented and spaced small circular
down-flowing liquid using downcomers to openings have been drilled or punched. These
conduct vapor-disengaged liquid from tray to trays are commonly referred to as ‘sieve trays’
tray and causing the vapor/liquid contact to occur because of the original practice of putting the
between cross-flowing liquid on the tray with maximum number of holes in any given tray
vapor flowing up through the tray. In other area. This original design produced a fairly
words, the vapor/liquid contact is intermittent inefficient operation at normal loading, and a
from tray to tray, and is therefore referred to as very inefficient operation with decreased vapor
being stagewise. Thus, for any given separation loading. About 50 years ago, engineers began
system, the degree of vapor/liquid contact will to suspect that the design approach had been in
be greater with a greater height of the packed error, and that the hole area in the trays should
section, or in the case of tray towers, a greater be limited by the hole velocity loading factor to
number of trays used. obtain maximum contact by frothing, as
It is generally considered that packing-type indicated in Figure 10.
internals may be used with relatively clean vapor The hole vapor loading factor (or perforation
and liquid systems where fouling is not a factor) is defined as the vapor velocity through
problem. Economics indicate that packing is the perforations adjusted by the square root of
applicable in small and modest sized towers. As the vapor density at the specific tower location
the towers become larger, packing becomes of a given perforated tray. With the parallel
complicated by the need for multiple liquid development of separation processes in the
redistribution points to avoid potential vapor/ petroleum refining, chemical and ethanol
liquid bypassing and reduction in efficiency. industries, the modern approach has developed
Structured packings (Fair et al., 1990; Bravo et to what is now called ‘perforated tray’ design.
al., 1985) are designed to minimize these The Fractionation Research Institute of the
problems by reducing the height requirement and American Institute of Chemical Engineers
controlling, to some extent, the distribution of diverted its efforts from bubble cap studies to
liquid. However, high fabrication and perforated tray testing; and have established a
specialized installation costs would indicate that basis for the design of perforated trays with high
these are applicable only for relatively low- efficiency and wide capacity range (Raphael
volume, high-value product processing. Katzen Associates, 1978).
Ethanol distillation: the fundamentals 331
Enriched
vapor Downcoming
Outlet weir Froth zone liquid
Inlet weir
V/L separation
V/L mixing
Downcomer
area
Vapor Stripped
liquid
Vapor
Rectifying
Reflux
Steam (energy)
Reboiler
Condensate
return to boiler
High boiling product
of cooling tower water with limited temperature distillation (similar to multiple effect
rise and minimal scale-forming tendencies. On evaporation) are also practiced. Pressure-to-
the other hand, where water is extremely scarce, atmospheric, atmospheric-to-vacuum, or
air-cooled condensers are used. pressure-to-vacuum tower stages are utilized,
with the thermal energy passing overhead from
one tower to provide the reboiler heat for the
Energy conservation next one. Two such stages are quite common
and three stage systems have also been utilized
The increasing cost of thermal energy, whether (Katzen, 1980; Lynn et al., 1986).
provided by natural gas, fuel oil, coal or biomass, Furthermore, the modern technique of vapor
is fostering an increased emphasis on heat recompression, commonly used in evaporation
recovery and a reduction in primary thermal systems, is also being applied to distillation
energy usage (Fair, 1977; Petterson et al., 1977; systems. Such a system can provide for
Mix et al., 1978). Conventional bottoms-to-feed compression of overhead vapors to a pressure
heat exchangers are now being supplemented and temperature suitable for use in reboiling a
with recovery of overhead vapor latent heat by lower pressure stripping tower. However, the
preheating feed streams and other intermediate compression ratios required for such heat
process streams. Techniques of multistage recovery may consume almost as much
Ethanol distillation: the fundamentals 333
electrical energy as would be saved in thermal The operation that has been most subject to
input. Alternative systems, using vapor critical comment is the distillation process. Many
recompression as an intermediate stage device relatively new ‘authorities’ in the field have
in the distillation system, have also been based their criticism on technologies that go
proposed. back 50-60 years, and have created an
unwarranted condemnation of distillation as a
viable process for low energy motor fuel grade
Control systems ethanol production. Systems developed over the
years will be described to show that much of
Control systems can vary from manual control, such criticism is unwarranted and unjustified.
through simple pneumatic control loops to fully
automated distributed control (Martin et al.,
1970). High level computer control has
facilitated the application of sophisticated PRODUCTION OF INDUSTRIAL ETHANOL
control algorithms, providing more flexibility,
Prior to the emphasis on motor fuel grade
reduced labor and higher efficiency with lower
ethanol, the major ethanol product utilized
capital investment. Such systems, when properly
worldwide was high purity, hydrous industrial
adapted to a good process design, have proven
ethanol, which is generally produced at a
more user-friendly than the control techniques
strength of 96 o GL (192 o US proof) ( oGL =
utilized in the past.
degress Gay Lussac = % by volume ethanol; US
proof = 2 x % by volume ethanol). Efficient
systems have been in commercial operation for
Economic design many years for the production of such high
In integrating the technology discussed, the final grade ethanol from ethylene, grain, molasses and
analysis must be economic. Alternative systems sulfite waste liquor. The basic distillation system
must be compared on the basis of investment is shown in Figure12.
requirements, recovery efficiency and relative In the case of synthetic ethanol (outside the
costs of operation. Thus, any heat exchangers scope of this publication), the beer stripping
installed for heat recovery must show a tower is not required and the refining system is
satisfactory return on the investment involved a simple three tower unit, which achieves 98%
in their purchase and installation. In comparing recovery of the ethanol in the crude feed as a
alternative separation systems, the overall first grade product. The final product may
equipment costs must be compared against contain less than 5 ppm total impurities and has
energy and other operating costs to determine a ‘permanganate time’ of more than 60 minutes.
which system offers the best return. Modern For the production of industrial or beverage
computer-assisted designs incorporate spirit products made by fermentation of grain,
economic evaluation factors so economic molasses or sulfite liquor, the system utilizes the
optimization can be determined rapidly. full complement of equipment shown in Figure
12. The beer feed is preheated from the normal
fermentation temperature in several stages,
recovering low level and intermediate level heat
Ethanol distillation/dehydration: specific from effluent streams and vapors in the process.
systems technology This preheated beer is degassed and fed to the
beer stripper, which has stripping trays below
Proven industrial technologies are available for
the beer feed point and several rectifying trays
distillation of various grades of ethanol from
above it. The condensed high wines from the
grain, sugarcane, molasses and other feedstocks.
top of this tower are then fed to the extractive
Improvements have been made over the years,
distillation tower, which may operate at a
particularly during development of the motor
pressure of 6-7 bars (87-101.5 psi). In this tower,
fuel grade ethanol industry. In such installations,
most of the impurities are removed and carried
a key requirement is the minimization of total
overhead to be condensed as a low grade ethanol
energy usage.
334 P.W. Madson
Preheater/
condenser Condenser
CW
Heads by-product
Beer feed
CW
Reflux
CW
Washwater
concentrating tower
Fusel oil
Condenser/ Condenser/ cooler Fusel oil
Reboiler Reboiler reboiler reboiler Reboiler by-product
Fusel oil
Steam Steam Steam decanter
Stillage
Waste water
stream, from which a small purge of heads A small heads draw is taken from the overhead
(acetaldehyde and other low boiling impurities) condensate, which contains the acetaldehyde
may be taken while the primary condensate flow fraction along with a small amount of the ethanol
is fed to the concentrating tower. The purified, produced. This may be sold as a by-product or
diluted ethanol from the bottom of the extractive burned as fuel. A fusel oil side draw is taken at
distillation tower is fed to the rectifying tower, high fusel oil concentrations through a cooler
which has an integral stripping section. In this to a washer. In the washer, water is utilized to
tower, the high grade ethanol product, whether separate the ethanol from the fusel oil, with the
industrial or potable, is taken as a side draw from washings being recycled to the concentrating
one of the upper trays. A small heads cut is tower. The decanted fusel oil may be sold as a
removed from the overhead condensate. Fusel by-product. The ethanol recovered from the
oils (mixtures of higher alcohols such as propyl, crude streams is taken as a side draw from the
butyl, and amyl alcohols and their isomers, which concentrating tower and fed back to the
are fermentation by-products or ‘congeners’) are extractive distillation tower for re-purification
drawn off at two points above the feed tray but and recovery of its ethanol content.
below the product draw tray to avoid a buildup In an early version of this system, installed
of fusel oil impurities in the rectifying tower. The more than 60 years ago for the production of
overhead heads cut and the fusel oil draws are potable ethanol from grain and from molasses,
also sent to the concentrating tower. all towers were operated at atmospheric pressure.
It should be noted that the rectifying tower is However, installations made within the past 40
heated by vapors from both the pressurized years utilize the multistage pressure system to
extractive distillation tower and the pressurized reduce energy consumption to a level of about
concentrating tower. 50% of the all-atmospheric system.
In the concentrating tower, the various streams The commercial installations utilizing the
of congener-containing draws are concentrated. multistage pressure, or ‘pressure cascading’
Ethanol distillation: the fundamentals 335
technique operate with a steam consumption of and minimize investment, a common condensing
3.0-4.2 kg of steam/liter (25-35 lb/gallon) of 96o and decanting system is used for the two towers.
GL ethanol. This may be compared to about 6 The entrainer used to remove water as a ternary
kg of steam/liter for earlier conventional (three component) azeotrope may be benzene,
distillation systems. heptane (C 6-C 8 cut), cyclohexane, n-pentane,
diethyl ether or other suitable azeotropic agents.
Production of anhydrous ethanol The entrainer serves to create a three component
azeotrope that boils at a temperature lower than
Systems have been designed and installed for any of the three individual components and
production of extremely dry and very pure lower than the ethanol/water binary (two
anhydrous ethanol for food and pharmaceutical component) azeotrope. Therefore the ternary
use, primarily in aerosol preparations. These mixture will pass overhead from the tower,
systems, as shown in Figure 13, yield ethanol carrying the water upward. Upon condensing,
containing less than 200 ppm water (99.98o GL), the mixture separates in a decanter into an
less than 5 ppm total impurities and more than entrainer-rich layer and a water-rich layer.
60 minutes permanganate time. The hydrous ethanol feed enters the
The two tower dehydrating system has been dehydrating tower near the top. The feed
operated in two super-anhydrous plants in contacts the entrainer in the upper section of the
Canada, and was used to produce motor fuel tower. The three component mixture in this
grade ethanol (99.5o GL) in four installations in section of the tower seeks to form its azeotrope,
Cuba (prior to the advent of the Castro regime). but is deficient in water and contains more
The dehydrating tower and the entrainer- ethanol than the azeotrope composition.
recovery tower are operated at atmospheric Therefore, the ethanol is rejected downward in
pressure. Thus, they may utilize either low the liquid and is withdrawn as an anhydrous
pressure steam, hot condensate or hot waste product from the bottom of the tower. The water
streams from other parts of the ethanol process joins the entrainer, passing upward as vapor to
to minimize steam usage. To simplify equipment form a mixture that is near the azeotrope
CW
Cooler
Anhydrous ethanol
to storage
Dehydration
190° proof ethanol feed tower
from distillation Entrainer
recovery tower
CW
Condenser
CW
Reboiler Cooler Reboiler
Low pressure steam
or hot condensate Decanter
Condensate return
Waste
water
Chapter 23
The equipment used during the dawn of the fuel product with benzene; and there was no way to
ethanol program was built using a strange totally protect plant workers from exposure to
combination of technologies borrowed from known or expected carcinogens.
different industries. Most of the basic production
expertise came from the beverage alcohol
industry, but there was no need to remove all SYNTHETIC ZEOLITES AND EARLY
the water from beverage alcohol. To prevent MOLECULAR SIEVE DEHYDRATORS
phase separation when blended with gasoline,
fuel grade ethanol needed to be almost An inventor named Skarstrom received a patent
completely dry. Standard distillation still leaves in 1957 for a device that utilized a synthetic
over 4% water in the ethanol, so a process called zeolite adsorbent that selectively removed water
‘azeotropic distillation’ was used in all the early from air and some other gasses and vapors
commercial ethanol plants to remove the final (Ruthven, 1984). A different zeolite could even
water from (dehydrate) the ethanol. Azeotropic be used in virtually the same device to separate
distillation uses a third component, typically oxygen and nitrogen from air (Ruthven, 1984).
benzene or cyclohexane, to ‘break’ the azeotrope These synthetic zeolites became known as
(the composition above which standard ‘molecular sieves’ due to the very precise pore
distillation becomes ineffective). size that enabled them to select and remove one
Azeotropic distillation systems tended to be molecule size from a bulk mixture containing
quite expensive, difficult to operate and adjust, molecules with a larger size or lower polarity. A
and consumed a significant amount of energy. properly designed system could dry air to a dew
Attempts were made to reduce the operating point of -100oF; and drying air became the first
difficulties and energy consumption, but the two major application for molecular sieves. By the
goals were incompatible. Multi-pressure early 1980s, at least a dozen companies offered
techniques could reduce energy consumption, molecular sieve air dryers.
but at the expense of more difficult operation If molecular sieves could be used to dry air,
and considerably higher capital cost. A process why not ethanol? Experiments were under way
upset could easily contaminate the ethanol
338 R.L. Bibb Swain
at that time by several inventors and companies molecular sieve dehydrator design that had been
to prove efficacy of that concept. In the August developed over the past three years by Delta-
1981 issue of Gasohol USA (the only trade T’s founder. The new design eliminated the
journal at the time), two companies advertised reliability and desiccant deterioration problems
molecular sieve dehydrators that dried ethanol that plagued earlier designs. Delta-T built a 150-
in the vapor phase. Ad-Pro Industries, Inc. of gallon/hr prototype at a small ethanol plant in
Houston, Texas, and Anhydrous Technology, Charles City, Virginia. After conducting a battery
Inc. from Sugarland, Texas, both offered small of tests to prove the concept and optimize the
skid-mounted systems for drying up to 200 design, the prototype continued to produce
gallons per hour. Two other companies, Pall ethanol commercially at the plant until the
Pneumatics and Shroeder Farms, were marketing Virginia ethanol incentive expired many years
molecular sieve ethanol dehydrators that dried later. Delta-T sold six dehydrators during its first
the ethanol in the liquid phase. The liquid phase eight months of operation, the largest of which
systems used too much energy, had severe was 800 gallons/hr.
problems with desiccant fouling and finally were In the meantime, larger ethanol plants
just not able to compete with the vapor phase continued to be built using azeotropic distillation
units. The two companies eventually sold dehydration technology. Molecular sieve
approximately a dozen of the small vapor phase dehydration was a relatively new technology
dehydrators. Since the devices were relatively with a somewhat troubled past; and Delta-T had
simple, many entrepreneurs built unauthorized no large reference unit to show a customer. In
copies for themselves and others. addition, the firms providing technology for the
The early vapor phase dehydrators had balance of the plant had their own azeotropic
problems of their own. Aside from a general lack distillation technologies to sell the customer.
of reliability, the molecular sieve desiccant They were quick to point out both the difficulties
would deteriorate within a year; and all the dust of getting uniform vapor flow through large beds
that resulted would contaminate the ethanol and that the technology was not proven at a larger
product. The commercial molecular sieve scale. Both were correct, but misleading.
desiccant was manufactured by mixing the Not able to sell a large dehydrator to an
molecular sieve crystalline powder (much like American ethanol plant, Delta-T formed a joint-
glass dust) with a clay binder, and forming the venture with an Italian engineering firm to build
slurry into beads or extruded pellets about 1/8 a 30 million gallon/year dehydrator for the
inch in diameter. The finished product had the largest distillery in Europe, located in Sicily.
look and toughness of limestone. Any movement Design specifications called for the feedstock
of a bed of the material resulted in abrasion ethanol to be as low as 160 proof (80% ethanol
between the particles. The resulting dust by volume), with a 199.4 minimum product
accumulated in the bed causing vapor proof. Performance guarantees were stringent,
channeling and product contamination. Bed but Delta-T badly needed a large reference unit,
movement could be caused by excessive so it contracted to provide the unit. Happily for
temperature swings (thermal expansion and all project participants, the new world-scale
contraction), excessive pressure drop across the dehydrator worked just as promised, and Delta-
bed (fluidization) and local disturbance by high- T finally had proven the design at large scale.
speed vapor entering the bed during cycle Molecular sieve dehydrators had always
transition (jetting). By 1987, Ad-Pro Industries offered the industry lower capital costs, lower
and Anhydrous Technology, whose designs were operating costs, and greater personnel safety than
subject to all these problems, were both out of azeotropic distillation. Now that a large unit was
business. on line and performing well, the last obstacle to
industry acceptance was overcome. At first,
ethanol plant developers would get their plant
NEW TECHNOLOGY BECOMES PROVEN ON technology from the older process design firms,
LARGE SCALE but they would come directly to Delta-T for their
molecular sieve dehydrator. When the process
In 1984, Delta-T Corporation was formed in technology firms realized they would no longer
Williamsburg, Virginia, to build and market a be able to sell azeotropic distillation dehydration
Development and operation of the molecular sieve: an industry standard 339
to their customers, they reluctantly adopted by the operating temperature and pressure.
molecular sieve technology. Today, all new Moisture capacity increases as pressure increases
ethanol plants are built with molecular sieve or temperature decreases, and vice versa. Since
dehydrators. a pressure swing adsorption dehydrator operates
at almost constant temperature (Ruthven, 1994),
we can desorb the water adsorbed on the
How does a molecular sieve dehydrator previous cycle by lowering the operating
work? pressure and passing a purge vapor through the
bed in the opposite direction to sweep the water
PRESSURE SWING ADSORPTION molecules from the free space surrounding the
desiccant beads.
Most modern molecular sieve dehydrators use The class of materials known as zeolites occurs
a process known in the industry as ‘pressure naturally, but most commercial molecular sieves
swing adsorption’ to remove water from a are man-made. Synthetic zeolites have a
vaporized feed stream. We will concentrate crystalline lattice structure that contains openings
herein on the removal of water from an ethanol (pores) of a precise size, usually measured in
vapor stream, but many other applications are angstroms (Å). Molecular sieves can be
possible with virtually the same device. The term manufactured with different pore sizes by using
‘pressure swing’ refers to the fact that the different chemistry and manufacturing methods.
dehydrator uses a relatively high pressure when A synthetic zeolite of type 3Å is used in most
water is being removed from the feed stream ethanol dehydrators, because the pores are 3Å
and a relatively low pressure when the molecular in diameter while water molecules are 2.8Å and
sieve desiccant is being regenerated (having ethanol molecules are 4.4Å. Therefore, water
water removed from the desiccant). Most molecules are strongly attracted into the pores
commercial designs have two or more beds of but ethanol molecules are excluded. Many other
desiccant and cycle the vapor flow through the adsorbent materials are available that have an
beds to provide continuous operation. While one affinity for water such as activated alumina.
bed is on-line drying the feed vapor, another However, other adsorbents have a wide range
bed is being regenerated. In some designs one of pore sizes and therefore are much less
or more additional beds are being depressurized selective than molecular sieves.
or repressurized in preparation for the next cycle Water is so strongly attracted to type 3Å
(Le Van, 1996). molecular sieve that for each pound adsorbed,
1,800 BTUs of heat are released. This effect is
referred to as the heat of adsorption. When you
HOW DOES ADSORPTION WORK? remove that same pound of water during
regeneration, you must supply 1,800 BTUs of
The ‘macroscopic’ principles of adsorption are
heat (referred to as the heat of desorption). Thus,
quite simple, but attempts to accurately explain
as feed vapor is dehydrated the bed of desiccant
how adsorption works often fail because the
heats up; and as the bed is regenerated it cools
author gets bogged down in technical jargon.
down. Type 3Å molecular sieve is capable of
The basic characteristic of an adsorbent material
adsorbing up to 22% of its weight in water, but
is a stronger affinity for one type of atom or
pressure swing adsorption dehydrators operate
molecule than for the other types in the vapor
at a much lower moisture loading and limit the
stream. In the case of a molecular sieve ethanol
dehydration period to prevent an excessive
dehydrator, we select an adsorbent with a strong
temperature swing.
affinity for water and little affinity for ethanol
and the other impurities typically contained in
the ethanol feed stream. As wet ethanol vapor
passes through the bed, the desiccant adsorbs Operational considerations
the water molecules but not the ethanol ENERGY NEEDS, PRODUCTION CAPACITIES
molecules. This process cannot continue
indefinitely, because the desiccant has a finite By limiting the adsorption period, temperature
capacity for water; and that capacity is affected swings are kept to a reasonable value and the
340 R.L. Bibb Swain
heat of adsorption can be effectively stored in Molecular sieve dehydrators can accommodate
the bed of desiccant as sensible heat. The heat a wide range of specifications, and can be built
is then available to supply the necessary heat of for any desired production rate. Some are
desorption during the regeneration period. The currently in service drying ethanol containing
ability to recover the heat of adsorption is why as much as 20% water and some can produce
molecular sieve dehydrators are so energy anhydrous ethanol with as little as 20 ppm water
efficient. Assuming the feed vapor comes directly in the final product. The smallest units are
from the rectifier column, the only energy used designed for a product rate of about two gallons
directly by the molecular sieve dehydrator is the per hour (mostly for lab and test work), while
steam required to superheat the vapor to bed the largest can produce 100 million gallons per
operating temperatures (0.1 to 0.2 lbs of steam year.
per gallon of anhydrous ethanol) and the
electricity to operate the pumps (0.02-0.03 Kw
hr per gallon). Many companies forget that A TYPICAL OPERATING CYCLE DESCRIBED
additional energy is required to redistill the
The following describes a typical operating cycle
liquid that results during regeneration as well as
for a 2-bed dehydrator drying ethanol containing
the electricity to operate the cooling tower and
5% water. We begin with Bed 1 on-line (drying
air compressor. A properly designed system will
the feed vapor) and Bed 2 being regenerated
have a totally inclusive energy consumption of
(see Figure 1). Wet ethanol vapor enters the top
about 4,000 BTUs per gallon of anhydrous
of Bed 1 under a modest pressure, passes
ethanol, assuming the feed ethanol contains 5%
downward through the bed and exits the bottom
water and the product is dried to 0.25% water.
as dry vapor. 60-85% of the vapor leaves the
Additional energy is required if the feed ethanol
system as anhydrous ethanol product; and the
is wetter or if the product is dryer.
Dehydrating Regenerating
Bed Bed
(High Pressure) (Low Pressure)
remainder is fed to Bed 2 as a regeneration purge controller (computer or PLC) to provide greater
stream. Bed 2 is maintained under a vacuum operating safety, reliability and energy efficiency.
that shifts the equilibrium conditions so that the Once adjusted to standard operating conditions,
adsorbed water is desorbed. Operating the molecular sieve dehydrator requires little
pressures, temperatures and flow rates are operator interface and can tolerate reasonable
adjusted so that regeneration is complete at the variations in feed rate or quality without need
point in the cycle when the beds change position for readjustment. Typical quality control
(valves are positioned to reverse the roles of the measures include sampling the product, feed and
two beds). One bed typically stays on-line regeneration liquid at regular intervals to assure
dehydrating the feed stream for 3-10 minutes the product meets specifications and that the
before being regenerated. When the beds system is operating efficiently.
transition from dehydrate to regenerate mode,
the pressure in the inlet header tends to sag,
which can upset the rectifier column if References
provisions are not made to mitigate the pressure
drop. After the beds switch, the bed that was LeVan, M.D. 1996. Fundamentals of adsorption.
being regenerated ‘pressures up’ to the Proceedings of the 5th International Conference
dehydrating pressure and the vaccuum system on Fundamentals of Adsorption. Kluwer
lowers the pressure in the other bed to the Academic Publishers, Boston, MA.
regenerating pressure. The cycle repeats itself Ruthven, D.M. 1984. Principles of adsorption
continuously. Since one bed or the other is and adsorption processes. John Wiley & Sons,
always receiving and drying the inlet vapor, the New York.
process seems to be continuous; but in reality it Ruthven, D.M., S. Farooq and K.S. Knaebel.
is more properly termed as ‘cyclic batch’. 1994. Pressure swing adsorption. VCH
While a molecular sieve dehydrator can Publishers, Inc., New York.
operate on a simple timed cycle, the best
systems are managed by an ‘intelligent’
Engineering ethanol fermentations
Water reuse in fuel alcohol plants: effect on fermentation 343
Chapter 24
W.M. INGLEDEW
Applied Microbiology and Food Science Department, University of Saskatchewan, Saskatoon,
Saskatchewan, Canada
The subject of water relationships in a fuel reincorporated into the new mash. In Figure 1,
alcohol distillery is an interesting and complex input water entering the system via the grain is
issue. Stringent emission controls are in part relatively easy to quantitate, as the average
forcing engineers to design and run newer moisture levels of corn or wheat are normally
distilleries to be virtually free of organics in stack known. Well water (part of the makeup water
emissions and in any water effluents (Bryan, used in the cook), recycled water such as
2003). This dictates that most water outputs in backset, carbon dioxide scrubber water, side
the distillery (except for the cooling tower stripper bottoms recycle water, and methanator
waters) are recycled inside the plant, and that recycle water, are all included. Some of the
organic materials in those waters are therefore volatile compounds in these streams, ethanol
not dispersed onto fields or into water bodies included, would be lost in cooking due to the
where they have a BOD (biological oxygen temperature employed and vented in part into
demand) or COD (chemical oxygen demand) the atmosphere. However, a significant portion
impact on the surrounding environment. ‘Zero of the ethanol remains in the pressurized slurry
discharge’ is a laudable goal in this industry, but during the cook and hold cycle, and then flashes
such design schemes have created significant off. The energy and steam vapor content is used
problems for microbiological fermentation. to run the distillation columns.
Incomplete understanding of such issues led to It should be noted that most of the existing
a comment I first made at the 1999 Fuel Alcohol plants do not incorporate biomethanator
Workshop: “A fermentor is not a garbage can”. technology, and therefore evaporator condensate
This comment, made due to concerns in certain water is added directly to the cook water tank.
fuel alcohol plants, became a catalyst for When biomethanation is accomplished, a side
discussions among engineers, biochemists and discharge line is made available for removal of
microbiologists associated with this some of the water from the plant when/if
multidisciplinary industry, and is beginning to necessary. If methanator water is discharged, well
lead to extensive analyses of water effluents water volumes are increased correspondingly.
from all conceivable sites in modern alcohol Minor input streams of much less concern in the
plants. plant include addition of enzymes (α-amylase
The pre-cook section of a generic dry milling at the mingler and at liquefaction; glucoamylase
fuel alcohol plant is given in Figure 1. It is in and urea at the fermentor; or ammonia/
this area or at the fermentor that waters are ammonium hydroxide at the mingler. Yeast and
344 W.M. Ingledew
55 Evaporator condensate
Carbon Side
dioxide Backset stripper
Discharge? scrubber water
water
90 40
20
Cook system
for a plant producing
Cook 4 Well 20 million gallons/year
56 water water
80
Corn
13
flour
Mash mingler Slurry
290 gpm
Figure 1. Water inputs and outputs from the pre-cook system of a typical dry mill plant (numbers are in US gallons per minute).
yeast nutrients are added at the fermentor. An (DMWG) and distillers dried grains with solubles
approximate mash bill in a dry mill plant is given (DDGS) contribute water output values in
in Table 1. relation to the degree of drying of the solids.
Less well quantified is stack moisture resulting
Table 1. An approximate mash bill for 50 tons/day of from drying of the grain. Other outputs include
mash in a dry mill fuel alcohol plant. evaporator syrup (if marketed separately from
wet or dried grains), dryer water emissions, and
Component in mash Tons of water stack emissions.
Grain 13.2 tons @ 15% 2.0
The theme of this chapter is that substrate is
Plant water 5.3* continually used and ethanol and carbon dioxide
Evaporator condensate 15.9* are continually produced in the system whether
Flash tank condensate 1.4* it operates by batch or continuous technologies.
Steam ~1.2* All other components, either unused in the
Calcium chloride 0.03 medium or made by the yeasts (or by
Wash waters with NaOH etc. Volume not determined* contaminants in the fermentation) are recycled.
Backset 12.9* Some waters that recycle are continually
Ammonia 0.06 increasing in concentration of the compounds
Yeast Variable**
present until the most inhibitory of these
*Components that should be monitored for chemical content. eventually affects fermentation rate by slowing
**Components that should be monitored for microbial the growth rate of the yeast, killing the cells or
content. altering yeast metabolism. This leads to loss of
ethanol concentration, reduced yields, sugar
Output water is more difficult to quantitate. Wet (substrate) ‘bleed’ into the beer well and a general
distillers grain, distillers modified wet grains lack of performance of the plant. It should be
Water reuse in fuel alcohol plants: effect on fermentation 345
noted that corn wet mill plants – for example Steep water is used as the primary source of
those that use starch slurry, corn steep liquor nutrients in fermentation of the ~40% starch
and backset – have the same general slurry obtained as a by-product in corn wet
considerations as dry mill plants. A complete milling plants. Although used only once in
and accurate audit of water in and water out will, theory, steepwater residues recycle in thin stillage
however, have to await a thorough examination or backset, and this thin stillage therefore
of all systems in selected facilities that can then contains all components that were not used by
be used as models. yeast or that were produced in fermentation but
In most plants, the volumes of water, water- not distilled out of the system. During each
based chemicals and the water contents of recycle, the concentrations of the materials
by-products can be read off computer screens recycled will increase. Note that Table 2 does
or estimated by the weight of the products and not provide ion analyses.
analytical data. What these plants lack, however,
is an exact quantitation on composition of these Table 2. Composition of light steep water.
streams, and the variability of contents and
volumes over time, as well as information on Characteristics
special purges of water such as seen with sodium
pH 3.98
hydroxide cleaning solutions or sanitizers. A Specific gravity (density meter) 1.0527
major concern with process input streams is toxic (13.7 g/100 mL)
or inhibitory components of the streams that Dry weight, g/100 mL 11.8
affect yeast and therefore the rate and extent of N (Kjeldahl method), g/100 mL 0.80
ethanol manufacture. Sodium ion and other α-amino N, mg/L 756*
components will be discussed later in this Total amino N, mg/L 1035*
context. These considerations put in the context Carbohydrate, g/100 mL 5.0*
of continuous fermentation systems are even Lactic acid, g/100 mL 3.82*
more problematic. Purges of waste streams of Acetic acid, g/100 mL 0.21*
undesirable content into fermentors can create *HPLC analysis
an upset to continuous fermentation trains, Thomas and Ingledew (unpublished data)
causing cascading changes from the upstream
site of entry throughout the rest of the train.
Moreover, the maximum concentration of such Table 3 provides an estimate of the amino acid
chemicals is always in the fermentor with the content in unused corn steep liquor, and
most crucial need for ideal conditions – the first therefore the value of this component in the
fermentor (or pre-fermentor) – where growth of nutrition of yeast. Yeasts only use amino acids,
yeast is an absolute requirement. More than a small peptides (<3 amino acids long),
week may be required to regain the balanced ammonium ion, and urea as nitrogen sources.
growth conditions characteristic of continuous The free amino acids are preferentially utilized,
fermentation. It is no wonder then that some but in a distinct order.
plants never really operate in a constant or
consistent mode (in balanced growth) for any Table 3. Free amino acids in light steep water.
extended time!
α-amino acid nitrogen (µmole/mL)
Some of the input components with potential
problems for fermentation are listed and detailed Asp 0.93 Tyr 0.94
below. Compositions and variabilities of the Glu 1.57 Val 3.91
components found in each recycled source are Ser 5.22 Met 1.90
Gly 2.62 Cys 0.19
all not yet well understood.
His 0.73 Ile 1.98
Thr 4.03 Leu 8.5
Ala 11.68 Phe 3.08
Light corn steep water Arg 2.03 Trp 0.57
Pro nd Orn 1.38
A representative composition of steep water from Lys 2.74
the wet milling of corn is recorded in Table 2. Thomas and Ingledew (unpublished data)
346 W.M. Ingledew
These data show that the amino acid spectrum Addition of extra usable nitrogen has been
in this light steep water is broad and varied, shown to improve the growth of yeast and the
providing good amino acid nutrition to the mash rate of fermentation of mashes (Ingledew, 1995;
in which it is used (note that a major dilution of 1999). More importantly, higher concentrations
all compounds would be made on addition of of alcohol can then be made. This is the basic
the corn steep water to the starch slurry used in principal behind very high gravity (VHG)
wet mills to make the mash). Steepwater is fermentation technology (developed in this lab)
normally used at about 12% of the mash volume being used today in the fuel alcohol industry
(Table 4). where concentrations of ethanol well above 16%
v/v can now be made under routine conditions
Table 4. Approximate makeup of mash in a 1000 gpm in a fuel alcohol plant.
mash continuous fuel alcohol distillery, showing major The quantity of lactic acid in corn steepwater
input waters and the ~40% starch slurry. is noteworthy. It originates from the bacteria that
are believed to play an important part in the wet
Components Gallons per minute
milling of corn – bacteria which grow and
Starch slurry 252 metabolize during the steeping process. After
Stillage 415* some recycling of thin stillage through
Stripper bottoms 68* sequential fermentations, the lactic acid levels
Steepwater 123* can become significant, inhibiting yeast growth
Enzymes, ammonia or urea, soda ash 5 and metabolism (Narendranath et al., 1997).
Well water 138* Lactic acid content limits the amount of corn
Yeast (depends on balance of the process) ??** steep liquor (and perhaps also thin stillage)
possible in fermentation mash makeup, as do
*Components that should be monitored for chemical content.
**Components that should be monitored for microbial
the foaming properties of these streams (leading
content. to entrainment of liquid in the carbon dioxide
environment in the headspace), which has been
Usable nitrogen is deficient in all grain samples known to cause ‘blowing’ of pressure relief
that could possibly be used for fuel alcohol valves or plates on the tops of fermentors,
production worldwide (Table 5). Nitrogen creating an immediate and dangerous carbon
deficiency is a main cause of stuck and sluggish dioxide problem for personnel in the plant.
fermentations in this industry. For this reason,
virtually all fuel alcohol plants add liquid or solid
urea, ammonia, or ammonium hydroxide, Thin stillage
although the amounts may not be optimized.
Thin stillage is used in both dry and wet mill
plants. Although it varies from plant to plant, it
Table 5. FAN levels in typical mashes from grains usable is generally low in utilizable nitrogen
for fuel alcohol production. compounds and in total nitrogen. In fact, there
are few, if any, positive reasons (other than water
Grain Free amino nitrogen (mg/L mash*)
recycling) to use thin stillage. Table 6 provides
Total FAN Usable FAN an analysis of thin stillage along with a grouped
Wheat 82 64 estimate of the major minerals in the corn (P, S,
Barley 84 62 K, Ca, Mg, Cu, Fe, Mn, and Zn only). Phosphorus
Hulless barley 124 100 levels ranged from 815 to 1762 mg/L while
Oats 193 159 potassium ranged from 705 to 2643 and
Hulless oats 184 130 magnesium levels in all stillages examined were
Rye 103 83 200-721 ppm. Other elements were much lower,
Molasses 267 141
including Zn, Fe, Cu, Mn. Sodium ion was not
Corn 70 58
Starch slurry ~0 ~0
measured at that time.
The lactic acid concentration in a wet mill
*All mashes normalized to 22% total solids for comparison. alcohol plant can also be problematic. In these
Thomas and Ingledew, unpublished data.
plants, it can be shown that the concentration of
Water reuse in fuel alcohol plants: effect on fermentation 347
lactic acid contributed by the light steep water Table 8. Backset samples from six fuel alcohol plants.
and the stillage can easily exceed the 0.8% (8
g/L) minimal concentration reported to affect Constituent Wheat Wheat Wheat Corn Corn Corn
yeast (Narendranath et al., 1997; 2001a; 2001b).
Total solids, % w/v
In this respect, the valuable contribution made Solubles 4.6 6.3 3.4 4.5 2.8 2.4
to yeast nutrition by light steep water is Insolubles 8.0 7.8 2.3 3.2 2.0 2.0
somewhat negated by the lactic acid present in
it and in thin stillage – acid which inhibits the Total N, % w/v 0.18 0.11 0.13 0.03 0.05 0.54
FAN, mg/L 96 79 51 118 75 88
growth and metabolism of yeast. The extent of Glucose, mg/mL 1.6 33 1.0 0.7 0.6 0.2
this contribution is shown in Table 7. Lactic acid, g/L 11 17 1.5 7 11 6
pH 4.2 3.6 4.7 4.7 3.7 4.0
Table 6. Proximate composition of corn thin stillage. Minerals, mg/L 5648 2085 2893 4149 3626 2946
Components g/L
A number of differences were noted among these
Total solids 48
Total nitrogen 0.53
six samples chosen at random from industry. Note
Crude protein 3.0 the levels of lactic acid, which we know begin
Free amino nitrogen (mg/L) 75 to exert a strong effect on yeast at 8 g/L. Note
Glucose 0.60 also that the free amino nitrogen values (FAN)
Lactic acid 11 are quite low, because any usable FAN in the
pH 3.7 stillage had already been made available to the
Minerals 3.64 yeast in one or more prior fermentations. FAN is
a measure of the free amino groups in amino
Jones and Ingledew, 1994
acids, small and large peptides and soluble
Table 7. Lactic acid concentrations in fermentors in wet
proteins. As yeasts are unable to use any peptides
mills entering the system from stillage and steep water. larger than a tripeptide (and the latter only slowly,
Patterson and Ingledew, 1999; Ingledew and
Mash constituent L/1000 Lactic acid Final mash Patterson, 1999), there will always be residual
L mash (g/L) concentration FAN in the backset, but there is little usable FAN!
(g/L) This is the basis of the comment made
previously that the quality of FAN (and thus the
Light steep water 134 38.2 4.7 quality of backset) is of less value than many in
Stillage < 415 11.0 4.57
the industry would believe.
Total 9.27
The glucose content of these backset samples
is also interesting. It can be concluded that one
Again, a complete audit of water composition wheat plant was having considerable problems
(including recycled waters) in all areas of the in fermentation at the time of the backset
distillery would allow us to determine the exact provision. Over 3% glucose was found in this
influence that each water stream (with its own backset; and this plant would have been suffering
particular chemical composition) might have on a loss of ethanol yield of about 15 g/L, a darker
the fermentation process. As not all of the above DDG after drying, and a reduced level of
materials have been repeatedly analyzed over a nitrogen in reused backset caused by Maillard
broad range of times and conditions so that the reaction complexing of sugars with nitrogen
significance of their contents is known, this sources during heating of the subsequent mash
chapter can only discuss what we do know in the jet cooker.
about some of the water components. As noted above, the concentrations of chemical
A few years ago, we analyzed six samples of components in thin stillage will increase as the
backset from batch fermentation plants. Half the backset recycles through successive
samples were from wheat plants and half from fermentations. This was well recorded by Chin
corn plants. These results are shown in Table 8 and Ingledew (1993; 1994) where a gradual
and were published by Jones and Ingledew, increase in lactic acid to 1.4% was seen in infected
1994. mashes and where losses in yeast viability of up
348 W.M. Ingledew
to 60% were shown. This occurred late in measured in the plants that collect data) are
fermentation after ethanol was totally made to found in Table 9.
the rather low alcohol levels of 7-7.5% v/v pre-
determined by the mash used in Chin’s work. Table 9. Analytical data on evaporator condensate.
The impact of backset on the fermentation is
because it contains any bacterial end products Component* g/100 mL
that are not removed in distillation (lactic acid
Ethanol ND**
and glycerol, for example), ions that were not Methanol (or compound with
taken up by the yeast in the previous same mobility)*** 0.169
fermentation(s), unused solubles from the Sugars
original mash, insolubles from previous mashes, DP3+ (MW of maltotriose and larger) 0.017
if any, and small amounts of volatiles which were DP2 (MW of maltose) 0.013
not completely distilled. As the backset DP1 (MW of glucose) 0.065
repeatedly recycles, the concentrations of some Acids
of these materials increase and some can become Lactic ND
inhibitory. Acetic 0.037
Succinic ND
Glycerol 0.015
is engineered with direct steam injections (rather 9 ppm sodium (Na) to over 20 ppm phosphorus
than as a reboiler), the side stripper bottoms will (P) and sulfur (S). Other ions including calcium
contain small amounts of organic residues. Table (Ca), iron (Fe) and magnesium (mg) were all
11 shows the composition of compounds so far below 0.016 ppm. Extensive analyses have not
identified in side stripper bottoms. This water been available to this author.
also shows the presence of a number of
components that have been entrained in the
water processed through the stills. The amounts Implications of wastewater on yeast
are not large. It should be noted that the fermentation
assessment of this sample has been done to date
only with the Aminex column and refractive Consideration of the inhibitory compounds
index HPLC (system commonly used for found in the above recycled waters led us to
fermentation monitoring). The possibility for considerable research and a diagram of the
other compounds like fusel alcohols, and other stresses that yeasts must endure. We now know
organics not measured or identified by this that many of the stress factors act in a synergistic
column is not ruled out. manner when more than one agent is found at
inhibitory concentrations (Narendranath et al.,
Table 10. Components of CO2 scrubber water. 2001). We also know that the presence of solids
Component* g/100 mL in the medium and the medium buffering
capacity both have an influence on the inhibition
Ethanol 2.52
Methanol (or compound with of the growth of yeast by organic acids (Thomas
same mobility) 0.115 et al., 2002), as does the temperature of the
Sugars fermention. Figure 2 (Ingledew, 1999) shows
DP3+ (MW of maltotriose and larger) 0.044 that a number of possible stress agents find their
DP2 (MW of maltose) 0.011 way into fuel alcohol fermentors via backset and
DP1 (MW of glucose) 0.041 by recycling of other wastewater streams within
Acids the plant. When concentrations increase to levels
Lactic ND** near or above the levels indicated (as single
Acetic ND**
inhibitors), the yeast will not grow or metabolize
Succinic ND**
Glycerol 0.006 at the same rate as is possible under ideal
fermentation conditions (Thomas et al., 2001).
*5 samples analyzed In some cases, the levels of compounds such as
**Not detected
acetic acid, lactic acid and ions like sodium are
Table 11. Components of side stripper bottoms.
so high in fermentation that normal ethanol
production is retarded. The immediate
Component* g/100 mL consequence in both continuous and batch
Ethanol 0.132 fermentation is that unused sugars are sent to
Methanol (or compound with the beer well. This ensures that lower ethanol
same mobility) ND** yields than those expected under ideal
Sugars conditions are seen.
DP3+ (MW of maltotriose and larger) 0.055
The concentrations of stress agents listed in
DP2 (MW of maltose) 0.022
DP1 (MW of glucose) 0.040
Figure 2 have, for the most part, been determined
Acids as accurately as possible. Unfortunately, when
Lactic 0.013 medium conditions (mash type and
Acetic 0.019 concentration, pH, presence of additional
Succinic 0.002 stresses) change, so will the level of each
Glycerol 0.035 ‘inhibitor’ examined that affects yeasts. We now
*22 samples analyzed know that the composition of the medium (i.e.
**Not detected corn mash) and the environmental conditions
imposed (pH, temperature etc.) will influence
Interestingly, side stripper water also has been the concentrations of any one inhibitor, which
found to contain mineral content ranging from can affect yeast growth and metabolism. In fact,
350 W.M. Ingledew
pH >0.8% Carbohydrate
level (not sugar)
3.0 to 4.0 23% but less
for growth
Ethanol
Sulfite >100 mg/L
(varies with strain)
Acetic acid
MASH >.05% w/v
>500 mg/L
Figure 2. Stress factors that influence growth and metabolism of yeasts. Much of this work was done with a synthetic fermenta-
tion medium and may represent the smallest values of a single stress factor that is inhibitory to yeast.
inhibition by lactic acid and acetic acid is pH potential to the industry: recycling treated water
dependent (Narendranath, unpublished; Thomas back to fermentors.
et al., 2002). As the knowledge of synergistic The anaerobic bioreactor in the biomethanator
effects among inhibiting compounds and the must be ‘fed’ a mix of nutrients needed to
magnitude of the effect of concentration all continue life of the bacteria on which the
improve with further research, this stress biomethanator is based – providing the food and
diagram will be amended. The values shown are environmental conditions that the bacterial
those we believe at present to be the minimum population in the sludge require to grow. In
levels where effects on yeast can be measured. addition, pH is adjusted with tank wash waters
rich in sodium ions (used as NaOH for tank and
line cleaning). Much of this sodium ion is
Biomethanators converted to sodium carbonate (Na2CO3) by the
carbon dioxide in the system. Regardless, the
A more recent addition to the technology of fuel sodium ion, all unused nutrients, and all end
alcohol production is the anaerobic biomethanator. products produced from the biomethanator that
A process flow diagram of the ICM/Phoenix were not converted to methane and carbon
wastewater treatment system is reproduced in dioxide move through the biomethanator
Figure 3 (Anon., 2003). unabated. This results in high levels of sodium
A biomethanator, in part, is a special type of ion and perhaps a series of yet-to-be-identified
fermentor, designed in this case to reduce by compounds in subsequent fermentors. It is
90% the variable BOD/COD levels in evaporator uncertain, as yet, whether the levels of sodium
condensate and other wastewaters. It produces experienced in plants both with and without
methane and CO2, the former of which is used methanators are high enough to inhibit yeasts.
to reduce boiler or dryer operation costs. Sulfur Evidence accumulated in this lab, however, using
can also be removed from treated wastewaters corn mash similar to that made in dry mills,
by conversion to hydrogen sulfide. This is in would indicate that large amounts of sodium ion
some respects an exciting innovation with great would lead to a purging requirement to eliminate
Water reuse in fuel alcohol plants: effect on fermentation 351
Biogas scrubber
Biogas H2S removal
Anaerobic
bioreactor Equalization tanks
Gas
High COD >5,000 mg/L
exchanger TSS <600 mg/L
Tc Blower
Waste water
Recycle FC
pHc tank
Recycle
Te
pump
System
feed
pump
Effluent FM pHc
Sump
Nutrients Caustic
Iron
ion buildup in water that would otherwise re- of waste sodium ion that is potentially dangerous
enter the plant via the methanator. if added to the cook or directly to fermentors.
The levels of sodium needed to inhibit The disposal of sodium ion even through the
laboratory fermentation of corn mash when no dryers via incorporation into DDG may require
other stressful agents are present would appear assessment.
to be between 2500 and 5000 ppm (Figure 4). Anecdotal evidence received from production
This should allow plenty of leeway for the use plants in the past (prior to methanator
of methanator effluent in fermentation (at least technology) indicated that levels of sodium ion
from the sodium standpoint). It should not be above 500 mg/L would inhibit yeast growth and
forgotten, however, that when more than one fermentation; and this was published in the stress
stressful agent is present, the effects are diagram (Figure 2) found in the last edition of
synergistic, and at this point, much lower The Alcohol Textbook (Ingledew, 1999). In the
concentrations of sodium (near 500–600 ppm) past, evidence published by Maiorella et al.
have been seen to have an effect on the yeast. (1984) showed that a sodium concentration of
When biomethanators are not used, it is ~5500 mg/L sodium was required to demonstrate
common practice to use a combination of negative effects on yeast as manifested by >20%
evaporator condensate water, carbon dioxide inhibition. Moreover, Watson (1970) showed
scrubber water, side stripper bottoms water, and that in batch culture, 0.25 to 1.5 M NaCl (5750
well water for the cooking process. Note that to 33700 ppm Na) was needed to affect growth
sodium hydroxide washwater is also unevenly rate and the yield of cells grown on glucose.
recycled to fermentors in volumes averaging Jones and Greenfield (1984) stated that for most
about 0.5% of plant flow rate. It is this volume tolerant strains of yeast, total inhibition of growth
352 W.M. Ingledew
25 0 ppm Na
2500 ppm Na
7500 ppm Na
15
10000 ppm Na
10
0
0 50 100 150 200
Time (hrs)
Figure 4. The effect of sodium ion as a single stress factor on the utilization of sugar during fuel alcohol production
(Hynes et al., 2003, unpublished data).
occurs above 1-2 M (23,000-46,000 ppm Na), action) are shown in Table 12 in comparison to
the level needed for non-sensitive yeasts. For control fermentations where inhibitory levels of
sensitive strains, 0.1 M or 2300 ppm was neither type of inhibitor were present.
needed. These latter values are in similar range
to Figure 4. Insensitivity to sodium is said to be Table 12. Plant data demonstrating the effects of both
via the diversion of energy needed to maintain organic acids and sodium ion on batch fermentations.
an electrochemical gradient of sodium ion Tank Ethanol Glucose Lactic Acetic Sodium Ethanol
against passive or facilitated diffusion into the no. acid acid loss
cells. The effect was not the same with different (% w/v) (% w/v) (% w/v) (% w/v) (ppm) (%)
yeasts measured. Explanation of the discrepancy A 12.4 0.03 0.35 0.02 <290 -
between the numbers that would definitely be B 12.8 0.24 0.29 0.04 <290 -
inhibitory and literature values may be found in H 10.8 4.70 1.0 0.32 <290 14.3
the work of Jones and Gadd (1990), who K 9.5 5.14 1.0 0.41 <290 26.2
indicated that the ratio between sodium and M 11.5 1.40 0.25 0.0 660 8.7
potassium is very important; and that at a pH of N 10.8 1.29 0.17 0.01 720 14.3
5.0 when the Na:K ratio is about 20, potassium
uptake is inhibited by 70%. At pH 4, sodium It can be seen from Table 12 that under normal
uptake becomes negligible; so they indicated that operational conditions (A and B), this alcohol
in an industrial medium containing high levels plant was capable of making over 12% alcohol
of sodium, a pH around 4.0 would be required under conditions where lactic and acetic acids
to overcome the deleterious effects of the were low and sodium was not present at alarming
sodium ion. The optimal concentrations of K+ levels. When organic acids were present at
ion of 2-4 mM may be in part due to the presence inhibiting levels (H and K), ethanol yields
of sodium in mash. dropped, sugar ‘bleed’ at end fermentation was
The differences seen in an industrial plant high, and the acids circulated into other
between organic acid problems (caused by fermentors in backset. This evidence strongly
bacteria and recycle of stillage) and ionic supports the synergistic or at least concerted
problems (probably caused by recycle of NaOH inhibitory effects of lactic and acetic acids. Under
wash waters by stillage and biomethanator conditions of high sodium in Tanks M and N,
Water reuse in fuel alcohol plants: effect on fermentation 353
where the source of sodium was from tank forget, however, that chemicals outside of those
washings recirculated with backset, ethanol made by microbes can also affect the process –
levels fell by 1-2 g/100 mL, and glucose levels sodium ions from tank and line washings are a
rose accordingly. Sodium levels of >500 mg/L case in point.
therefore appeared to significantly lower ethanol The message for the ethanol industry is that
output of this plant. The magnitude of ethanol ‘zero effluent’ plants (those which do not release
loss in these examples demonstrates the significant volumes of water with high BOD or
problems a plant can experience when there is COD or high concentrations of salts) are not
no opportunity to remove recycled thin stillage easily designed – especially when the prime
from the system in order to lower the levels of mode of water recycle is via the fermentation
recycled chemicals and preserve the delicate system. Fermentors (biologically fragile reactor
biological conditions in the fermentor. It is the systems) are susceptible to all chemicals added
author’s opinion that methanators can lead to to them. Engineering to save on water usage is
such elevated levels of ions (but not organic contrary to the preservation of the physiological
acids) unless there are undetermined toxic conditions required by the yeast to ensure
bioproducts of methane-producing microbes complete and rapid fermentation.
that find their way back into the fermentation
system. This is as difficult a situation to monitor
and rectify, as is the present case with continual References
recycle of backset.
Anonymous, 2003. The ICM/Phoenix
biomethanator. http://www.phiosystems.com/
Conclusions: the importance of experience.htm.
elimination or preventing buildup of Bryan, T. 2003. Green Machines. Ethanol
inhibiting chemicals Producer 9(7):12-15, 52, 55-56.
Chin, P.M. and W.M. Ingledew. 1993. Effect of
It has been estimated that the fuel alcohol recycled laboratory backset on fermentation
industry in general loses from 1 to 4% of of wheat mashes. J. Agric. Food Chem.
potential alcohol volume each year. The North 41:1158-1163.
American industry produces at this writing Chin, P.M. and W.M. Ingledew. 1994. Effect of
almost three billion gallons per year. A 1% loss lactic acid bacteria on wheat mash
is 30 million gallons (113 million liters), with a fermentations prepared with laboratory
value of approximately $33 million. A 1% loss backset. Enz. Microb. Technol. 16:311-317.
in a plant such as the one in Table 12 would Ingledew. W.M. 1993. Yeasts for production of
only be 0.12% alcohol. Such a loss is unlikely fuel ethanol. In: The Yeasts Vol. 5. 2nd Edition.
to be noticed due to variability in chemical Academic Press, N.Y. pp 245-291.
quality control assays and the starch levels in Ingledew, W.M. 1995. Chapter 7. The
incoming substrate. A 4% loss over the whole biochemistry of alcohol production. In: The
North American industry is 120 million gallons Alcohol Textbook. Second Edition. (T.P. Lyons,
(452 million liters) with an approximate value D.R. Kelsall, and J.E. Murtagh, eds).
of $132 million. In certain plants (sometimes
Nottingham University Press. Nottingham, UK.
under particular but not infrequent circumstances
pp. 55-79.
as seen in Table 12), losses greatly exceed these
percentages. A major cause of such losses is Ingledew, W.M. 1999. Chapter 5. Alcohol
bacterial degradation of sugars that should have Production by Saccharomyces cerevisiae: a
been converted by yeast to ethanol. All efforts yeast primer. In: The Alcohol Textbook. Third
should therefore be made in plants to avoid Edition.(K. Jacques, T.P. Lyons and D.R.
bacterial action. The action of bacteria or wild Kelsall, eds). Nottingham University Press.
yeasts, or the effects of their end products of Nottingham, UK. pp. 49-87.
metabolism, which remain after heat or other Ingledew, W.M. and C.A. Patterson. 1999. Effect
agents kill the microbes that made them, are of nitrogen source and concentration on the
significant causes of yield loss. One must not uptake of peptides by a lager yeast in
354 W.M. Ingledew
continuous culture. J. Amer. Soc. Brew. Chem. lactic acid on the growth of Saccharomyces
57:9-17. cerevisiae in a minimal medium. J. Indust.
Jones, R.P. and G.M. Gadd. 1990. Ionic nutrition Microbiol. Biotechnol. 26:171-177.
of yeast – physiological mechanisms involved Narendranath, N.V., K.C. Thomas and W.M.
and implications for biotechnology. Enz. Ingledew. 2001b. Acetic acid and lactic acid
Microb. Technol. 12(6):402-418. inhibition of growth of Saccharomyces
Jones, R.P. and P.F. Greenfield. 1984. A review cerevisiae by different mechanisms. J. Amer.
of yeast ionic nutrition. Part 1: Growth and Soc. Brew. Chem. 59:187-194.
fermentation requirements. Proc. Biochem. Patterson, C.A. and W.M. Ingledew. 1999.
19(2):48-60. Utilization of peptides by a lager brewing yeast.
Jones A. and W.M. Ingledew. 1994. J. Amer. Soc. Brew. Chem. 57:1-8.
Fermentation of very high gravity wheat mash Thomas, K.C., S.H. Hynes and W.M. Ingledew.
prepared using fresh yeast autolysate. 2001. Effect of lactobacilli on yeast growth,
Bioresource Technol. 50:97-101. viability and batch and semi-continuous
Maoirella B.L., H.W. Blanch and C.R. Wilke. alcoholic fermentation of corn mash. Journal
1984. Feed component inhibition in ethanol of Applied Microbiology. 90:819-828.
fermentation by Saccharomyces cerevisiae. Thomas, K.C., S.H. Hynes and W.M. Ingledew.
Biotech Bioeng 26:1155-1166. 2002. Influence of medium buffering capacity
Narendranath, N.V., S. H. Hynes, K.C. Thomas on inhibition of Saccharomyces cerevisiae
and W.M. Ingledew. 1997. Effects of growth by acetic and lactic acids. Applied and
lactobacilli on yeast-catalyzed ethanol Environmental Microbiology 68:1616-1623.
fermentations. Appl. Environ. Microbiol. 63: Watson, T.G. 1970. Effects of sodium chloride
4158-4163. on steady-state growth and metabolism of
Narendranath, N.V., K.C. Thomas and W.M. Saccharomyces cerevisiae. J. Gen. Microbiol.
Ingledew. 2001a. Effects of acetic acid and 64:91-99.
Understanding energy use and energy users in contemporary ethanol plants 355
Chapter 25
JOHN MEREDITH
President, Ro-Tech, Inc., Louisville, Kentucky, USA
Results of calculations quantifying the thermal Electrical energy. Kilowatt hours used by the
and electrical energy needed to convert starch plant. Demand and energy cost components of
and extract ethanol from grain are often typical utility electrical bills are not considered
controversial and depend on the assumptions in this analysis. These two components are
behind the calculations. Looking through the measured by utility companies and used to
literature and surveying historical operating costs determine monthly electrical bills. This cost is
for beverage distilleries and ethanol plants also a function of a number of factors outside
reveals a wide range of results reported for the scope of this chapter (plant location, energy
energy use per gallon of ethanol produced. It is market at any particular time, etc.).
also noteworthy that total energy use per gallon
of ethanol produced in beverage plants in the Feedstock. Energy use in ethanol plants is
late 1900s was significantly greater than the dependent on feedstock. For our purposes,
energy use per gallon in today’s new fuel ethanol feedstock will be No. 2 corn at approximately
plants. 15% moisture. Some ethanol plants have
experimented with using ‘wet corn’ at ~30%
moisture as a feedstock. This technique would
Terms defined and assumptions for eliminate the energy need for drying the corn to
energy use calculations 15% moisture at the elevator. However to date,
these efforts have been mostly unsuccessful due
For purposes of this chapter, the criteria below to problems in handling and preparing the corn
will apply to the calculations and conclusions for cooking.
which follow: Other feedstocks require similar energy input
to produce ethanol with some variation. Wheat
Gallon of denatured ethanol. One US gallon of is virtually identical to corn in energy
ethanol containing 4.76% denaturant and in requirements except as affected by moisture
agreement with ASTM D4806. content. Rye and milo (sorghum) require some
additional energy due to higher slurry viscosities,
Thermal energy. Based on natural gas used at which affect heat transfer and pumping. Sugar
1000 BTU/ft3 at standard conditions (60°F, 1 atm cane-based feedstocks use less energy (no
of pressure). dryhouse) compared to grain-based feedstock
plants with DDGS facilities.
356 J. Meredith
Boundaries. Energy use for purposes of this Energy use for co-product handling, drying and
chapter starts at grain unloading at the ethanol storage will be included. Although some older
plant and finishes with product at the point ready beverage plants show a net loss in the production
to be pumped to a waiting truck or rail car. Some of DDGS when comparing costs to revenues,
studies have investigated energy use starting drying co-products to a DDGS product is
with corn in the field and finishing with ethanol considered profitable in newly constructed, more
at a receiving terminal separated from the plant. energy efficient, ethanol plants. Energy use in
A recent study by the US Department of the dryhouse is a major component of the plant
Agriculture stated that ethanol plants deliver 34% energy balance.
more energy to the end user of the fuel than is
used in growing, harvesting and distilling corn Beverage plants versus fuel ethanol plants.
into ethanol. Ethanol releases 90,030 BTU for While this chapter focuses on energy use in fuel
each gallon burned. If 90,030 BTU represents a ethanol plants, the relationships between energy
34% increase, then the total energy required to use and plant system design apply equally to
make ethanol would be approximately 67,000 both industries. When comparing beverage
BTU. Modern plants require approximately plants to similarly designed fuel ethanol plants,
36,000 BTU of thermal energy and 1.1 KWH beverage plants consume more energy per gallon
(3800 BTU/gallon) of electrical energy per owing to the specifics of producing a defined
gallon of denatured ethanol. This totals 39,800 product for human consumption. Distillation is
BTU/gallon, which leaves over 25,000 BTU/ often more complicated in order to produce a
gallon for energy users outside the plant product with controlled amounts of congeners.
boundaries including farm equipment involved In addition, heat recapture techniques, thick
in planting and harvest, energy to transport the slurries and hot water recycling are not as
corn from the field to an elevator for drying and prevalent due also to quality concerns.
then to the ethanol plant, energy needed to dry
corn to ~15% moisture, and energy expended
in transporting ethanol from plant to receiving Understanding thermal and electrical
terminal. energy usage ratios
Despite the focus some place on the overall
energy balance of ethanol production, it is worth BTU (THERMAL ENERGY) PER GALLON OF
remembering that ethanol is not added to fuel ETHANOL
for its energy value. The Clean Air Act of 1990
required adding oxygenates to gasoline in some Modern dry milling mid-sized (40 million gallons
locations to reduce emissions. Ethanol is an per year) ethanol plants are being constructed
oxygenate, and is compared in Table 1 to other in the US based on an expected thermal energy
oxygenates. Methyl tertiary butyl ether (MTBE), requirement of 36,000 BTU (natural gas) per
the oxygenate of choice until a few years ago, gallon of ethanol. This energy requirement is
has been found to be a pollutant, which has led based on denatured ethanol. In other words, the
to increased interest in ethanol as a replacement. ratio is calculated using a mix of 100 gallons of
undenatured plant product with 5 gallons of
Table 1. Comparison of four fuel oxygenates. denaturant (low octane gasoline) yielding 4.76%
denaturant in the mix. Some plants look at the
Oxygen (%) ratio before denaturing, which would be 36,000
BTU/gallon ethanol x 105 gallons denatured/100
Ethanol 34.73 gallons undenatured ethanol = 37,800 BTU/
Methanol 49.90
gallon. Likewise, other plants use ‘steam BTUs’
MTBE (Methanol and isobutylene) 18.15
ETBE (Ethanol and isobutylene) 15.66
instead of ‘natural gas BTUs’. ‘Steam BTUs’
introduce boiler efficiency into the equation as
follows:
Understanding energy use and energy users in contemporary ethanol plants 357
ELECTRICAL ENERGY (KILOWATT HOURS) and recapture, some of the extra water is
PER GALLON OF ETHANOL evaporated twice, requiring over 16,000 BTU
for each gallon of water as it moves through these
Electrical use in a modern mid-sized plant is processes:
expected to be between 0.7 and 1.2 KWH/gallon
ethanol. Using 1.1 KWH/gallon as a • cooking (raising the fluid’s condition from a
conservative average for denatured ethanol, and saturated liquid at ambient temperature to a
calculating the value when denaturing is saturated liquid at approximately 230°F)
removed from the equation, the following is
obtained: • distilling and purifying (changing fluid state
from a saturated liquid (beer) at 90°F to a partly
saturated vapor at 200 proof and the rest to a
1.1 KWH 105 gal denatured ethanol 1.155 KWH
x = saturated liquid at approximately 220°F)
gal ethanol gal undenatured ethanol gal undenatured ethanol
• dryhouse operation (changing conditions from
a saturated liquid at 220°F (stillage) to a
saturated vapor at 160°F to 230°F)
RELATIONSHIP BETWEEN THERMAL AND
ELECTRICAL ENERGY USE
Modern ethanol plants reduce the water and
To further complicate thermal and electrical backset added to the corn meal in the meal slurry
energy use statements, the usage ratios can be tank to a level acceptable in terms of conversion,
dramatically changed by plant design techniques yield and general handling characteristics. There
that transfer energy sources from thermal to is, however, a tradeoff between concentrating
electrical (or electrical to thermal) for certain the slurry to reduce evaporation costs and the
plant processes. For example, the entire subsequent increase in enzyme and chemical
dryhouse evaporator thermal energy load can cost to maintain the proper fluid viscosity and
be shifted to an electrical load by replacing a pH needed for cooking and fermentation.
multiple effect steam driven evaporator with a Keeping the mixture of corn meal and water
mechanical vapor recompression type with in the slurry tank at a ratio of 1 lb corn meal: 2
electric driver. lbs water + backset total provides a concentrated
slurry that saves energy and leads to good
conversion and alcohol yield. This 1:2 mixture
Factors affecting energy use and will yield an ethanol concentration in the beer
efficiency of approximately 14% abv. Diluting the mixture
to 1:2.5 gives an ethanol concentration in the
MASH DILUTION beer of approximately 12% abv. The plant
running at 12% abv will use as much as 20%
Any extra water added to the corn meal slurry more steam in the cooking, distilling and
system before cooking must be heated three dryhouse operation compared to the plant
times in the ethanol plant. Without heat recycle running at beer concentrations of 14% abv.
358 J. Meredith
The new high temperature and alcohol in ethanol plants. The process designer has a
resistant yeasts can operate at 95°F and 18% choice in selecting one or more of these features
abv (or more) in a fermentor, further decreasing with return on investment (operating cost vs
energy requirements. However, slurry handling capital cost tradeoff) the governing factor.
problems and heat transfer problems increase
with these thicker beers. • Preheat the water used in cooking or use
recycled hot water:
NON-ENERGY DEPENDENT YIELD ENHANCERS - use treated evaporator condensate
- use rectifying column bottoms
Alcohol yield per bushel of corn is calculated
- use distillation column overhead condenser
and expressed as follows:
water
- interchange with mash to be cooled after
1000 kg corn at 71% starch = 710 kg starch.
cooking
At 100% efficient conversion 362.8 kg ethanol
would be produced. • Preheat the beer for distillation:
At 90% efficient conversion 326.5 kg - interchange with mash to be cooled after
(718.3 lb) ethanol would be produced. cooking
= 108.8 gallons (6.6 lb/gallon) - interchange with rectifying column bottoms
= 108.8 gallons per 1000 kg corn • Pre-heat or pre-evaporate thin stillage:
= 2.77 gallons per bushel (56 lbs) of corn
- interchange with rectifying column
Typically ethanol plant yields are closer to 2.65 overheads
gallons/bushel after distillation and final - interchange with molecular sieve overheads
processing. If alcohol yields per bushel of grain - interchange with evaporator condensate
increase due to non-energy related procedures,
then thermal and electrical energy ratios will • Recover the saturated liquid flash from the
improve. For example, Alltech’s solid state bottom of the stripping column and return
enzyme RhizozymeTM increases alcohol yield back to stripping column
per bushel of grain by 10% or more versus the • For plants with thermal oxidizers on dryer
use of traditional enzymes. This 10% increase vents
in yield per bushel of grain will have a directly
proportional 10% impact on the thermal and - use regenerative type thermal oxidizer and
electrical energy ratios before denaturing. provide time for clean up (bake out) in the
operating cycle
- for non-regenerative types, use waste heat
CAPITAL IMPROVEMENTS, EQUIPMENT boiler on oxidizer discharge and produce
SELECTION AND ENERGY USAGE RATIOS steam for the plant
• Interchange molecular sieve inlet and outlet
The ratios are dependent on capital
streams
improvements built into plant design that reduce
energy use. Heat recovery, heat interchange, • Use steam turbine drives and use resulting
etc. will reduce energy use ratios. Elimination low pressure steam for distillation and
of the dryer by selling wet distillers grains with dryhouse systems
solubles (WDGS) instead of dry distillers grains
Maximizing use of these techniques allows some
with solubles (DDGS) will dramatically improve
ethanol plants to approach the 36,000 BTU/
the thermal energy ratio. This may not, however,
gallon and 0.7 KWH/gallon ratios noted above.
improve the plant profits depending on the
Without these energy saving improvements,
relative sale price of WDGS versus DDGS.
thermal energy use will increase dramatically
with ratios of 75,000 BTU/gallon and higher a
HEAT RECLAMATION TECHNIQUES possibility.
Listed below are design features presently in use
Thermal energy (100%)
Thermal G
energy
(Natural gas) 70% of
gas
Electrical energy (100%)
E
Electrical E
energy
E 10% of E 4% of E 4% of 1% of E G E G
electrical electrical electrical electrical 1% of
35% 30% of
gas electrical
of elec.
45% of
electrical G
Utilities
Boiler, Chiller,
G = Natural gas (thermal)
Cooling tower,
10%
S = Steam (thermal) 5% 5% 70% Building services
of steam
of steam of steam of steam
E = Electrical
P = Production Steam
10%
of steam
Note: Energy for emission abatement is not included returns to
boiler and 100%
buildings of steam
Distribution of energy use in an ethanol are heated by live steam while fuel ethanol plants
plant use reboilers. Reboilers eliminate beer dilution
associated with live steam and also permit
Table 2 details energy use by subsystem in an recapture of clean condensate heat by the plant’s
ethanol plant. An overview is depicted in Figure boiler system.
1. Energy use distribution in a beverage plant is Purifying the alcohol water mixture after
similar. Data in Table 3 were taken from a new distillation from approximately 192 proof to 200
grassroots beverage plant design in 1990. proof is accomplished in most ethanol plants
using molecular sieve technology, which
Table 2. Ethanol plant energy use by subsystem1,2. requires thermal energy to vaporize the feed to
the sieve and then more energy to superheat the
Thermal Electrical vapor to over 300°F before entering the sieve.
(% of total) (% of total) Some energy can be saved by sending rectifying
column overhead vapor direct to the sieve
Grain receiving and milling Negligible 10-15
Cooking and liquefaction 5-8 1-5
superheater without first condensing the
Fermentation 1 1-5 rectifying column vapor. However, combining
Distillation/molecular sieve 55-60 1-5 the continuous process rectifier with the batch
Product loadout Negligible 1-5 process sieves can present control problems for
Evaporation and drying the ethanol plant operator.
(including emissions) 40-45 30-40 Dryhouses use thermal energy to boil and/or
Utilities 5-7 40-50 dry the stillage, removing condensed water in
Buildings 1 1 the evaporator and vapor from the dryer stack.
1
Depending on equipment selection, various
Based on data collected by Ro-Tech from plants constructed
since 1995.
combinations of steam, natural gas, propane or
2
Emission abatement no included. fuel oil are used to provide the thermal energy.
Large electrical energy use points (large motors)
Table 3. Beverage plant energy use by subsystem. in an ethanol plant are concentrated in utility
areas (chillers, air compressors, cooling towers,
Thermal Electrical boiler fans). If the evaporator is a mechanical
(% of total) (% of total) vapor recompression system, then the dryhouse
becomes a major electrical user.
Grain receiving and milling Negligible 16 The grain handling and milling areas will use
Cooking and liquefaction 7.4 5
Fermentation 1.0 1
significant amounts of power depending on use
Distillation and product of pneumatic systems as well as the complexity
handling 32.1 2 of the grain receiving, storage, transfer and
Evaporation and drying milling systems. However, whereas the modern
(including emissions) 53.5 41 fuel ethanol plant approaches energy usage
Utilities 5 34 ratios of 36,000 BTU/gallon and 0.7 KWH/
Buildings 1 1 gallon, the 1990 beverage plant in Table 3 had
a similar electrical ratio but the thermal ratio was
73,000 BTU/gallon (calculated as one gallon of
The primary thermal energy use points are 200 proof ethanol).
distillation/purification and the dryhouse
followed by the cooking process. Cooking is a
direct live steam injection process using a Overall ethanol plant energy balance
combination of heat, time and enzymes to
liquefy the starch. Indirect heat transfer systems Energy inputs and energy exits for an ethanol
have been tried with little success. Distillation plant are listed in Table 4. Chemical and physical
systems in ethanol plants include stripping and properties (including temperature, pressure and
rectifying columns (sometimes combined into specific heat) of each item must be known to
one column) yielding alcohol at a concentration calculate the proper enthalpy (BTU/lb) associated
approaching the azeotrope of 95.6% ethanol by with entering or exiting energy.
weight. In most beverage plants, these columns
Understanding energy use and energy users in contemporary ethanol plants 361
Fuel Wastewater emissions to land application, outfall (stream), or sewer (from make up water
treatment, from process (evaporator, CIP, other), from boiler, scrubbers and cooling tower
Electricity blowdowns)
Denaturant Air emissions and waste heat to atmosphere (from dryer or oxidizer, tank and vessel vents, cooling
towers, baghouses, from thermal losses through exterior vessel walls and building walls, from
Yeast exhaust air (buildings)
Chemicals
(for CIP, fermentation nutrients)
Chapter 26
JOHN MEREDITH
Backset
in a dry mill ethanol plant Thin stillage
(centrate)
Thin
stillage
Dry Syrup
Methane
Sell wet
Distillers dried grains Sell wet
(no solubles)
Dry Dry
Wet distillers grains
Distillers dried grains with solubles
(WDGS) Solubles
with solubles
(DDGS)
corn meal as the primary product and beverage • whole stillage heat recovery
spirit as a secondary product. Thick stillage
• whole stillage liquid/solid separation
developed into an excellent livestock feed
ingredient. One gallon of thick stillage contains • backset handling and heat preservation
~0.2 lbs. of protein. In the 1950s, farmers • cake handling
typically paid $8-10 per 1000 gallons for thick
stillage. • evaporation
Over time, the small corn milling operations • syrup handling
were gradually replaced by larger plants
producing beverage alcohol as the primary • drying
product. Storing and handling large quantities • dry solids handling and storage
of thick stillage introduced problems as the
distilleries increased in size. Whole stillage • wastewater handling
leaving distillation is a hot, acidic and viscous • emission handling
fluid with limited shelf life (Table 1). Rich in
protein and other nutrients for bacterial growth,
stillage will decompose rapidly depending on Dryhouse reliability is also very much a function
storage conditions. of fermentation efficiency in the front of the
plant. Unconverted starch and sugar will quickly
Table 1. Whole stillage properties at the point of foul heat transfer surfaces in the dryhouse
distillation. causing capacity reduction (poor heat transfer)
Saturated liquid temperature ≥220°F
and downtime.
Total solids 6-16% Secondly, selection of the process system must
Dissolved solids 2-8% address capital cost, revenues and operating
pH 4.5 or less (corrosive) costs such that the dryhouse remains a profit
General composition Fiber, protein, fats/oil, water, center. The capital cost for building the dryhouse
yeast cells, acetic and lactic depends on equipment selection as well as design
acid, other organic capacity. Increasing the backset return to
compounds including cooking and fermentation reduces the
glycerol, minerals evaporative load on the dryhouse and thus the
capital cost needed for construction.
For a variety of reasons drying processes
developed in the 1990s to remove the water
from the thick stillage. Drying eliminates the Selection of equipment and the dryhouse
necessity to maintain a livestock operation. Shelf process
life is retained; and the need to handle an acidic
and corrosive material precluded. The result is In order to maximize the return on investment
a valuable distillery co-product with less storage in dryhouse equipment, present value economic
and transport volume needed. analysis should be completed for the various
dryhouse options. Each co-product option carries
with it operating costs, as noted in Table 2.
Dryhouse design and co-product Equally important to operating costs are
production equipment reliability and longevity. All process
materials in the dryhouse are corrosive and will
Reliability and maximum economic return are attack mild carbon steel. The aggressiveness of
the two key issues that must be addressed in the attack is somewhat dependent on moisture
designing the dryhouse for modern ethanol and content. For example 30% total solids wet cake
beverage alcohol plants. Process systems must is more corrosive than 90% total solids DDGS.
operate with minimum downtime for cleaning Stainless steel 304 is the recommended
and repair. Production must not be limited by material for most all applications. Copper
problems in the dryhouse. To that end, each bearing steel ‘cor-ten’ can be used in some of
subsystem or unit operation listed must be the less corrosive applications such as DDGS
reliable:
366 J. Meredith
storage bins. Heated surfaces in some dryers can The 8% (or more) total solids in the thick
be carbon steel without corrosion problems. In stillage is composed of 3% (or more) dissolved
some cases, particularly if sulfur is present, solids plus the remaining suspended solids.
higher grades of stainless steel must be used. Dissolved solids are defined as the solids that
will pass through a No. 4 Whatman filter paper
using a Buchner funnel and vacuum pump.
Equipment selection by process system Good liquid/solid separation is key to running
an energy efficient, low maintenance, dryhouse.
WHOLE STILLAGE HEAT RECOVERY As will be discussed later in this chapter,
removing water in the dryer uses at least four to
Excepting vacuum distillation systems, whole five times more energy than removing water in
stillage leaving distillation is a superheated liquid the evaporator. The amount of evaporation load
at approximately 220°F. If the first process that can be shifted to the evaporator is almost
following distillation is screening or completely dependent on the concentration of
centrifuging, then flash heat (220-212°F) is often suspended solids in the evaporator feed (dilute
lost. Available heat recovery processes include liquid noted above, also called thin stillage or
heat interchange with fluids to be heated (beer centrate). Liquid/solid separating equipment
to stripper column, cook water) and eductor determines this concentration.
recycle to the base of the stripper column. It Early dryhouse designs included combinations
should be noted, however, that if the dryhouse of screens and presses. One common design
includes evaporators and dryers, cooling of the uses an inclined screen with moving blades or
whole stillage below the atmospheric boiling paddles. Stillage is pumped to the top of the
point causes the need for additional thermal screen. The thin stillage passes through the
energy later in the drying process. screen while the cake is guided down by the
paddles and exits at the end of the screen, usually
into a hopper to be fed into a press. These
LIQUID/SOLIDS SEPARATION techniques result in reduced suspended solids
In order to produce any of the co-products content in thin stillage compared to the thick
(except thick stillage), the thick stillage (or whole stillage. However with these systems, suspended
stillage, see Figure 1) must be separated into a solids in the thin stillage remains at levels (2.5%
somewhat diluted liquid that will be partly used or more) that limit evaporator operation to syrup
as backset and then can be further processed, (evaporator product) levels of 25-28% total
and a bulk solid material typically called wet solids.
cake. Thick stillage (or whole stillage) total solids
content in beverage plants typically runs 7-10% Centrifuges
compared to 14% and higher for fuel ethanol Centrifuges gradually replaced many of the
plants. Beverage plants run ‘thinner’ beer screen/press systems in the mid 1900s. Various
primarily for quality reasons.
Dryhouse design: focusing on reliability and return on investment 367
centrifuge configurations were installed; bowl, where they are ‘plowed up the beach’
however, the horizontal solid bowl type with (narrowing of helical conveyor) and out of
separating forces exceeding 3000 times the the liquid layer into discharge ports.
force of gravity performed best on distillery
c. The clarified liquid is continuously removed
stillage.
by overflowing adjustable weirs (plate dams)
Two companies, Sharples/Alpha Laval and
inside the machine to the discharge point.
Westfalia, have supplied centrifuge equipment
to many ethanol and beverage spirit plants
(Figure 2). It is key to understand that the centrifuge must
There are three continuous actions occurring be fine-tuned at each plant location to reach
in the centrifuge (reference Sharples optimum performance. Various on site
procedures): adjustments are required and may take hours or
days to reach design performance:
a. Slurry is introduced into the revolving bowl
of the centrifuge through a stationary feed • The milling system at the front end of the plant
tube at the center of rotation. By centrifugal affects particle size and centrifuge
force, the solids (more dense than the liquid) performance. Screen analysis from corn mills
are thrown to the wall of the bowl and the should be obtained and compared to other
liquid forms a concentric inner layer in the plants with good centrifuge performance.
bowl. • Depth of pond inside the centrifuge (plate
b. A helical screw conveyor located inside the dams are adjustable).
revolving bowl rotates in the same direction • Bowl RPM – example 2600.
as the bowl but at a slightly lower RPM. This
conveyor moves the solids to one end of the
368 J. Meredith
• Delta RPM – example 4 to 6 RPM (back drive and bulk solids in the dryhouse from process to
determines delta RPM between bowl and process. Selecting bulk solids conveying
conveyor). equipment is the most challenging. Two methods
are common: mechanical conveying and
• Beach angle and length.
pneumatic conveying.
• Conveyor design including wear part Mechanical conveyors including screw
protection (tiles and carbide wear parts). conveyors, belt conveyors, drag conveyors and
bucket elevators are in service in dryhouses.
Plant operators should monitor total, suspended Selection of the type of conveyor is dependent
and dissolved solids for the process streams on several factors:
entering and leaving the centrifuge. Low
suspended solids in the centrate is the first • Length of run
objective. Values approaching 1% have been • Elevation change
obtained depending on the plant. Low moisture
• Properties of the bulk solids (moisture
in the cake is the second objective. Cake from a
content, bulk density, angle of repose, fouling
properly selected centrifuge typically runs 30%
tendencies)
to 31% total solids.
• Flow rate
Centrifuge installation
Manufacturers provide detailed and extensive Screw conveyors are commonly used for all
installation instructions. However a few items applications with the exception of high flow rate
are worth noting. Vents must be provided to systems. In these high flow systems, cost is a
atmosphere at each of the two discharge points. factor and often belt or drag conveyors are less
Considerable amounts of flash vapor discharges expensive than large stainless steel screw
from the centrifuge. The liquid will cool conveyors. Bucket elevators have limited
somewhat (centrifuges are big ‘fans’) from application in dryhouses due to the properties
entrance temperature. Secondly, the supporting of most of the bulk solids, which lead to fouling
structure for the centrifuge must handle static of the buckets. However, bucket elevators have
and dynamic loads and must be designed by a been used successfully in handling 90% total
licensed structural engineer. These large, heavy, solids material.
fast-rotating machines can impose dynamic Pneumatic conveying becomes a viable
loads in excess of three times the static load. method when conveying distances are excessive
Centrifuges are capital cost intensive and high and/or elevation change is large, however,
maintenance machines. Spare parts including a certain design issues must be addressed:
spare bowl assembly are necessary for plants
that run continuously. Lastly, it is interesting to • Limit the applications to low moisture bulk
note that screen/press systems will give a lower solids. DDGS, solubles and DDG (no
cake moisture (up to 38% total solids) than solubles) are all successfully transferred
centrifuges. However, the high thin stillage pneumatically.
suspended solids for the screen/press system • Dryhouse bulk solids generally contain fiber
shifts loads to the dryer and raises overall and are abrasive. Use flat back ells and
dryhouse operating costs in most cases. conservative air:solids ratios.
Depending on the application and selection of
co-product types, the centrifuge will not always • Pneumatic conveying introduces the need for
be the low cost life cycle option, however. air/solids separation at the end point. This
requires the use of cyclones, bag houses and
vent filters and introduces an air emission
BULK SOLID CONVEYING AND STILLAGE point.
PUMPING These considerations may eliminate viable
pneumatic conveying at some locations.
Equipment must also be selected to move liquids
Dryhouse design: focusing on reliability and return on investment 369
Stillage pumping systems in the dryhouse do taking advantage of the enthalpy of evaporated
not require exotic, extremely costly equipment. vapor at reduced pressure, or by vapor
Thick stillage and thin stillage can be pumped compression (thermal or mechanical) and then
with centrifugal pumps using an ‘open impeller’. using this evaporated vapor for further
Syrup should be pumped with a positive evaporation.
displacement pump (lobe type is suitable). Evaporators find applications in many
All piping systems in the dryhouse should be industries. Only certain types of evaporator
free of internal dead spots. Reduced port ball configurations however, have been used
valves should not be used. Blind Ts should be successfully by ethanol and beverage spirit
avoided. Butterfly valves, ball valves, full port plants. Choices must be made among:
mag-meters and ‘DU-O’ check valves are all
suitable. Rupture discs are preferred to relief • Vessel configuration (see Figure 3).
valves due to fouling.
• Forced circulation vs. natural circulation.
• Rising film tubular vs. falling film tubular.
EVAPORATION
• Plate equivalents of tubular evaporators.
Thin stillage containing 1% or more suspended • Multiple effect vacuum with thermal
solids and 3% or more total solids can be recompression vs. mechanical vapor recom-
concentrated in an evaporator to yield a product pression turbine drive or mechanical vapor
(syrup) that ranges from 25 to 50% total solids. recompression motor drive (Figures 4 and 5).
Evaporators allow reuse of thermal energy by
• Combination evaporator/distillation systems.
Separator
Long tube
vertical
Tubes
Separator
Short tube
vertical Tubes
Separator
Tubes
Horizontal with
remote separator
Separator
Vertical with
remote separator
(Perferred configuration for Tubes
falling film systems in dryhouses)
STEAM
CONDENSER
T.C.
STEAM
Condensate
Condensate Condensate
Water
SYRUP Condensate
FEED
STEAM
TUBE TUBE
BUNDLE BUNDLE
Separator Separator
Condensate Condensate
COMPRESSOR
FEED SYRUP
DRIVER
Rising film and plate equivalents of tubular four effect units may approach 6 to 7 lbs vapor
evaporators have failed and/or caused excessive per pound of steam. The cost of the fourth effect
maintenance in ethanol and beverage spirit must be justified by the increased thermal
plants and should be avoided. Forced circulation efficiency. Likewise, depending on heat transfer
falling film tubular vessel designs are the most surface provided, a large mechanical vapor
user-friendly designs for plant dryhouses. The recompression unit might evaporate 100,000 lb/
key to falling film systems is the design of the hr of vapor with a power requirement of 1000
distribution head at the top entry to the heat kW. Less efficient units would have a lower first
exchanger. Each tube should receive the same cost.
design GPM. Multiple effect and recompression
systems are equally acceptable for the dryhouse.
The final choice depends on life cycle DRYERS
economics.
Dryhouse reliability and operating cost/return
Additional evaporator design considerations issues hinge as much on the selection of the
dryer as any other system in the facility. Dryers
Ethanol plant evaporators are often integrated are major users of energy and the largest source
with the distillation and molecular sieve systems of air emissions in many ethanol and beverage
to save energy. spirit plants. Further, dryer failure will shut the
plant down and in some cases dryer failures are
Thin stillage reboiler. The first effect of the not quickly remedied. Selection of the dryer
evaporator becomes the reboiler for the stripping depends on various process and operating
column. This saves energy compared to the use considerations:
of live steam on the stripper and is more user-
friendly (less fouling) than using a whole stillage • Energy available (gas, oil, steam).
reboiler on the stripper.
• Cost of energy options ($/BTU).
Hot vapor condenser as first effect of evaporator. • Co-product to be dried (DDGS, syrup).
The first effect of the evaporator replaces one • Capital cost/operating cost tradeoffs (dryer
of the condensers off the rectifying column or efficiency vs. dryer cost).
off the molecular sieve – also an energy saver.
• Corrosion.
Optimum run cycle should be balanced vs. syrup • Operating time between shutdowns for
concentration cleaning.
Evaporator cleaning is expensive, causing lost • Preheat methods for air including heat
production and the need for energy input without recovery (consider flue gas, steam
product output. Caustic and other chemical costs condensate, recycle exhaust).
are also incurred. Each plant should develop the
optimum evaporator run time before cleaning • Environmental issues (dryer emissions)
by balancing syrup concentration, operating - scrubber
costs and cycle times between cleanups. Overall - thermal oxidizer
costs to produce 35% syrup may exceed costs
- regenerative thermal oxidizer
to produce 30% syrup because of shortened
cycle times at 35% syrup. - remove particulates, VOCs, odor
Energy costs vary depending on evaporator
design; and as with other plant systems, must Types of dryers
be analyzed based on capital cost/operating cost
trade offs. Typically, a triple effect thermal As with evaporators, ethanol and beverage spirit
recompression evaporator will evaporate 4.1 lbs plants have used different dryer types dating to
vapor per pound of steam consumed. However, the early 1900s (Table 3).
372 J. Meredith
Table 3. Dryer types, energy sources and co-products Each vendor uses in-house design techniques to
produced. lift and move the grain through the rotating drum.
Type/description Energy source Typical co-product Efficiencies without air recycle average 0.6 to
0.7 lbs evaporated per pound of steam. Inlet air
Steam tube Steam DDGS
(rotating drum) temperatures may approach 300°F with exit
Steam tube Steam DDGS temperatures of approximately 200°F.
(rotating tube bundle)
Direct fire Gas, oil DDGS Steam tube rotating tube bundle
(rotating drum)
Flash dryer Steam, gas, oil DDGS This is a variation of the steam tube rotating
Drum/dehydrator Steam Solubles drum. In the rotating tube bundle, the drum is
Spray drier Steam, gas, oil Solubles fixed and the tube bundle rotates. More heat
transfer surface is located in a smaller drum
compared to the rotating drum. Efficiencies are
Steam tube rotating drum similar to the rotating drum. Feed conditioning
to prevent fouling is critical to the rotating tube
This dryer is one of the original designs for bundle type because of the close clearances
distillery dryhouses and many installations between the tubes.
around the world exist today. Steam tubes
(typically 3 or 4 inch pipe) are fixed and located Direct fire rotating drum
in one or more concentric rings inside the drum.
Drum diameters may exceed 10 feet, and drum Direct fired rotating drum dryers also have a long
lengths are often 60 feet or longer. Steam enters history of service in the ethanol and beverage
a distribution head and condensate leaves from spirit industries. These dryers have a fixed
the same end. As with all dryers, a well mixed, furnace section for heating the drying and
properly conditioned feed is necessary to prevent conveying air stream, followed by the feed
fouling (see following discussion on feed introduction point, which is in turn followed by
conditioning). the rotating drum. Drum designs vary, with both
single pass and multiple pass units available.
FAN
HEAT
RECOVERY WET
(OPTION)
CYCLONE
PRODUCT
DRYING
DRUM DRY PRODUCT
EXHAUST
FAN STACK
BURNER
KILN
HEAT
EXCHANGER HEAT
RECOVERY
(OPTION)
Unlike steam tube dryers, direct fire rotating between the hot drying and conveying air stream
drum dryers must include cyclones to separate and the grain. Other features are similar to the
the dry product from the conveying air stream. direct fire rotating drum including: fixed furnace,
Modern versions of these dryers include air feed entry point, cyclones and possibility of air
recycle to capture and reuse energy. recycle. Some vendors use the term ‘ring dryer’
Depending on design, inlet air temperatures instead of flash dryer.
may vary from 600 to 1200°F. Exit temperatures
are typically 200°F. The elevated inlet Drying syrup into solubles
temperatures, compared to steam tube systems,
will result in additional dryer stack emissions. Drying evaporator syrup is more challenging
Efficiencies are similar to steam tube systems, than drying the mixed wet cake/syrup
again depending on design. One direct fire combination. The syrup is a viscous liquid prone
rotating drum drying system by Ronning to foul contact surfaces. In contrast, the mixed
Engineering Co., Inc. shows energy use of 1350 wet cake/syrup combination is a solid and can
BTU per pound of water evaporated. A typical be conditioned with DDGS recycle before
Ronning Dryer System is illustrated in Figure 6. drying. In addition, dry solubles are extremely
A schematic of a direct fire rotating drum system hygroscopic and difficult to handle.
with air recycle and incineration of emissions Two technologies, the drum dehydrator and
by ecoDry is illustrated in Figure 7. the spray dryer, have been used to produce dry
solubles. In drum dehydrators, the syrup is
Flash dryers deposited as a film on the outside of a heated
(steam) rotating steel drum. The resulting dry
Flash drying technology utilizes a stack or duct sheet is continuously removed by a blade and
system instead of a rotating drum for contact then milled or ground into a powder. Drying
374 J. Meredith
syrup with a spray dryer is not common practice. cooking system. Beverage plants do not recycle
Difficulties include reaching the very low due to quality concerns. The recycled condensate
moisture required for solubles (often 4-5%) must be conditioned with chemicals and/or make
without fouling the dryer. The resulting product up water to prevent contamination and
can be hydroscopic and difficult to handle. unacceptable pH levels.
However, Alltech has successfully used this Whole stillage, thick stillage and thin stillage
technology and continues to improve the (evaporator feed) all have a BOD (Biological
process at two plants. Oxygen Demand) of over 20,000, compared
with approximately 250 for domestic waste.
Feed conditioning for dryers Evaporator condensate BOD will range from
1000 to as high as 5000. The condensate is acidic
As noted above, wet cake moisture is typically (usually lactic and acetic acids) with a pH of 4.0
70% and syrup moisture may be as high as 75%. or less. EPA direct stream discharge laws usually
Feeding this mixture without conditioning to limit BOD to 15 to 30.
most dryers will cause rapid fouling and in some Various wastewater treatment systems are in
cases may lead to dryer fires. The dryer feed place in most ethanol plants and virtually all
should be conditioned and the moisture content beverage plants to condition some or all of the
decreased by recycling some dry product and wastewater to acceptable levels. Requirements
mixing it with the wet cake and syrup. Feed vary from location to location and also depend
moisture contents of 25 to 30% will enhance on access to municipal sewers. When municipal
performance in most DDGS dryers. The sewers are an option, dryhouse wastewater is
exceptions are drum dehydrators and spray pre-treated before discharge to the sewer.
dryers, which are fed syrup ‘as is’. Wastewater pre-treatment requirements also
Selecting the wet cake/syrup/recycle mixer is vary, but a reduction in BOD to 200 would be
again a trade-off between capital cost and typical.
operating performance. Systems that have been Wastewater systems must address the
used include: following:
• U-Trough screw conveyor with mixing blades. • BOD reduction.
• Pugmill/double screw conveyor with mixing • Suspended solids reduction.
blades (horizontal).
• Temperature.
• Vertical enclosed screw conveyor.
• Acidity.
• Solid/liquid continuous in-line blenders.
• Reduction of other regulated contaminants.
• Ribbon blenders.
• Sludge handling and disposal.
Considering mixing efficiency, horsepower and
corrosion/erosion control, the double screw Wastewater treatment options include:
pugmill is a good selection.
• Aerobic digestion with lagoons or tanks.
• Sequential batch reactors.
WASTEWATER AND EMISSIONS
• Anaerobic conversion to methane.
Wastewater
• Land application.
Evaporator condensate (water removed from
evaporator feed plus in some cases mixed in
steam condensate) and CIP (clean-in-place) Dryhouse air emissions
systems are the two primary sources of Sources for air emission in dryhouses include
wastewater from ethanol and beverage spirit the dryer stack, cyclones, bag houses and tank
dryhouses. Many ethanol plants recycle all or vents. Until about 1999, the US Environmental
part of the evaporator condensate back to the
Dryhouse design: focusing on reliability and return on investment 375
Chapter 27
Ethanol production and the modern livestock feed industry: a relationship
continuing to grow
KATE A. JACQUES
North American Biosciences Center, Alltech Inc., Nicholasville, Kentucky, USA
The rapid increase in production of ethanol in The primary market for DDGS today is
North America is paralleled by the rise in co- livestock feeding. Of the >3.5 million mT of
products produced, particularly those from the DDGS produced in North America (figures for
dry milling ethanol process. While there is the year 2000), around 800,000 mT are exported
undoubtedly great potential for new products and the remaining 2.7 million mT are used in
based on various fractions of stillage, true US and Canadian animal feeds (Markham, 2003;
realization of that potential will take both time Figure 1). If ethanol production from dry mills,
and considerable industrial creativity. At present, and consequently DDGS, production is to double
the immediate issue is finding ways to maximize as predicted in a few short years, then clearly
return for the co-products currently produced, we must actively pursue ways to increase its
which is mainly distillers dried grains with utilization if DDGS is to maintain its valuable
solubles (DDGS). role in the economics of ethanol production.
8
Exported DDGS
Projected
7
North American DDGS production
Million metric tonnes
0
1990 1995 2000 2005
Figure 1. Production and exports of DDGS in North America (adapted from Markham, 2003).
378 K.A. Jacques
There is a growing amount of animal nutrition ‘New generation’ DDGS vs old feed
research in recent years that illustrates that DDGS industry perceptions
could play a much larger and more useful role
in modern food and companion animal nutrition. Much of the feed industry still perceives DDGS
Work with dairy and beef cattle, pigs, poultry as an ‘alternative protein source’ commodity
and even fish shows that the ‘new’ DDGS has with limitations in quality, consistency or
higher nutrient content, greater flexibility in composition that restrict its use as a feed
formulations, can be used at higher inclusion ingredient. Much of this impression, however,
levels and in general provides a cost-effective stems from the DDGS produced using
means of meeting the goals of modern nutrition. technologies common 20 to 30 years ago, but is
While the impact of this work is growing, two perpetuated by industry references that provide
key issues must be addressed if DDGS use in ‘book values’ based on ‘old DDGS’. For
the feed industry is to increase significantly. One example, energy content in DDGS from modern
is a greater understanding on the part of the ethanol plants is higher than in ‘old’ DDGS; and
ethanol industry of what today’s feed industry both are higher than values published in the most
values in ingredients. The second is a change in recent edition of Nutrient Requirements of Swine
the feed industry perception of DDGS. (Table 1, Shurson et al., 2003). Similarly, values
The contemporary animal feed industry is very for total protein and certain limiting amino acids,
different from the industry in the 1970s and particularly lysine, are notably higher in new
1980s when much of the initial nutrition research DDGS (Table 2).
with DDGS was conducted. At that time, the Dated reference values underscore the need
primary approach to DDGS use was a for suppliers of DDGS to provide complete
straightforward replacement of other dietary analytical data. Diet formulation software
protein sources. Today both the animal and the incorporates all nutrient fractions of each
context in which it is fed have changed. Modern ingredient; and reliance on ‘book values’ of any
livestock genetics have given us higher sort by nutritionists is avoided. Part of the quality
performance expectations for meat, milk and egg control program for feed manufacturers is
production; nutritional knowledge has grown, keeping ingredient composition updated such
and consequently diet formulations have gotten that fluctuation in nutrient content of an
more complicated. Feeding goals have changed ingredient will not result in a change in animal
as well. There is a focus on animal health aspects performance.
of nutrition as well as performance, a general The lighter or ‘golden’ color of DDGS from
move away from animal protein sources in modern plants is desirable in that it indicates less
feeding programs in many parts of the world heat damage to nutrients, which particularly
and the environmental impact of food animal affects amino acid digestibility. The dark color
production is a critical consideration. Distillery of earlier DDGS often indicated overheating and
co-products can have an impact in all of these carmelization of nitrogenous components and
areas; and getting that information into the sugars that escaped fermentation. In modern
marketplace can increase overall co-product processes less sugar escapes fermentation,
usage to mutual benefit. stillage viscosity tends to be lower, and more
Table 2. Comparison of amino acid composition of DDGS (dry basis) in old generation DDGS, new generation DDGS,
and values published in NRC (1998)1,2.
Old generation DDGS New generation DDGS DDGS NRC (1998)
Arginine, % 0.92 (18.7) 1.20 (9.1) 1.22
Histidine, % 0.61 (15.2) 0.76 (7.8) 0.74
Isoleucine, % 1.00 (9.1) 1.12 (8.7) 1.11
Leucine, % 2.97 (12.4) 3.55 (6.4) 2.76
Lysine, % 0.53 (26.5) 0.85 (17.3) 0.67
Methionine, % 0.50 (4.5) 0.55 (13.6) 0.54
Phenylalanine, % 1.27 (8.1) 1.47 (6.6) 1.44
Threonine, % 0.98 (7.3) 1.13 (6.4) 1.01
Tryptophan, % 0.19 (19.8) 0.25 (6.7) 0.27
Valine, % 1.39 (2.3) 1.50 (7.2) 1.40
1
Values in parentheses are coefficients of variation among ethanol plants.
2
Shurson et al., 2003
water can be removed by evaporation, which livestock diets. For corn, which is normally 9-
reduces heat input in the dryhouse. Ergul et al. 10% crude protein, resulting DDGS become
(2003) found that color was a good predictor of 28-30% (or more) crude protein. Though the
amino acid digestibility in chickens and that a energy in starch is removed during fermentation,
lighter, more yellow color was associated with the fat remains. Concentration, added to the fact
higher lysine, cystine, and threonine that fat contains three times the energy as starch,
digestibility. brings DDGS energy content nearly equal to that
of corn. This indicates that DDGS can be used
to replace a portion of the corn or other energy
Nutritional value of DDGS grain. The fibrous fraction of corn is concentrated
to 7-8% crude fiber in corn DDGS. Phosphorus,
PROTEIN, ENERGY AND PHOSPHORUS notably low in corn (0.28%) is raised to 0.7-0.8%
CONTENT in DDGS. More importantly, the phosphorus in
DDGS becomes ‘available’ for digestion
Fermentation of the starch in cereal grains, following ethanol fermentation. Pigs, poulty and
which constitutes about 66% of the weight, other monogastric species lack the phytase
concentrates remaining nutrients (fat, protein enzyme needed to hydrolyze phosphorus from
and minerals) by approximately a factor of 3. the phytin form in which it exists in cereal grains
In nutritional terms, this makes DDGS a good (Figure 2). General composition of grain and
source of protein, energy and phosphorus, the DDGS for corn, wheat and grain sorghum are in
three most expensive nutrients added to Table 3.
HO
Figure 2. Structure of phytin phosphorus in cereal grains. CAN WE INCREASE USE OF DDGS IN DAIRY
AND BEEF CATTLE?
Table 4. Average, minimum, maximum, and mean Table 5. Predicted nutrient content and coefficients of
nutrient values (dry matter basis), of grains and solubles variation in DDGS from plants with the least or most
fractions from six Minnesota ethanol plants1. variation in protein, fat or phosphorus content1,2.
Dairy cattle
Beef cattle
Pigs
Poultry
Figure 3. Utilization of DDGS by the North American livestock feed market (Markham, 2003).
382 K.A. Jacques
converting it to microbial protein. On the other have been suggested as the reason that the
hand, some feed protein sources are higher in feeding value (higher performance and
biological value (amino acid profiles useful to improved efficiency) of the distillery co-product
the animal) than microbial protein and of greater has been observed in several trials to be higher
benefit to the animal if they ‘by-pass’ rumen than equal amounts of the original grain
fermentation. Because it has already been (Klopfenstein, 1996; Klopfenstein and Grant,
through fermentation, over 50% of DDGS 2001).
protein is ‘by-pass’ protein (Schingoethe, 2001), The fiber in DDGS from corn results in >40%
which compares favorably with other available neutral detergent fiber (NDF), which is the sum
sources (Table 6). of cellulose, hemicellulose and lignin. While
particle size is too small for the NDF in DDGS
to provide the physiological effects needed by
Table 6. Comparison of the crude protein and rumen- cattle from fiber, over 60% of the NDF is in the
undegradable protein contents of various feedstuffs. form of hemicellulose, and therefore
nutritionally useful.
Crude protein Rumen undegradable
(%) protein (%)
Wet distillers grains
Forage sources Cost of transport and rapid spoilage make use
Alfalfa hay 19 28
Alfalfa silage 19 23
of wet distillers grain practical only where large
Corn silage 9 31 concentrations of cattle are relatively near an
ethanol plant, but where such opportunities exist,
Distillery co-products the reduced energy requirements for drying
Corn DDGS 28 54
make the product attractive for both buyer and
Oilseed meals seller.
Soybean meal 47 35 Apart from moisture content, the nutrient
Canola meal 42 28 content of DDGS and wet distillers grains are
Sunflower meal 26 26
very similar, provided that the same amounts of
Animal sources solubles are added. Interestingly, some of the
Fishmeal 67 60 studies with growing cattle have suggested an
Feather meal 89 71 advantage in growth performance of feeding wet
Blood meal 92 82
vs dry distillers grains. Speculation has been that
Meat and bone meal 42 61
normal drying may reduce energy value to some
extent, however this has been difficult to prove
(Klopfenstein, 1996; Klopfenstein and Grant,
While substitution of DDGS for other protein 2001).
ingredients is most common, use of DDGS to A very real concern when feeding wet grains
replace energy sources in ruminants may is molding and the associated potential for
provide benefits to rumen/animal health as well mycotoxin contamination, which can both
as boost dietary inclusion rates. Starch overload reduce performance and cause animal health
in the rumen of a dairy cow or beef finishing problems. Addition of preservatives at the
animal is a common cause of acidosis, low milk alcohol plant may extend the shelf life of wet
fat content and variation in feed intake and grains, especially during the summer.
performance. Distillers grains (wet or dry) are Experiments with CakeGuard™ (Alltech Inc.),
similar in energy content to the original grain, a preservative based on a mixture of organic
but the energy value is contributed primarily by acids, have shown that use rates of 2 kg/t
concentration of the fat along with some prevented mold growth for two weeks when
digestible fiber. In addition, bypass protein has stored at room temperature. In the field,
a higher energy value than the same protein CakeGuardTM is being used with success at 1 kg/
degraded in the rumen. Lower incidence of tonne to extend stability of wet cake beyond the
acidosis and the impact of energy level and form typical four days expected.
Ethanol production and the modern livestock feed industry 383
1. 00 E+ 07
Control 1 kg/T 2 kg/T 5 kg/T
1. 00 E+ 06
Mold (CFU/g)
1. 00 E+ 05
1. 00 E+ 04
1. 00 E+ 03
1. 00 E+ 02
1. 00 E+ 01
Week 1 Week 2
Figure 4. Effect of CakeGuardTM addition rate on mold growth in wet distillers grains weeks 1-5 through storage.
PIGS, POULTRY AND DISTILLERS GRAINS WITH The higher levels of inclusion listed in Table 7
SOLUBLES assume that diets are formulated based on the
digestible amino acid (as opposed to total
The pig sector arguably represents the largest protein) and available phosphorus content of diet
potential for increased use of DDGS. Until ingredients (Shurson et al., 2003). This is an
recently, very little DDGS has been used in diets important point to note in terms of marketing
fed pigs. The primary reasons include variability DDGS. Nutritionists, particularly those feeding
in quality and nutrient content among sources, pigs and poultry, need to know the amounts of
poor amino acid digestibility due to overheating digestible amino acids in an ingredient, not
during drying, concerns about the high fiber simply crude protein content.
content, and cost competitiveness with the corn, DDGS, like corn, is fairly low in the amino
soybean meal and dicalcium phosphate acid lysine, which is the first limiting amino acid
(inorganic phosphorus source) they replace for growth. Modern ethanol plants, with process
(Shurson et al., 2003). Research with all classes controls that help maximize starch extraction,
of pigs at the University of Minnesota and other fermentation efficiency, and other upstream
midwestern universities has demonstrated operations, produce less variable DDGS with
conclusively that the higher energy, digestible higher levels of more digestible lysine (see Table
amino acid and available phosphorus content 2). As processes further improve and quality
of DDGS from modern ethanol plants allows control of co-products reaches higher standards,
inclusion at levels in excess of 20% in many an increasing number of pig feed manufacturers
diets (Table 7). will have the confidence in DDGS needed to
boost inclusion levels from the currently
Table 7. Use rate recommendations for high quality conservative 5 and 10%.
DDGS in diets for pigs.
Production Phase Maximum % of diet
Nursery pigs (>7 kg) 25 ENVIRONMENTAL IMPACT
Grow-finish pigs 20
Developing gilts 20 Increasing use of DDGS in pig diets can
Gestating sows 50 ultimately have a large and beneficial
Lactating sows 20 environmental impact. Owing to the indigestibility
Boars 50 of the phosphorus in corn, the phosphorus
Shurson et al., 2003
content of pig manure is a major concern.
Phosphorus carried in runoff from land used to
384 K.A. Jacques
spread manure threatens surface waters, causing stored at customer farms. One adsorbent is based
algae blooms in streams. Regulations limiting on a yeast cell wall component modified to
the amounts of phosphorus that can be added to maximize the range of toxins adsorbed, which
a given land area are being put into place in are thereby prevented from harming the animal
virtually all major agricultural regions, which in (Table 8).
turn can limit the number of animals reared and In recent years, the use of in-feed antibiotic
the overall profitability of the farm. Given both growth promoters has markedly declined. Both
the expense of adding inorganic phosphorus to a consumer issue and a potential health concern
animal feeds and the desire to reduce the because it might contribute to resistance of
environmental impact of manure, the high pathogens to similar drugs used in human
available phosphorus content of DDGS has medicine, feed manufactures are removing such
much to offer pig feed manufacturers. compounds from feeds and ‘antibiotic-free’ is a
marketing claim made for some food animal
products as well as animal feeds. Further, the
POULTRY exclusion of all or some antibiotic growth
promoters (virginiamycin is a relevant example)
Use of DDGS in chickens and turkeys has is a legal requirement in some arts of the world.
historically been limited, however its value in The possibility of any measurable antibiotic
providing ‘unidentified growth factors’ that residue, no matter how small the amount, will
influence hatchability and efficiency have long increasingly restrict market opportunities for
been recognized. While the fiber content of ethanol co-products.
DDGS limits its use to a greater extent in poultry
than in other species, research with the product Table 8. Mycotoxin binding efficiency of Mycosorb1.
from modern plants has shown that when high
quality DDGS is used in diets formulated on an Amount bound2
available amino acid basis, performance and (%)
economic advantages are the result. Work
demonstrating these responses in turkeys has Total aflatoxins 85.23
resulted in an increasing amount of DDGS being B1+B2+G1+G2)
Zearalenone 66.66
included in turkey diets in Minnesota. Citrinin 18.41
T2-Toxin 33.39
Fumonisin 67.00
Contaminants: mycotoxins and
1
antibiotics Minus values of control group
2
Alltech Inc.
A quality concern for use of high levels of
distillery co-products is the threat of mycotoxin
contamination. Mycotoxins present in the grain Replacing animal proteins: opportunities
mashed for fermentation are not destroyed for export?
during fermentation and distillation, since heat
stability varies among toxins. In addition, Good quality vegetable protein sources are in
moisture conditions in storage are critical for increasing demand across the world. While it
prevention of surface mold. As long as DDGS can be said that this would be true in any case,
is dry to <15% moisture and is cool entering one of the changes in animal feeding programs
storage, the potential for mold growth is no in Europe following BSE was a reduction in use
greater than in other feed ingredients. It should of animal by-product ingredients. Higher prices
be noted, however, that establishing procedures and lower availability of fish meal have
for mold prevention (along with mycotoxin increased use of soybeans as protein sources in
analysis) is something purchasers increasing poultry feeds in both Europe and South America.
look for from ingredient suppliers. Commercial While shipping costs, the acceptability of by-
feed mixes often include mycotoxin adsorbents products from genetically modified crop varieties
as a precaution against mold growth in feed and other issues affect export patterns, the trend
Ethanol production and the modern livestock feed industry 385
Chapter 28
KARL A. DAWSON
North American Biosciences Center, Alltech Inc., Nicholasville, Kentucky, USA
Introduction
Traditional ethanol production systems are production of a single product, ethanol. The
designed to efficiently produce alcohol from simple value of the plant was largely defined on the
sugars derived from grains and products from sugar basis of its ability to efficiently produce alcohol.
processing. In their natural form, these sugars are There was little consideration given to the ‘by-
in the form of polysaccharides that must be products’ that are produced, even though these
subjected to degradative processing before they materials can account for more than a third of
can be subjected to the metabolic activities of the organic materials entering the plant. The
microorganisms during active fermentation. Since markets for these materials have often been
polysaccharides are generally derived from considered secondary to that of ethanol
complex plant materials, no single processing step produced. With increased emphasis on economic
or enzyme treatment will release all of the sugars. accountability and the need for controlling waste
As a result, the conversion of plant material to streams and environmental impacts, it is now
useful sugars is generally incomplete, and a large becoming critical to update some of the basic
proportion of the material initially entering the processes used to produce ethanol. Advances in
ethanol processes is not converted to ethanol and biotechnology make these updates particularly
remains unfermented. One of the key challenges timely, since our new understanding of
for the modern alcohol industry is defining systems biochemical systems and specific micro-
for economically converting these unfermented organisms has provided us with many new tools
materials into useful or value-added co-products that can be used to upgrade the basic processing
that can contribute to the profitability of the systems used in ethanol production.
fermentation plant. This chapter will describe some The term ‘biorefinery’ has been used to
concepts that can be used to extend the describe chemical production facilities that use
bioprocessing capabilities and flexibility of alcohol biological systems (microbial fermentations and
production plants to make them more sustainable enzymatic conversions) to efficiently catalyze
and economically viable. key chemical transformations (Abbas and
Cheryan, 2002). As we have moved into the age
The biorefinery concept of molecular biology and metabolic engineering,
we now see that innovative applications of
In the past, processes in traditional alcohol biology can have a tremendous impact on both
distilleries have been optimized for the efficient the economic and the practical engineering of
388 K.A. Dawson
the ethanol production process. In broader with minimum amounts of waste (Figure 1).
terms, biorefineries can be seen as very versatile Instead of thinking of traditional processes that
facilities that are not bound to the production of produce only one or two products from a single
an individual product, but can instead use a substrate (Figure 2a), it is important to think
variety of substrates and many different about the value of systems that produce many
processes to produce a wide range of products different products (Figure 2b) from a variety of
INPUT OUTPUT
• Corn • Alcohol
• Grains • CO2
• Molasses • Glycerol
• Corn stover • Chemicals
• Paper waste • Solubles
• Sawdust • Yeast
• Bagasse • Light grains
• Fiber crops
• Municipal waste Feed and fiber
(animal feeds)
Figure 1. A biorefinery.
Saccharified Feedstock
(1 ton)
Distillation Ethanol
(630 lb)
Clarification and
Drying
Animal Feed
(710 lb)
Figure 2a. A bioprocessing system that focuses on the efficient production of ethanol.
Biorefineries: the versatile fermentation plants of the future 389
Feedstock
(1 ton)
Water
Distillation Ethanol
(630 lb)
Clarification
Separation/refining
Secondary fermentation/
Plant oils Carbon dioxide
enzymatic treatment
(100 lb) (660 lb)
and extraction
Figure 2b. A bioprocessing system that focuses on production of a range of products with minimal waste.
390 K.A. Dawson
substrates. In some cases, this may be at the years, there has been considerable interest in the
expense of ethanol production. Ideally, the development of specific enzyme systems for
economic value for the biorefinery would come producing fermentable substrates from fibrous
from its versatility. Biorefineries that can use a plant materials and waste products. Special
wide variety of feedstocks and produce a variety projects have focused on the development of
of co-products will have the ability to adapt their inexpensive cellulase and xylanase enzyme
processes to address external economic systems with improved catalytic activities. Major
pressures. development efforts using both fiber-digesting
In addition, there are a number of other basic enzymes and new thermochemical processes
pressures that make the biorefinery concept have allowed for the saccharification of non-
attractive for modern ethanol production traditional lignocellulosic substrates (Kaylen et
systems. In particular, environmental issues and al., 2000). However, there is also concern about
the need for controlling waste streams from the use of the pentose sugar derived from the
chemical production systems are becoming the hemicellulose fraction of fibrous plant material
driving force for changes in ethanol production. because these sugars are not used by
In this respect, it is becoming possible to use Saccharomyces cerevisiae in traditional ethanol
biotechnology to design biorefineries in a way production systems. Feedstocks may also
that will allow them to be emission-free, efficient include waste products from the sugar and starch
in their use of energy and water, and yet flexible refining industry, milk sugars in whey, corn fiber,
enough to profitably provide useful products forest products and even municipal and food
from inexpensive substrates. industry wastes. Innovative applications of
biotechnology and engineering principles are
needed to obtain maximum benefit from this
Feedstocks for the biorefinery array of substrates.
high fiber and high protein content that remains into the animal feed market. Probably the best
after the preparation of the readily fermentable opportunity for developing the biorefinery
sugars and starch for fermentation. Distillers concept is related to the innovative use of the
grains account for the largest single co-product distillers grains. Since distillers grains are most
from the ethanol industry in the United States. often valued as a protein source for food
Close to 10 million tons of these DDG is animals, strategies that increase the protein
produced in the United States each year. Because content, alter the availability the protein, and
of the high fiber content, many of these co- improve amino acid composition may have
products are typically used as feeds for considerable value. Such strategies may also
ruminants (beef and dairy cattle). However, improve the profile of the key nutrients for
many of them also contain useful substrates that animal production and allow for a more
could provide a potential starting point for economical application of these spent grains in
further biotransformation into high value monogastric animals (pigs) and poultry diets. A
products. Such transformations provide higher number of approaches are currently available for
value feeds for other livestock production modifying distillers grains and improving their
systems. overall value as co-products. These can be
simple supplementation strategies, but many of
them are true biotransformations that could be
Changing the value of traditional co- carried out in a biorefinery setting.
products in a biorefinery One approach for improving the value of distillers
grains would be to develop supplementation
Distillers grains are considered to be the strategies that improve nutrient profiles. This can
standard co-products of the alcohol fermentation be accomplished by adding deficient or unique
process in dry mill, ethanol production systems, amino acids, vitamins, and minerals in a way that
but can be quite variable in quality. While there complements the basic nutritional value of spent
has been some attempt to differentiate distillers grains and other dietary components. Since the
grains obtained from different sources, most of results of these strategies are often reflected in
these materials are currently sold as commodities improved animal performance, it is often difficult
392 K.A. Dawson
to evaluate the economic value of such Probably the most effective way to enhance
approaches based on simple nutrient analysis. the value of distillers grains would be to expose
Instead, it must be based on the improved them to a more complex biotransformation using
performance of the livestock (Table 2). Early a secondary fermentation system. This process
studies with one such supplementation strategy could use selected strains of fungi and bacteria
using proprietary vitamin and mineral mixes in combination with a short fermentation period
have indicated that it is possible to get a three- to further alter the grain composition. This
to-one return on processing costs when such process could increase the protein content,
strategies are used to improve milk production increase the lysine content of the protein, and
in dairy cattle. improve the overall digestibility of the protein.
Another approach for improving the value of This fermentation process may also decrease the
the distillers grain could focus on the relatively fiber content of the spent grains and increase
poor digestibility of the protein and fiber fractions their overall energy availability. Such
in the spent grains. Current estimates of the manipulation could provide a much more
availability of the most limiting amino acids in balanced protein to the animal in the form of a
distillers grains is estimated to be about 65% in highly digestible feed ingredient. Simple
swine diets (Dale and Batal, 2002). This suggests fermentations that increase protein content from
that a large portion of the protein ingested by around 30% to up to 45%, increase amino acid
the pig is not subject to digestion and is availability from 65 to 90%, and improve lysine
unavailable as a nutrient. A biotransformation content (from 1.35% to 2.7%) can improve the
with an enzymatic treatment that increases the value of the spent grains as a feed ingredient.
amino acid availability from 65 to 90% could This can result in up to a 60% increase in the
significantly alter the value of spent grains. basic economic value of the distillers grains as
Currently, biotechnology may provide specific a feed supplement (Table 2). These examples
enzyme complexes that can bring about the clearly show the value of strategic applications
required improvements in amino acid of microbial induced biotransformations in a
availability. If such modified grains are biorefinery setting.
formulated into a diet for a pig in place of
soybean meal, savings in the cost of a ton of
feed can be demonstrated (Table 2). In this case, Products from the biotransformation
biorefinery steps that allow for the pre-treatment processes
of distillers grains with an enzyme may increase
the basic economic value of the distillers grain Commercially, a number of products are
as a feed ingredient in pig diets by as much as
20 to 30%.
Table 2. Potential value of developing fermentation co-products from spent distillers grains for swine.
Strategy Goal of strategy Decrease in feed cost Added value per ton of distillers
relative to soybean meal grainsa
Strategic supplementation Improved or more balanced Depends on cost of Value only obtained through
with nutrients or dietary nutrient profile supplement enhanced performance of pigs
adjuncts (minerals, 1:3 return on investment has
vitamins, amino acids, been reported
or enzymes)
produced using the basic metabolic activities of ability to contribute to bioconversion of basic
living organisms (Table 3). Many of these feedstocks and metabolic intermediates within
products are produced in well-defined biorefineries.
fermentation systems that convert low quality
substrates into high value, vendable products or
chemical precursors for other manufacturing YEAST
processes. Alternatively, some of these
biotransformations can be achieved by applying While the metabolic activities of many different
specific enzyme systems. By applying new types of microorganisms have been examined
fermentation processes, biotechnology can for functional roles in biorefinery systems,
provide tools that can be used for the production currently the vast majority of commercial
of new inexpensive enzyme systems for more fermentations depend on active yeast cells. These
cost-effective biotransformations. Innovative organisms are the basic biocatalysts that are
applications of some of the fermentation central to the modern fermentation industry. The
processes or enzyme treatments into integrated most important yeasts in this respect are from
production systems will be important to the the species Saccharomyces cerevisiae. There are
development of flexible biorefineries (Figure several thousand strains of S. cerevisiae
2b). available to the fermentation industry. They can
be differentiated biochemically and by their
Table 3. Some industrial products currently produced basic fermentation activities. However, a number
using fermentation technologies that might be produced of non-conventional yeast have been tested and
in a biorefinery using microbial biocatalysts. used as biocatalysts for the production of
specialized fermentation products (Table 4).
Ethanol Representative strains from these different yeast
Methanol species carry out a variety of biochemical
Citric acid processes that make them attractive for driving
Glycerol
Xylitol and other sweeteners
fermentation and biotransformation processes in
Astaxanthin an integrated biorefinery setting.
Lactic acid Metabolic engineering is a concept that uses
Polylactide polymers powerful genetic engineering approaches and
Emulsifiers and surfactants microbial physiology in strain development
Vitamins (riboflavin) strategies for improving the biocatalytic
Fatty acids and oils capabilities of microorganisms (Olsson and
Enzymes Nielsen, 2000). Metabolic engineering of yeast
Specific pharmaceuticals and specific heterologous proteins now allows for the development of fermentation
Single cell protein
processes that integrate much of our knowledge
about physiological, metabolic, and biochemical
properties of this group of organisms using
specific genetic engineering technology to
Microorganism use in biotransformations improve specific bioprocesses. This is becoming
The processing power in modern biorefineries a very successful way of changing our traditional
is often limited by the metabolic power of fermentation technologies. For example,
microorganisms or biocatalysts available. recently, genetic modification techniques have
Modern biotechnology is now capable of made a variety of yeast strains available that have
engineering new metabolic systems that increase very specific metabolic activities (Gong et al.,
the flexibility of biorefineries. These are based 1999, Mouiruzzaman et al., 1997). These
on both traditional or naturally occurring molecular tools are extremely powerful and
microorganisms and genetically modified make it possible to initiate metabolic engineering
organisms. Microorganisms used for bio- programs that allow for new processing
transformation fall in to three major groups: strategies. Genetic modification has been
classical yeast, fungi, and bacteria. Many types especially important in developing strains of
of organisms have been examined for their yeast that can use the xylose derived from
394 K.A. Dawson
Table 4. Some non-conventional yeast uses available for use in modern biorefineries.
Yeast Applications
Candida boidinii Vendable chemicals and enzymes production from fatty acids, methanol and ethanol
Candida butyrii Ethanol production from xylose
Candida guilliermondii Xylitol and citric acid production from pentoses and cellulose
Candida shehatae Ethanol production from xylose
Candida tenuis Ethanol production from xylose
Candida tropicalis Xylitol and other vendable chemicals produced from pentoses and petroleum base substrates
Candida utilis Wide variety of organic chemicals from fibrous substrates
Cryptococcus laurentii Enzyme and organic acid production from waste oils
Debaryomyces hansenii Xylitol and citric acid production from xylose and hydrocarbons
Geotrichum spp. Enzyme production
Hansenula polymorpha Protein production from methanol
Kluyveromyces fragilis Thermotolerant ethanol production from fiber hydrolysates
Kluyveromyces lactis Ethanol, enzyme and heterologous protein production from whey
Kluyveromyces marxianus Ethanol production from whey, xylose, and pectin
Pachysolen tannophilus Ethanol and xylitol production from xylose from pentoses
Phaffia rhodozyma Carotenoid pigment production from corn by-products and wood hydrolysates
Pichia pastoris Heterologous protein and enzyme production from methanol and oxygenated hydrocarbons
Pichia stipitis Ethanol production from xylose and xylans (hemicellulose hydrolysates)
Schizosaccharomyces pombe Ethanol production from simple sugars
Schwanniomyces occidentalis Halophilic biotranformations of starch to organic chemicals
Yarrowia lipolytica Citric acid production from plant oils, hydrocarbons, and petroleum products
lignocellulose digestion (Ho et al., 1999, on a variety of substrates and are readily
Krishnan et al., 1999) and the development of manipulated to produce a variety of useful end
S. cerevisiae strains that contain ß-glucosidase products. Fungi are often used to economically
and directly produce ethanol from cellulose (Cho produce industrial enzymes. This enzyme
and Yoo, 1999). When combined with our production often takes advantage of the ability
understanding of molecular genetics, our of these organisms to grow on solid substrates,
sophisticated understanding of yeasts and their such as wheat bran, with relatively low moisture
physiological activities make them attractive as contents (around 50%). This characteristic makes
biocatalysts in biorefineries. it possible to grow these organisms in surface
culture systems (solid state fermentation) with
exceptionally high yields of enzyme. This has
FUNGI been the basis for developing large scale ‘Koji’
or solid state fermentation systems (SSF).
The metabolic flexibility and unique substrate Enzyme complexes produced in a SSF system
range of fungi make them potential contributors have a number of unique characteristics,
to the modern biorefinery. However, relative to including a broad range of side or auxiliary
the yeast, very few species have been evaluated activities, that make them more powerful than
for their contributions to commercial production purified enzymes obtained from traditional liquid
systems (Table 5). These organisms can grow fermentation systems. These types of enzyme
systems could be attractive co-products in a Such organisms can provide a wide variety of
biorefinery system where partially dried distillers biochemical activities that may be integrated into
grains are available as a substrate. For example, unique fermentations systems in a biorefinery.
it is possible that DDGS could be used as a Molecular manipulation and genetic
substrate for producing an enzyme such as engineering have also allowed for the
Rhizozyme™, which would make it unique not development of many new strains of bacteria
only in ability to totally replace other with unique fermentation capability. For
saccharifying enzymes (saving $0.04-$0.05 example, strains of bacteria that contain active
gallon), but bring about an increase in yield since lignocellulose-degrading enzymes have been
hemicellulose/cellulose would be degraded. developed and can be used to provide innovative
It is also possible to think about metabolic microbial activities in biorefineries. Genetic
engineering of fungal cultures to make them manipulation of Escherichia coli provides one
more useful as tools in a modern biorefinery. of the most powerful tools for modern
While mutant strains of fungi have been used biotechnology and provides the best studied
for many years to produce commercial enzymes, system for manipulating microbial processes.
systems for routinely manipulating their genetic Manipulated strains of this organism are
makeup are still lacking. New frontiers will open commonly used to produce many types of
up when reliable technologies for metabolic pharmaceutical products, enzymes, and
engineering of fungi become readily available. diagnostic reagents. Specific strains of this E.
coli that produce ethanol have also been
developed (Ingram et al., 1999). Some of these
BACTERIA strains can use a variety of substrates, including
some of the pentose sugar derived from the
Bacteria have a long history of contributing to degradation of the hemicellulose fraction of plant
commercial fermentation processes and are often fiber (Dien et al., 1999).
considered in the development of unique In recent years, there has been considerable
processes in biorefinery systems. The ability of interest in using heat resistant thermophilic
bacteria to use multiple substrates and their bacteria in industrial fermentation processes
susceptibility to genetic manipulation make them (Scopes, 1997). Many of these bacteria naturally
attractive candidates for biocatalytic roles. produce mixtures of ethanol and organic acids,
However, bacterial fermentations in commercial but tend to be less ethanol tolerant than yeast.
settings generally tend to be less robust than those These organisms are uniquely adapted to carry
associated with yeast and fungi. As a result, care out fermentation at temperatures higher than
must be taken in preparing substrates and 50oC. Application of these types of organisms
facilities that will support bacterial fermentations. may allow for an increase in the temperature
Many types of naturally-occurring bacteria have range for biocatalytic activities and allow for
been examined for their potential role in high temperature strategies in biorefineries.
commercial fermentation systems (Table 6).
Other tools from biotechnology the basic tools needed to develop these integrated
biological based processing systems for use in
In addition to allowing for the development of biorefineries.
new strains of microorganisms, molecular
biology and metabolic engineering techniques References
are also beginning to provide other new systems
and processes that can be used for controlling Abbas, C.A. and M. Cheryan. 2002. Emerging
chemical processes in biorefineries. Specifically, biorefinery opportunities. Appl. Biochem.
new enzymes and enzyme expression systems Biotech 98-100:1147.
have been strategically designed for specific Bon, E.P.S. and S. Picataggio. 2002. Enzyme and
metabolic transformations and are currently microbial biocatalysts. Appl. Biochem.
being produced and evaluated in a number of Biotech. 98-100:163.
applications. Much of this work is focusing on Cho, K.M. and Y.J. Yoo. 1999. Novel SSF process
the development of cellulases and xylanases that for ethanol production from microcrystalline
will allow for more efficient conversion of cellulose using the d–integrated recombinant
fibrous biomass into fermentable sugars. Current yeast, Saccharomyces cerevisiae L2612dGC.
research programs in these areas not only look J. Microb. Biotech. 9(3):340-345.
at increasing the yields of enzymes, but also at
Dale, N. and A. Batal. 2002. Ingredient analysis
the production of more efficient enzyme
table: 2002 Edition. Feedstuff 74:16-22.
systems. Tools such as proteomic analysis,
molecular flux analysis, genomic analysis and Dien, B.S., L.B. Iten and R.J. Bothast. 1999.
directed molecular evolution will be important Conversion of corn fiber to ethanol by
in developing the next generation of enzymes recombinant E. coli strain FBR3. J. Indust.
for use in biorefineries (Bon and Picataggio, Microb. Biotech. 22:272-581.
2002). The development of such enzyme-based Gong, C.S., N.J. Cao, J. Du and G.T. Tsao. 1999.
sugar production platforms has been a key part Ethanol production from renewable resources.
of the strategies for economically converting Adv. Biochem. Engin./Biotech. 65:207-241.
renewable fibrous biomass to ethanol. It can be Ho, N.W.Y., Z. Chen, A.P. Brinard and M. Sedlak.
expected that similar techniques can be used to 1999. Successful design and development of
develop other enzyme systems to carry out key genetically engineered Saccharomyces yeasts
transformation in biorefineries. The alcohol for effective cofermentation of glucose and
industry can look forward to the application of xylose from cellulosic biomass to ethanol.
many new enzymes and enzyme complexes that Adv. Biochem. Engin./Bioctech. 65:164-191.
will increase the substrate ranges and resistance Ingram, L.O., H.C. Aldrich, A.C.C. Borges, T.B.
to some of the environmental extremes Causey, A. Martinez, F. Morales, A. Saleh, S.A.
associated with industrial fermentations. These Underwood, L.P. Yomano, S.W. York, J.
will improve the efficiency of ethanol production Zaldivar and S. Zhou. 1999. Enteric bacterial
and improve the value of co-products. catalysts for fuel ethanol production. Biotech.
Progress 15:855-866.
Kaylen, M., D.L. Van Dyne, Y.S. Choi and M.
Conclusions Blasé. 2000. Economic feasibility of producing
ethanol from lignocellulosic feedstocks.
The future of many fermentation plants will
depend on their ability to rapidly adapt to Bioresource Tech. 72:19-32.
changes in the cost of substrates and to obtain Krishnan, M.S., N.W.Y. Ho and G.T. Tsao. 1999.
the maximum value from the products they Fermentation kinetics of ethanol production
produce. The concept of developing flexible from glucose and xylose by recombinant
biorefinery systems that allow for the use of a Saccharomyces 1400(pLNH33). Appl.
variety of feedstocks and that have the ability to Biochem. Biotech. 77-79:373-388.
produce several high value co-products will be Lynd, L.R., J.H. Cushman, R.J. Nichols and C.E.
important when determining the value of alcohol Wyman. 1991. Fuel ethanol from cellulosic
production systems. Biotechnology will provide biomass. Sci. 251:1318-1323.
Biorefineries: the versatile fermentation plants of the future 397
Mouiruzzaman, M., B.S. Dien, C.D. Skory, Z.D. Scopes, R.K. 1997. Ethanol from biomass: The
Chen, R.B. Hespell, N.W.Y. Ho, B.E. Dale and potential use of thermophilic organisms in
R.J. Bothast. 1997. Fermentation of corn fibre fermentation. Australasian Biotech. 7:296-299.
sugars by an engineered xylose utilizing Torre, P. 1999. Co-products from ethanol
Saccharomyces yeast strain. World J. Microb. fermentation: Alternatives for the future. In:
Biotech. 13:341-346. The Alcohol Textbook, 3 rd Edition (K.A.
Olsson, L. and J. Nielsen. 2000. The role of Jacques, T.P. Lyons and D.R. Kelsall, eds).
metabolic engineering in the improvement of Nottingham University Press, Nottingham, UK,
Saccharomyces cerevisiae: utilization of pp. 335-346.
industrial media. Enzyme Microb. Tech.
26:785-792.
The Alcohol Alphabet 399
Italic type denotes words (or minor variations of words) defined under separate entries.
John Murtaugh
Sept 12, 1936 - Jan 25, 2003
Beginning with his dissertation work on the cooking process through a lifetime of consulting, educating and
inspiring others, Dr. John Murtaugh moved the ethanol industry forward. He began this glossary with the
2nd edition of the Alcohol Textbook. Like the rest of John’s work, it has an important and very practical
purpose. As such, it will be updated and continued in this and successive editions of the book.
Acidity A quantitative measure of the amount The principal alcohol in fuel and beverage
of acid present. use is ethanol (otherwise known as ethyl
alcohol). The lower molecular weight alcohol
Acid-acid process Term used in starch (methanol) and higher alcohols are more
processing when acid hydrolysis is used to toxic.
accomplish both the initial liquefaction and
the final saccharification to simple sugars. Alcohol Fuel Permit (AFP) A permit issued by
the Bureau of Alcohol, Tobacco and Firearms
Acid-enzyme process Term used in starch allowing the holder to engage in the
processing when acid hydrolysis is used to production of ethanol solely for fuel use.
accomplish the initial liquefaction, and an
enzyme such as amyloglucosidase is used for Aldehyde A member of a class of organic
the saccharification to simple sugars. compounds considered to be derived by the
removal of hydrogen atoms from an alcohol.
Acid hydrolysis The hydrolysis of a polymer Aldehydes tend to be produced as congeners,
by the use of acid. In the case of starch or by-products of fermentation; and having a
hydrolysis, acids may be used as an lower boiling point than ethanol, they tend to
alternative to enzymes in either (or both) the vaporize more readily and may accumulate
liquefaction or saccharification processes. as a ‘heads’ fraction in the distillation process.
Acid washing A process in which yeast Alkali A compound capable of neutralizing
recovered from a finished fermentation is hydrogen ions, usually by generating
acidified to pH 2.2 for ~20 minutes using an hydroxyl ions which combine with hydrogen
acid like phosphoric acid, to reduce the level ions to form water.
of bacterial contamination prior to recycling
into a new fermentation. Alkalinity Specifically in water, a measure of
the total bicarbonate and carbonate content,
Active dry yeast A yeast preparation made from determined by titration to an endpoint using
compressed yeast by careful and controlled various indicators such as phenylphthalein,
removal of moisture to ~5% w/v, such that methyl orange, etc.
10-50 billion yeast/gram remain viable.
Storage for a number of months allows plants Alpha-amylase (α α -amylase) An enzyme used
to eliminate risky propagation ‘in-house’. in the liquefaction of starch in the grain
mashing process prior to saccharification and
Aguardiente An unaged alcoholic beverage fermentation. α-amylase hydrolyzes the long-
produced in Central and South America by chain starch molecules into short-chain
the distillation of beer derived from the dextrins. These are more suitable for
fermentation of sugarcane juice or molasses. subsequent saccharification by other enzymes
It is similar to crude rum. to fermentable glucose. (α-amylase is an
Alcohol A member of a class of organic endoenzyme in that it works from the inside
compounds containing carbon, hydrogen and of the amylose molecule). In beverage alcohol
oxygen. Considered to be hydroxyl derivatives production α-amylase may be derived from
of hydrocarbons produced by the malt (sprouted barley), but in fuel ethanol
replacement of one or more hydrogen atoms production the enzyme is obtained solely as
by one or more hydroxyl (OH) groups. Under a bacterial product. The enzyme molecule
the International Union of Pure and Applied contains a calcium atom, which is essential
Chemistry (IUPAC) naming system, the name for its activity.
given to an alcohol is derived from the parent American Society for Testing and Materials
hydrocarbon with the final ‘e’ changed to ‘ol’. (ASTM) A scientific and technical
Thus methane-methanol, ethane-ethanol, etc.
The Alcohol Alphabet 401
Arabinose A pentose sugar comprising a major the volume of liquid residue to be evaporated.
constituent of hemicellulose. Chemical Improperly handled, it may be a major source
formula C 5 H 10 O 5 . It is not fermented by of bacterial contamination, and may contain
normal strains of distillers yeasts. Also known high levels of chemicals that cause stuck/
as pectin sugar or gum sugar. sluggish fermentations.
Atomic weight The relative mass of an atom Bacteria Microscopic organisms of the kingdom
based on a scale in which a specific carbon Monera usually characterized by small size,
atom (carbon 12) is assigned a mass value of vigorous biochemical activity and lack of a
12. Also known as relative atomic mass. true nucleus.
Azeotrope The term used to describe a constant Bacterial contamination The condition
boiling mixture. It is a mixture of two (or occurring when undesirable bacteria become
more) components with a lower boiling point established in a fermenting mash and reduce
than either component alone. For example the ethanol yield. The bacteria use available
water, which boils at 100 oC, and anhydrous sugars to produce various compounds
ethanol, which boils at 78.5 o C, form a (congeners), particularly acids, which may
constant boiling mixture (azeotrope) at inhibit yeast activity. In severe situations,
78.15 oC. The vapor of the mixture has the bacterial contamination may cause serious
same composition as the liquid and therefore economic losses.
no further concentration can be achieved by
normal distillation. Under normal pressures Balling (or Brix or Plato) A scale used to
it contains approximately 97% by volume measure the specific gravity of a liquid in
ethanol (1940 proof). It is very expensive in relation to that of a solution of sugar in water.
terms of energy to attempt to reach 1940 proof, Each unit on the scale is equivalent to 1% by
so 190 0 proof is generally considered to be weight of sugar. Thus a mash of 20o Balling
the practical, economic azeotrope limit for fuel has the same specific gravity as a 20% w/w
ethanol distillation. sugar solution. The scale is frequently
considered to indicate % dissolved solids in a
Azeotropic distillation A distillation process liquid, although this is only true of solutions
in which a liquid compound (entrainer) is of pure sucrose. Traditionally, the term
added to the mixture to be separated to form ‘Balling’ has been used in grain distilleries
an azeotrope with one or more of the and breweries, while ‘Brix’ has been used in
components. Normally, the entrainer selected sugar mills and rum or molasses alcohol
is easily separated from the component to be distilleries. The measurement is accomplished
removed. For example, when benzene is used by use of a Balling (or Brix/Plato) hydrometer,
in azeotropic distillation to dehydrate ethanol, refractometer or more recently density meter.
the overhead condensate phase-separates to
yield a water-rich layer that can be withdrawn Barbet time See Permanganate time.
and a benzene-ethanol layer, which is Bargeload Generally refers to a river barge with
refluxed. a capacity of 10,000 barrels or 420,000
gallons.
Barrel A liquid measure equal to 42 US gallons
B or 5.6 cubic feet. Or, a wooden container used
Backset Recycled thin stillage. It may be added for the aging and maturation of alcoholic
to the cooker or to the fermentor and may beverages. Barrels used for whisky maturation
serve as a partial source of nutrients. It reduces are made of oak wood and have a capacity of
the water required for mashing and reduces
The Alcohol Alphabet 403
about 52 US gallons. Barrels may only be extracts the ethanol from the beer and a
used once for aging bourbon whisky, so there concentrating or rectifying section, which
is a worldwide trade in used bourbon barrels normally takes the ethanol up to 1900 proof
for aging other alcoholic products such as (95o GL). Beer stills may consist of a single
Scotch whisky and rum. A US beer barrel tall column, or two or more columns standing
contains 31 US gallons. side by side, linked by vapor pipes.
Base losses The percentage of ethanol lost in Beer stripping column See Beer still.
the stillage at the base of a beer stripping
column or a rectifying column. It is virtually Beer well The holding vessel into which finished
impossible to achieve zero base losses, and it beer is transferred prior to distillation.
would be wasteful of steam. The base losses Benzene Colorless, flammable, aromatic
are generally monitored and controlled in an hydrocarbon liquid. Chemical formula C6H6.
optimal range determined by steam costs, etc. Boils at 80.1OC and freezes at 5.4OC. Used as
BATF Abbreviation for US Bureau of Alcohol, an entrainer for the dehydration of ethanol
Tobacco and Firearms. by azeotropic distillation. Known to be
carcinogenic.
Batch cooking Cooking a set amount of grain
meal, water and backset (if any) in a single Beta-amylase (ß-amylase) An enzyme that
vessel as a discontinuous operation. One or hydrolyses the long-chain amylose molecules
more batch cooks may be used to fill a single in starch into fermentable maltose, the dimer
batch fermentor. Batch cooking is mainly (or double-molecule) of glucose. It is an
confined to the beverage alcohol industry and exoenzyme in that it works from an outer end
smaller plants in the fuel ethanol industry. of the molecular chain. It is found in malt
(sprouted barley) in association with α-
Batch distillation Distilling batches of beer in amylase. With the advent of microbial
a discontinuous operation. It is not commonly amyloglucosidase enzymes, malt amylases
used in fuel ethanol production, but is used are generally only used in the production of
in the beverage ethanol industry particularly heavily-flavored beverage alcohol.
for the production of special types of heavily-
flavored distillates. Betaglucan (ß-glucan) Gum-like polymers of
ß-linked glucose as in cellulose instead of the
Batch fermentation The fermentation of a set α-linked glucose units as in starch (amylose).
amount of mash in a single vessel in a Betaglucan is commonly present in barley
discontinuous operation. mashes. It is not broken down by α-amylase
and causes foaming problems due to its
Beer The name given to the product of alcohol viscous elastic nature.
fermentation. In grain alcohol production,
beer may contain from 9->16% ethanol (v/v). Betaglucanase (ß-glucanase) An enzyme which
hydrolyses betaglucan. It is frequently used
Beer preheater A heat-exchanging device used in barley mashing to reduce foaming and
for heating beer before it enters a beer still. viscosity problems.
Usually, it also serves as the first condenser
for the beer still, so that the overhead vapors Beverage alcohol Any form of distilled ethanol
heat the beer. with or without congeners considered by law
to be fit for human consumption. The laws in
Beer still The distillation unit used for the initial many countries confine beverage alcohol to
removal of ethanol from finished beer. It that produced by fermentation as opposed to
generally consists of a stripping section that alcohol of synthetic origin.
404 The Alcohol Alphabet
BTU Abbreviation for British Thermal Unit. approximately 75% of the dry weight of all
vegetation.
BTX Abbreviation for the three related octane
enhancers, benzene, toluene and xylene. Carbon dioxide A colorless non-flammable
gas. Composition CO2. It does not support
Bubble cap A contacting device used on some human respiration; and in high concentrations
distillation plates. It consists of a cylindrical it causes asphyxiation. It is approximately 1.5
chimney set in a hole in the plate and covered times the weight of air, and tends to
by a dome-shaped cap, which deflects the accumulate in floor drains, pits and in the
vapors rising up the chimney to cause them bottoms of unventilated tanks. It is produced
to pass through the liquid layer on the top by various means, notably the combustion of
side of the plate. fuels in an excess of air and is a by-product
Bushel A unit of dry volume equal to 2150.42 of yeast fermentation. It may be recovered
cubic inches or 1.244 cubic feet. When used from fermentations and compressed to a liquid
to measure grain, bushel weight depends on or solid (dry ice), or used to carbonate
the type and condition of the grain. In the case beverages.
of corn, bushel volume generally averages Carbon monoxide A colorless, odorless,
about 56 lbs by weight. This has led to the flammable gas (‘coal gas’). Composition CO.
use of a distillers bushel, which represents 56 It is poisonous if inhaled, as it combines with
lbs of grain, regardless of type or volume. blood hemoglobin to prevent oxygen transfer.
Butanol (butyl alcohol) A minor constituent of It is very slightly lighter than air. It is
fusel oil. Chemical formula C 4H 9OH. Four produced by the incomplete combustion of
isomers exist. They are all colorless, toxic fuels with a limited oxygen supply, as in auto
flammable liquids. N-Butanol may be engines. It is a major component of urban air
produced as a co-product with acetone and pollution, which can be reduced by blending
ethanol by the fermentation of selected an oxygen-bearing compound (or oxygenate)
carbohydrates with the anaerobic bacterium such as ethanol into hydrocarbon fuels.
Clostridium acetobutylicum. Butanols are Carbon steel A steel deriving its particular
used as solvents and chemical intermediates. properties from its content of carbon. These
By-products Products that are secondary to the steels may range from ‘low-carbon’, (having
principal product of a process. In ethanol less than 0.25% carbon), to ‘high-carbon’.
production, carbon dioxide and distillers Low carbon or ‘mild’ steel is easily hot-
dried grains are normally considered by- worked and rolled, and is used in steel plates
products; but in certain circumstances they and beams. It is easily welded, but is readily
may be viewed as co-products, in that they subject to rusting and other forms of
may contribute significantly to the overall corrosion.
process economics. Caribbean Basin Initiative (CBI) The program
arising from the Caribbean Basin Economic
Recovery Act passed by the US Congress in
C 1983 to encourage industrial development in
the Caribbean islands and Central America.
Carbohydrate Any of a group of compounds Under this program, fuel ethanol and other
composed of carbon, hydrogen and oxygen products from the region may enter the US
including the sugars, starches, dextrans and without being subject to customs duties.
celluloses. They are the most abundant class
of organic compounds in nature, constituting Cassava A root crop with a high starch content
406 The Alcohol Alphabet
grown in the tropics and subtropical regions. centrifugal pumps unsuitable for moving
Known in Brazil as manioc, it is used as an heavy viscous liquids such as molasses and
alternative to sugarcane as a feedstock for syrups.
ethanol production. It is also processed for
food as ‘tapioca’. Centrifuge A machine for separating insoluble
liquids or solids from liquids by the
CBI Abbreviation for Caribbean Basin application of centrifugal force. Two main
Initiative. types of centrifuges are basket and decanter.
In basket centrifuges, solids are retained in a
CCC Abbreviation for Commodity Credit rotating perforated basket while liquid passes
Corporation. through the perforations. In decanter
CDA Abbreviation for completely denatured centrifuges, the mixture is thrown against a
alcohol. solid-walled cylinder and the heavier solid
particles collect against the wall for removal
Cell recycle The process of recovering yeast while the lighter liquid collects in the central
from fermented beer to return it to the starting part. Centrifuges are commonly used in
vessel of a continuous fermentation or to a ethanol plants for yeast recovery and in
new vessel in a batch fermentation system. It stillage dewatering.
may include an acid washing step to reduce
bacterial contamination. Chelating agent A chemical that combines with
and solubilizes water hardness ions to prevent
Cellulase An enzyme capable of hydrolyzing precipitation and scale formation. Chelating
long-chain cellulose molecules into simple agents are also used in detergent formulations
sugars or short-chain polymers. to remove scale deposits. EDTA (ethylene
diamine tetraacetic acid) is a chelating agent.
Cellulose The principal polysaccharide in living
plants. It forms the skeletal structure of the Chemical oxygen demand (COD) A laboratory
cell wall, hence the name. It is a polymer of test to determine the oxygen requirements for
glucose units coupled by ß-type linkages into the chemical digestion of effluents. It should
chains of 2,000-4,000 units. Cellulose be distinguished from the BOD test, which
normally occurs with other polysaccharides determines the total of both chemical and
and hemicelluloses derived from sugars such biological oxygen requirements. COD is
as xylose, arabinose and mannose. measured in a rapid test that involves heating
the effluent in the presence of an oxidizing
Celsius (or centigrade) A temperature scale in
agent such as potassium dichromate and then
which (at normal atmospheric pressure) water
determining the amount of oxygen absorbed
freezes at zero degrees and boils at 100
by the effluent.
degrees.
Chlorine dioxide Chemical formula ClO2. It is
Centrifugal pump A machine for moving
a strongly-oxidizing, yellow-to-reddish-
liquids by accelerating them radially outwards
yellow gas at room temperature. It has an
by means of a rotating impeller contained
unpleasant odor similar to that of chlorine and
within a casing. It is the most common and
reminiscent of nitric acid. It is unstable in
most versatile type of pump in normal
light. It reacts violently with organic materials
industrial use. It has an advantage in that it
and is easily detonated by sunlight or heat in
does not force a positive displacement of
concentrations greater than 10% at atmos-
liquids, so it may be used in systems where
pheric pressure. Boils at 11oC and freezes at
output must be throttled or completely closed
-59 oC. Chlorine dioxide may be used as a
by control valves. This feature makes
sterilant, and may be produced in situ for
The Alcohol Alphabet 407
sterilizing yeast mashes by addition of sodium chemical sterilizing of equipment at the press
chlorite solution in the presence of acids or of a button.
chlorine (or hypochlorite solution). It is
considerably more effective as a sterilant than Cleaning-out-of-place (COP) The manual
straight chlorine. disassembly, inspection and cleaning of
processing equipment and systems. Some
Chromatography A method for separating a degree of COP is usually required with CIP
mixture of chemical compounds into (cleaning-in-place).
individual components by selective
distribution between two immiscible materials Closed receiver See Receiver.
(or phases), one stationary and the other CMS Abbreviation for condensed molasses
mobile. The phases are selected so that the solubles.
mobile phase will carry the various
components through the stationary (or solid) Coccus (plural: cocci) A type of bacteria with
phase at differing rates to give separation. Gas spherically-shaped cells. They may occur as
chromatography may be used to separate single cells, clusters or long chains.
ethanol from the other congeners produced
COD Abbreviation for chemical oxygen
in fermentation and to measure them
demand.
quantitatively. High performance liquid
chromatography may be used to follow starch Column A vertical cylindrical vessel containing
hydrolysis in grain alcohol production. a series of perforated plates or other contact
devices through which vapors may pass to
CIP Abbreviation for cleaning-in-place system.
effect a separation of liquid mixtures by
Citrus molasses A by-product of the citrus juice distillation.
industry. Citrus residue, mainly peel, is treated
Co-mingled tank The term used for fuel ethanol
with lime and then passed through a press.
tanks at refineries or pipeline terminals where
The press liquor is then evaporated to a
two or more suppliers may share the same
viscous, dark brown molasses of about 72 o
tank for storing ethanol.
Brix. Citrus molasses is similar to cane
blackstrap molasses, having about 45% total Commodity Credit Corporation (CCC) An
sugars. However, it has more protein and a agency of the USDA established to stabilize
much lower ash content. Citrus molasses may and protect farm incomes and prices by
be diluted and fermented for ethanol maintaining balanced and adequate supplies
production, but it may need pretreatment to of agricultural commodities.
reduce the content of d-limonene (commonly
referred to as ‘citrus-stripper oil’), which tends Completely denatured alcohol (CDA) A term
to inhibit yeast growth. used by the BATF to describe ethanol made
unfit for human consumption by addition of
Cleaning In processing systems, cleaning is specified denaturants such as methyl isobutyl
removing soils in the equipment to a degree ketone, kerosene or gasoline.
that is acceptable for the process.
Compressed yeasts A paste of viable yeasts of
Cleaning-in-place system (CIP) A system approximately 30% total solids prepared by
designed to permit process equipment to be filtration of yeast slurry. Refridgeration is
cleaned without disconnecting or required at 4°C, and such yeasts loose viability
dismantling. A sophisticated automatic CIP constantly with a shelf life of less than 3
system may provide a sequence of water weeks.
flushing cycles, detergent washing and
408 The Alcohol Alphabet
Concentrating column A column where by a pump to a steam jet heater (jet cooker), a
hydrous ethanol is concentrated to decrease holding vessel or lengths of piping (to provide
water content. A beer still may consist of a some residence time at the cooking
stripping column in which dilute ethanol is temperature), one or more flash vessels (to
removed from beer and a concentrating cool the cooked mash), a holding vessel for
column, which receives ethanol vapor from enzymatic liquefaction and a heat exchanger
the stripping column and concentrates it up for final mash cooling. Continuous cookers
to about 1900 proof (95oGL). If provision is are more common in fuel ethanol plants than
made in a concentrating column to remove in beverage alcohol plants.
impurities such as fusel oil, then it is more
correctly a rectifying column. Continuous fermentation A system into which
cooked mash may be fed continuously to be
Condensate Liquid condensed from vapor in a fermented and then discharged to the beer well
condenser. and distillation system. Continuous
fermentation systems generally consist of a
Condensed molasses solubles (CMS) The term series of interconnected tanks sized to provide
used to describe molasses stillage sufficient residence time for the fermentation
concentrated by evaporation. The molasses to proceed to completion. The ethanol
residue (after fermentation and distillation) industry is divided on fermentation systems,
may be concentrated to about 60° Brix (or with more gallonage produced by continuous
approximately 60% solids) to be sold as a fermentation than by batch fermentation.
substitute for molasses in animal feeds as a
caking agent and dust suppressant. It contains Continuous distillation A process using
high concentrations of salts. specially-designed equipment to permit a
volatile component such as ethanol to be
Condenser A heat exchange device connected separated by distillation from a continuous
to the vapor discharge pipe of a column to flow of an aqueous solution such as beer.
permit the vapor to be cooled and condensed
to a liquid. Condensers are commonly Control loop A portion of a process control
cylindrical vessels containing tubes through system that includes a sensing device
which cooling water is passed. connected to a signal transmitter, a controller,
another signal transmitter and an actuator. For
Congeners Chemical compounds produced example, a pressure control loop on a
with ethanol in the fermentation process. They distillation column may have a pressure
are frequently referred to as impurities. sensor connected through a controller to a
Common congeners are methanol steam flow regulating valve.
acetaldehyde, esters (such as ethyl acetate)
and fusel oils (higher alcohols, particularly Cooker A device for heating a slurry of grain
amyl alcohols). Fermentation conditions may and water to a sufficiently-high temperature
be adjusted to control congener formation for sufficient time to release and gelatinize
depending on the requirements for the end the starch in the grain and thereby render it
product. susceptible to enzymatic hydrolysis. Live
steam is normally used for heating the slurry;
Constant boiling mixture See Azeotrope. and pumps or agitators are used to ensure
Continuous cooker A system into which a mash mixing and even heating. Cooking may be
of water, grain and enzymes may be fed performed continuously or in a batch mode.
continuously to be cooked and discharged to Cooling tower A tower or other type of structure
the fermentation system. Continuous cookers where air (the heat receiver) circulates in direct
generally consist of a slurry tank connected
The Alcohol Alphabet 409
or indirect contact with warmer water (the often important physical and mechanical
heat source) to cool the water. Cooling towers factors in the corrosion process.
are used in ethanol production plants to
recirculate cooling water and to minimize the Co-solvent A liquid required to keep another
amount of water used from wells, rivers or liquid in solution in a third liquid. For
public sources. example, in the Dupont Waiver for the use of
5% methanol in gasoline, 2.5% ethanol (or a
Cooper A person who makes or repairs wooden higher alcohol) is required as co-solvent to
barrels. improve the miscibility of the methanol in the
gasoline.
Cooperage A place where wooden barrels are
made or repaired. Also used to refer to a Crude Oil Windfall Profits Tax Act of 1980
supply of barrels (i.e., the product of the work US federal legislation that extended to 1992
of a cooper). the excise tax exemption for ethanol-gasoline
blends granted under the Energy Tax Act of
COP The abbreviation for cleaning-out-of- 1978. It also introduced a blender tax credit
place. of 40 cents per gallon of ethanol as an
Co-products Where the economics of a alternative to the excise tax exemption.
production process depend on the value of Cyclohexane A colorless, flammable, alicyclic
more than just the primary product, the hydrocarbon liquid of chemical formula
secondary products are referred to as co- C 6H 12 , that boils at 80.3 oC, and freezes at
products rather than as by-products. For 6.5oC. It is used as an alternative to benzene
example, distillers dried grain may be as an entrainer in the dehydration of ethanol
considered a co-product of the production of by azeotropic distillation.
ethanol from dry milled grain.
Cordials See liqueurs.
Corn steep liquor The concentrated liquid that
D
has been used for steeping and softening corn DDG Abbreviation for distillers dried grain.
prior to wet milling. It contains extracted low
molecular weight chemicals from the corn, DDGS Abbreviation for distillers dried grain
and serves as a nutrient source in subsequent with solubles.
fermentation of the starch stream.
DE Abbreviation for dextrose equivalent.
Corn whisky Defined by the BATF as ‘whisky
Deadleg A length of mash piping closed either
produced at under 160 0 proof from a
temporarily by a valve, or permanently to
fermented mash containing not less than 80%
leave a dormant pocket of mash that may
corn grain’. If corn whisky is stored in oak
become a source of bacterial contamination.
barrels, the BATF stipulates that the proof
should not be more than 125 o, and that the Dealer tank wagon (DTW) Used in reference
barrels be used, or uncharred if new. to fuel ethanol and gasoline sales, it is
Furthermore, the whisky cannot be subjected generally of 7,000-8,000 gallons capacity.
to any treatment with charred wood.
Decanter Vessel used for the separation of two-
Corrosion The destruction, degradation or phase liquids. In a fusel oil decanter, an upper
deterioration of material (generally metal) due fusel oil phase is separated from a lower
to the reaction between the material and its aqueous ethanol phase. In a benzene column
environment. The reaction is generally reflux decanter, the upper, mainly-benzene,
chemical or electrochemical, but there are
410 The Alcohol Alphabet
phase is separated from the lower, mainly- The department includes an Office of Alcohol
water, phase. Fuels.
Deficit Reduction Act of 1984 US federal Dephlegmator Name commonly used for the
legislation that increased the excise tax first of two or more condensers attached to
exemption on ethanol-gasoline blends from the overhead vapor line of a distillation
5 cents (as set by the Surface Transportation column. It literally means an entrained liquid
Assistance Act of 1982) to 6 cents per gallon. separator.
It also increased the alternative blender tax
credit to 60 cents per gallon of ethanol. Desiccant A substance that absorbs water and
can be used for drying purposes. For example,
Defoamer See Antifoam. potassium aluminosilicate is used as a
desiccant in molecular sieve systems for
Dehydration The process of removing water ethanol dehydration.
from a substance, particularly the removal of
most of the remaining 5% of water from 1900 Detergent package A combination of detergents
proof ethanol in the production of absolute and corrosion inhibitors normally added to
or anhydrous ethanol. fuel ethanol to impart a cleaning action to
ethanol-gasoline blends and thereby reduce
Demethylizing column Occasionally referred engine fuel injector blockages.
to as a supplementary column, it is a
fractionating column used to remove Dewatering The removal of water (or other
methanol in the production of neutral spirit liquids) from a solid material, particularly the
and is located after the rectifying column. The use of screens or centrifuges in the initial
demethylizing column is heated indirectly via separation of thin stillage (liquid) from the
a reboiler. The impure spirit enters part way solids contained in whole stillage in DDG
up the column and the methanol is removed processing.
in the overhead vapor together with some
ethanol, while the bulk of the ethanol Dextran A non-fermentable, large, branched-
descends to be removed at the base of the chain polymer of glucose molecules produced
column as a product relatively free of from sucrose in molasses by bacterial
methanol. contamination (mainly Leuconostoc
mesenteroides). It gives a ropey appearance
Denaturant A substance added to ethanol to to molasses when stirred or poured, and
make it unfit for human consumption so that reduces ethanol yield on fermentation.
it is not subject to taxation as beverage
alcohol. The BATF permits the use of 2-5% Dextrin Short-chain polymers of glucose
unleaded gasoline (or a large numbered molecules produced by the partial hydrolysis
similar, specified substances) for use as of starch with α-amylase or acid in the initial
denaturants for fuel ethanol. (See also stage of conversion to fermentable sugars.
specially-denatured alcohol). Dextrose An alternative name for glucose.
Department of Energy (DOE) A department Dextrose equivalent (DE) A measure of the
of the US federal government established in degree of hydrolysis of starch. It is no longer
1977 to consolidate energy-orientated considered as significant as previously in
programs and agencies. The department’s ethanol production with the more general
mission includes the coordination and acceptance of simultaneous saccharification
management of energy conservation, supply, and fermentation. The use of HPLC to
information dissemination, regulation, measure fermentable carbohydrate directly
research, development and demonstration. makes DE determination necessary.
The Alcohol Alphabet 411
Diethyl ether The traditional anaesthetic ether, by differences in boiling point by boiling and
which is employed as an entrainer in some recondensing the resultant vapors. In ethanol
azeotropic distillation processes for fuel production, it is the primary means of
ethanol dehydration (as an alternative to the separating ethanol from aqueous solutions.
more commonly-used benzene). It is a
colorless, flammable liquid that boils at Distilled spirits permit (DSP) A permit issued
34.6 o C and freezes at -117 o C. Chemical by the BATF allowing the holder to engage in
formula (C2H5)2O. It readily forms explosive the production or warehousing of
mixtures with air. undenatured ethanol for beverage, industrial
or fuel use.
Differential pressure cell (DP cell) A device
for measuring the difference in pressure of Distillers bushel 56 lbs of any grain, regardless
liquid on either side of a restricting orifice. It of volume.
is used in flow measurement. (See Orifice Distillers dried grain (DDG) The dried residual
meter). by-product of a grain fermentation process.
Dimer A compound produced by linking It is high in protein, as most of the grain starch
together two molecules of a simpler has been removed. It is used as an animal feed
compound (or monomer). It is the simplest ingredient. By strict definition, DDG is
form of polymer. For example, maltose is a produced only from the solids separated from
dimer composed of two linked glucose whole stillage by centrifuging or screening.
molecules. In practice, the term is commonly used to
describe the entire dried stillage residue,
Disaccharide A compound sugar that yields making it synonymous with DDGS.
two monosaccharide units on hydrolysis. For
example, lactose yields glucose and Distillers dried grain with solubles (DDGS)
galactose, sucrose yields glucose and The product derived by separating the liquid
fructose, while maltose yields two glucose portion (thin stillage or solubles) from grain
units. whole stillage by screening or centrifuging,
then evaporating it to a thick syrup and drying
Disc and donut column A beer distillation tower it together with the grain solids portion.
patented by Raphael Katzen with trays
alternately consisting of annular shelves with Distillers dried solubles The product derived
open centers (donuts) and discs (of a slightly from separating the liquid portion (thin
larger diameter than the donut holes) placed stillage) from grain whole stillage,
centrally below the holes. The effect is to give evaporating it to a thick syrup and then drying
a circular curtain of liquid descending from it to a fine powder.
tray to tray, through which the rising vapors Distillers feeds The by-products of fermentation
must pass. It has the advantage of being of cereal grains. See DDG and DDGS.
unaffected by the solids content of the beer
feed or any scaling that may occur in a typical Distillers wet grain (DWG) The wet grain
tray column. residue separated from whole stillage by
screening or centrifuging. It has a very limited
Distilland Material to be distilled, such as beer. storage life and is generally used only where
Distillate The portion of a liquid (distilland) cattle feedlots are located near ethanol
removed as a vapor and condensed during a production plants.
distillation process. Distillery A building or premises where alcohol
Distillation The process by which the is distilled.
components of a liquid mixture are separated
412 The Alcohol Alphabet
Distillery run barrels Recently-emptied, used to vinasse that has been stored for some time
whiskey barrels that have not been sorted to to allow bacterial development prior to being
remove those with or without defects. used as backset in the production of heavily-
flavored rums.
DMA 67Y A detergent produced by Dupont. It
is added to fuel ethanol to ensure that the Dupont waiver The waiver received by Dupont
resultant blend with gasoline will have clean- from the EPA in 1985 for a blend of gasoline
burning characteristics similar to detergent containing a maximum of 5% methanol plus
gasolines supplied by major oil companies. a minimum of 2.5% ethanol or other approved
co-solvent, together with an approved
Downcomer (or downpipe) A device used in proprietary corrosion inhibitor.
distillation columns to allow liquid to descend
from one plate to another. Downcomers may DWG Abbreviation for distillers wet grain.
take the form of one or more round, oval or
rectangular section pipes, or chordal baffles.
The downcomer usually has a weir-type seal
E
at the bottom to prevent vapors passing
upward. Ebulliometer A laboratory instrument used to
measure alcohol concentration in water-
DP cell See differential pressure cell.
ethanol mixtures. The mixture is heated to
Dry degermination A process for the removal boiling and the temperature of the vapor is
of germ from grain without the need for measured accurately and corrected for
steeping and wet milling. It may involve some barometric pressure. The ethanol
pretreatment of the grain to raise the moisture concentration is read from a table of
content before processing. It is used in concentration vs boiling point.
Scotland for corn milling in grain whisky
ED Abbreviation for extractive distillation.
plants and is used in the production of corn
flakes; but is not commonly used in the US Effect In the context of evaporators, the term
ethanol production industry. is used to describe one vessel in a series. For
example a quadruple effect evaporator
Dry milling In the ethanol production industry
consists of four linked vessels.
dry milling refers to the milling of whole dry
grain where, in contrast to wet milling, no EITC Abbreviation for energy investment tax
attempt is made to remove fractions such as credit.
germ and bran. Milling may be carried out
with various types of equipment, including Endoenzyme An enzyme acting on internal
hammer mills and roller mills. portions of a large polymeric molecule rather
than around the periphery. For example, the
DSP Abbreviation for Distilled Spirits Permit. α-amylase enzyme hydrolyzes linkages within
amylose and amylopectin molecules. In
DTW Abbreviation for Dealer Tank Wagon.
contrast, amyloglucosidase acts as an
Dual-flow plate (or tray) A perforated plate exoenzyme by only hydrolyzing the outermost
similar to a sieve plate, without downcomers. linkages.
The perforations are of such a size and open
Energy Policy Act of 1992 US federal
area that the liquid descends by weeping
legislation that amended the provisions of the
through the holes.
Omnibus Reconciliation Act of 1990 to allow
Dunder A Caribbean synonym for vinasse or the proration of the excise tax exemption of
molasses stillage. It is commonly used to refer 5.4 cents per gallon on ethanol gasoline blends
The Alcohol Alphabet 413
that use less than the standard content of 10% Enzyme Any of a class of complex proteinaceous
ethanol. This was to meet the requirements substances (such as amylases and lactases)
of the Clean Air Amendments Act that produced by living organisms that catalyze
required that oxygenate blends with 2.1% or chemical reactions without being destroyed.
2.7% oxygen be used in gasolines in Enzymes may act outside the producing
designated areas suffering from severe air organism and maybe used in industrial
pollution. (These oxygen rates correspond to processes such as saccharification.
ethanol usages of 6% and 7.7%,
respectively). EPA Abbreviation for Environmental Protection
Agency.
Energy Security Act of 1980 US federal
legislation that supported the emerging fuel Ester The product derived by the reaction of an
ethanol industry by establishing an acid with an alcohol or other organic
independent Office of Alcohol Fuels within compound having hydroxyl groups. For
the Department of Energy and authorized example, ethyl acetate is an ester produced
programs of loan guarantees, price by reacting acetic acid with ethanol. Esters
guarantees and purchase agreements with fuel tend to accumulate in distillation in the heads
ethanol producers. at the top of the tower.
Energy Tax Act of 1978 US federal legislation ETBE Abbreviation for ethyl tertiary butyl ether.
that instituted the first excise tax exemption Ethanol Otherwise known as ethyl alcohol,
for gasoline blended with 10% fermentation alcohol, grain-spirit or neutral spirit, etc. A
ethanol. It exempted the blends from the tax clear, colorless, flammable oxygenated
of 4 cents per gallon. It also created an energy hydrocarbon. Chemical formula: C 2H 5OH. It
investment tax credit (EITC) of 10% that has a boiling point of 78.5oC in the anhydrous
applied to equipment for converting biomass state. However, it forms a binary azeotrope
to ethanol in addition to the standard 10% with water with a boiling point of 78.15oC at a
investment tax credit. composition of 95.57% by weight ethanol.
Entrainer A substance used to assist in the Ether One of a class of organic compounds in
dehydration of ethanol by azeotropic which an oxygen atom is interposed between
distillation. Examples are benzene, cyclo- two carbon atoms in the molecular structure.
hexane and pentane. (See azeotropic Ethers may be derived from alcohols by the
distillation). elimination of water. Ethers such as diethyl
Environmental Protection Agency (EPA) A ether and isopropyl ether may be used as
US government agency established in 1970. entrainers in the dehydration of ethanol by
EPA responsibilities include the regulation of azeotropic distillation. Other ethers such as
fuels and fuel additives, including ethanol MTBE and ETBE may be used as octane
gasoline blends. enhancers in gasoline. Ethers are dangerous
fire and explosion hazards. When exposed to
Enzymatic hydrolysis The hydrolysis of a air they form peroxides that may detonate on
polymer by the use of enzymes. In the case heating.
of starch hydrolysis, an α-amylase enzyme
may be used in the initial hydrolysis to Ethyl acetate Chemical formula CH3COOC2H5.
achieve liquefaction and an amylo- A clear, volatile, flammable liquid with a
glucosidase enzyme may be used to complete characteristic fruity odor. It has a pleasant,
the hydrolytic saccharification to fermentable sweet taste when diluted. Boils at 77 oC and
sugars. freezes at -83oC. It is produced by the reaction
414 The Alcohol Alphabet
of ethanol with acetic acid and is a major operated under vacuum (to obtain a lower
component of the esters that give rum its boiling point of the liquid) and may consist
characteristic odor. of a series of interlinked vessels or effects
operating under differing degrees of vacuum.
Ethyl alcohol See Ethanol.
Extractant A substance such as ethylene glycol
Ethyl carbamate Otherwise known as urethane or glycerol be used in extractive distillation
or carbamic acid ethyl ether. A carcinogenic processes for the dehydration of ethanol.
compound produced by heating urea with
ethanol under pressure. Chemical formula Extractive distillation A process where an
NH2COOC2H5. It is very soluble in ethanol extractant is added to a mixture being distilled
or water. Traces of ethyl carbamate may be to change the volatility of one or more
formed in the distillation of beers produced components. The less volatile mixture will then
with the use of urea as a yeast nutrient. It does descend in a continuous distillation column
not present a significant hazard if such ethanol while the more volatile components may be
is only used for fuel purposes, but may removed in the condensed overhead vapors.
present problems in the production of In the use of extractive distillation in the
whiskies. dehydration of ethanol, liquid extractants such
as ethylene glycol, or glycerol, may be used.
Ethylene glycol A slightly viscous, sweet- Salts such as potassium and sodium acetates
tasting, poisonous liquid. Chemical formula may also be used alone in molten form or in
CH 2 OHCH 2 OH. It has a boiling point of mixtures with glycerol, etc. Anhydrous ethanol
197.6oC, and is considerably hygroscopic. It is recovered in the overhead condensate while
is commonly used as an antifreeze. It may the water combines with the extractant to
also be used as an extractant in extractive emerge from the bottom of the column. (This
distillation processes for the dehydration of is the reverse of the situation in azeotropic
ethanol. (See extractive distillation). distillation). The extractant is then separated
Ethyl tertiary butyl ether (ETBE) A colorless, from the water in another column (or an
flammable, oxygenated hydrocarbon. evaporator) and is recycled.
Chemical formula C 2H 5OC 4H 9 . It may be In the production of neutral spirit, light rums
produced from ethanol and tertiary butanol or whiskies, extractive distillation may be
(TBA) or isobutylene. It is of similar structure used to remove fusel oils and some other
to methyl tertiary butyl ether (MTBE), having congeners in the condensed overhead vapors.
similar octane-enhancing properties, but has In this instance the extractant is water, as some
a significantly lower effective Reid vapor of the congeners with lower volatility than
pressure in blending with gasoline. ethanol in a concentrated state may have
Evaporation The process by which a substance higher volatility than ethanol when diluted
in the liquid state is converted into the vapor with water and therefore rise up the extractive
state. distillation column while the ethanol
descends.
Evaporator A device used to evaporate part or
all of a liquid from a solution. In the case of Exoenzyme An enzyme restricted to acting on
grain stillage processing, evaporators may be the outer end of large polymeric molecules
used to concentrate screened thin stillage to and cleaves molecules one by one. (See
a syrup of about 35% solids, which may then Endoenzyme).
be fed to a dryer. Evaporators are normally
The Alcohol Alphabet 415
in suspension by a rising column of gas rather higher boiling points than ethanol and are
than resting on a conventional grate. The generally removed in the distillation process
system gives greater heat transfer and higher to avoid accumulation in the rectifier. They
potential combustion efficiency. The system may be subsequently added back into the
has the advantage that limestone may be anhydrous product for motor fuel grade
added to the coal fuel to absorb sulfur gases ethanol.
to reduce emissions.
Fusel oil decanter A device used to separate
Fossil fuel Any naturally-occurring fuel of an accumulations of fusel oil from ethanol based
organic nature that originated in a past on the fact that fusel oil has a lower miscibility
geologic age such as coal, crude oil or natural in water than ethanol and can therefore be
gas. removed by dilution with water. It generally
consists of a tank with windows or sight glasses
Fractional distillation A process of separating to permit the operator to observe and control
mixtures such as ethanol and water by boiling the separation.
and drawing off the condensed vapors from
different levels of the distillation tower.
Fructose A fermentable monosaccharide G
(simple sugar) of the chemical formula
C6H12O6. Its chemical structure is similar to Galactose A monosaccharide of chemical
that of glucose, but it is sweeter to the taste. It formula C6H12O6 that along with glucose is a
may be produced from glucose by enzymatic constituent of the disaccharide lactose. It is
isomerization, as in the production of high an isomer of glucose, but is less-readily
fructose corn syrup (HFCS). fermented by yeasts to ethanol.
Fuel grade ethanol See motor fuel grade Gas chromatography (GC) A technique for
ethanol. separating chemical substances in which the
sample is carried by an inert gas stream
Fuel ethanol Usually denotes anhydrous through a tube (or column) packed with a
ethanol that has been denatured by addition finely-divided solid material. The various
of 2-5% unleaded gasoline and is intended components in the sample pass through the
for use as an automotive fuel in blends with column at differing velocities and emerge
gasoline. from the column at distinct intervals to be
Fungible Literally means ‘interchangeable in measured by devices such as a flame
trade’. Commonly used to denote products ionization detector or a thermal conductivity
suitable for transmission by pipeline. Ethanol is detector. (Where the solid material in the
not considered fungible in this sense because it column is pretreated with a liquid to achieve
would absorb any water accumulating in the component separation, the process may
pockets in a pipeline. be referred to as gas liquid chromatography).
The technique may be used for separating and
Fusel oil Term used to describe the higher measuring the amount of ethanol and the
alcohols, generally the various forms of various by-products formed in fermentation.
propanol, butanol and amyl alcohol that are
congeners or by-products of ethanol Gas liquid chromatography See gas
fermentation. Normally, predominantly chromatography.
isoamyl alcohol. Their presence in alcoholic Gasohol (or gasahol) A trade name registered
beverages is known to be a cause of by the Nebraska Agricultural Products
headaches and hangovers. The fusel oils have Industrial Utilization Committee, later
The Alcohol Alphabet 417
renamed the Nebraska Gasohol Committee. pump containing two intermeshed gear
(The committee was responsible for laying the wheels in a suitable casing. The counter
groundwork for the development of the rotation of the gears draws fluid between the
present day US fuel ethanol industry). The gear teeth on one side and discharges it on
trade name, in either spelling, covers a blend the other side. It is commonly used for viscous
of anhydrous ethanol ‘derived from liquids such as molasses and stillage syrup.
agricultural products’ with gasoline (not
necessarily unleaded). The committee has Gelatinization In reference to the cooking of
freely granted permission for commercial use starchy feedstocks, gelatinization is the stage
of the trade name provided it is not used for in which the starch granules absorb water and
blends containing alcohols other than ethanol. lose their individual crystalline structure to
become a viscous liquid gel. Gelatinization
Gasoline A volatile, flammable, liquid is significant in that it is the preliminary
hydrocarbon mixture suitable for use as a fuel process necessary to render starch susceptible
in internal combustion engines. Normally to enzymatic hydrolysis for conversion to
consists of a blend of several products from fermentable sugars.
natural gas and petroleum refining together
with anti-knock agents and other additives. It Gin Defined by the BATF as ‘a product obtained
is a complex mixture of hundreds of different by original distillation from mash, or by
hydrocarbons, generally in the range of 4-12 redistillation of distilled spirits, or by mixing
carbon atoms per molecule. The components neutral spirits with or over juniper berries and
may have boiling points ranging mainly other aromatics, or with or over other extracts
between 30 and 200 oC with blends being derived from infusions, percolations or
adjusted to altitude, season and legal maceration of such materials, and includes
requirements. mixtures of gin and neutral spirits. It shall
derive its main characteristic flavor from
Gasoline extender The term used to describe juniper berries and be bottled at not less than
ethanol when it is simply used as a partial 80 0 proof. Gin produced exclusively by
replacement for gasoline without any original distillation or redistillation may be
consideration for its value as an octane further designated as distilled’.
enhancer or oxygenate.
GL Abbreviation for Gay Lussac.
Gay Lussac (GL) The name given to a scale of
the concentration of ethanol in mixtures with GLC Abbreviation for gas-liquid chromatography.
water where each degree is equal to 1% by Glucoamylase An enzyme that hydrolyses starch
volume (i.e., 1 o GL is equivalent to 2 o US into its constituent glucose units. (See
proof). It takes the name from the French Amyloglucosidase).
chemistry pioneer, Joseph-Louis Gay Lussac.
Glucan See Betaglucan.
Gay Lussac equation The equation for the
fermentation of sugar by yeast to carbon Glucanase An enzyme that hydrolyses glucan.
dioxide and ethanol, established by the French (See Betaglucanase).
chemist Joseph-Louis Gay Lussac in 1815:
Glucose A fermentable sugar otherwise referred
C 6 H 12 O 6 → 2CO 2 + 2C 2 H 5 OH. (See
to as dextrose. It is a monosaccharide and has
Stoichiometric yield).
the formula C6H12O6. Glucose is the ultimate
GC Abbreviation for gas chromatograph. product in the hydrolysis of starch and
cellulose, which are both polymers of
Gear pump A positive displacement rotary hundreds or thousands of glucose units.
418 The Alcohol Alphabet
Gram-positive See Gram stain. Heat of vaporization The heat input required
to change a liquid at its boiling point to a vapor
Gram stain A widely-used microbiological at the same temperature.
staining technique that aids in the identification
and characterization of bacteria. It was Hemicellulose Term used to describe non-
devised by Danish physician, Hans Christian cellulosic polysaccharide components of
Gram. Bacteria are described as Gram-positive plant cell walls. The most common
if their cell walls absorb and retain the stain hemicelluloses are composed of polymers of
and Gram-negative if they do not. The xylose (a 5-carbon sugar, or pentose) together
technique is particularly useful in examining with a uronic acid (a sugar acid), arabinose
bacterial contaminants in fermentations, as (another pentose) and mannose (a hexose).
most Gram-positives are susceptible to control Hemicelluloses have no chemical relationship
by penicillin. to cellulose. They frequently surround the
The Alcohol Alphabet 419
cellulose fibers and increase their bonding and High performance liquid chromatography
tensile strength. The presence of hemicelluloses (HPLC) An advanced form of liquid
(and lignins) in close association with cellulose chromatography used for the separation of
tends to impede the extraction and hydrolysis complex mixtures of non-volatile materials
of wood cellulose to sugars for ethanol for analytical purposes. It involves very small
production, as well as its degradation in particles packed in columns through which
nature. the liquids flow and high pressures to increase
the rate of flow and shorten the time required
Hexose A class of monosaccharides (simple to achieve the separation. The process is
sugars) containing six carbon atoms in the coupled with other devices to identify the
molecule. They have the formula C 6H 12O 6. constituents. It may be used in the analysis of
Common examples are glucose, fructose and the sugars and dextrins produced by
galactose. hydrolysis of starch.
HFCS Abbreviation for high fructose corn High test molasses (HTM) See molasses.
syrup.
Hogshead A wooden barrel with a capacity of
Hiag process A process developed in Germany approximately 66 US gallons (250 liters) used
in the 1930s for the dehydration of ethanol in Scotland for aging whisky. It is usually
by extractive distillation using a mixture of constructed from shooked bourbon whiskey
sodium and potassium acetates as the barrels of 52 gallons capacity by using
extractant. additional staves and larger heads and hoops.
High boilers Term used to describe the HPLC Abbreviation for high performance liquid
impurities produced in ethanol fermentation chromatography.
(congeners) that have higher boiling points
than ethanol. Otherwise referred to as tails, HTM Abbreviation for high test molasses.
they include the higher alcohols, propanols,
butanols and amyl alcohols or fusel oils. In Hydrocarbon A member of a class of organic
the context of gasoline, the term refers to chemical compounds containing only hydrogen
components with boiling points considerably and carbon. It should be clearly differentiated
above the mid-range. from carbohydrates, which contain oxygen as
well as hydrogen and carbon. The main natural
Higher alcohols Alcohols having more than two sources of hydrocarbons are petroleum, coal,
carbon atoms. They exist in various isomeric natural gas and bitumens.
forms. As the number of carbon atoms
increases, so does the number of isomers, but Hydrolysis Literally means the breakdown,
at a greater rate. The lower members of this destruction or alteration of a chemical
group, namely propanol, butanol and amyl substance by water. In the case of starch and
alcohol are major constituents of fusel oil. other polymers of glucose, a molecule of water
is divided between two adjacent glucose units,
High fructose corn syrup (HFCS) A product in order to cleave the linkage. For example,
in which a large percentage of the glucose maltose (C 12 H 22 O 11 ), which contains two
derived from starch hydrolysis has been glucose rings, requires the addition of a
converted into its sweeter-tasting isomer molecule of water (H2O) to yield two separate
fructose by use of enzymes. It is frequently glucose molecules (C6H12O6). The hydrolysis
used as a substitute for cane or beet sugar as may be accomplished with the use of acids
a sweetener in soft drinks, etc. It is a seasonal or enzymes.
alternative to ethanol production in some corn
processing plants.
420 The Alcohol Alphabet
vapors are poisonous and at low Jobber A gasoline wholesaler. Some jobbers
concentrations may cause headaches and may operate under contract to one or more
dizziness. major oil companies, distributing gasoline to
branded gas stations. Other jobbers may
Isomer One of a series of two or more molecules operate independently, buying gasoline from
with the same number and kind of atoms and various sources for distribution to unbranded
hence the same molecular weight, but retail outlets. Independent jobbers are
differing in respect to the arrangement or frequently major buyers and blenders of fuel
configuration of the atoms. For instance, ethanol.
glucose and fructose have the formula
C6H12O6, but different molecular structures.
Isomerase An enzyme that can convert a K
compound into an isomeric form. For instance,
the glucose isomerase enzyme converts Karl Fischer titration A method to chemically
glucose into its sweeter-tasting isomer determine the amount of water present in a
fructose in the production of high fructose sample of ethanol and/or other substances.
corn syrup (HFCS). When correctly practiced, the method
provides an extremely accurate measurement
Isomerization The process of converting a of very small quantities of water in ethanol
chemical compound into its isomer such as even if gasoline denaturant is present. (See
converting glucose to fructose in the titration).
production of high fructose corn syrup
(HFCS). In petroleum refining, the term Kerosene One of the three permissible
describes methods used to convert straight- denaturants for fuel ethanol as specified in
chain to branched-chain hydrocarbons, or BATF regulations. It is a refined petroleum
acyclic to aromatic hydrocarbons to increase fraction used as a fuel for heating, cooking
their suitability for high octane gasoline. and for jet engines. It has a boiling point range
somewhat higher than that of gasoline,
Isopropyl ether (IPE) An ether (otherwise generally between 180oC and 290oC.
known as di-isopropyl ether) used in some
fuel ethanol plants as an entrainer in the Kjeldahl method An analytical method for the
dehydration process as an alternative to determination of nitrogen in organic
benzene, etc. It is a colorless, volatile liquid compounds. As nitrogen is an essential
with chemical formula (CH3)2CHOCH(CH3)2, element in protein, the method may be used
which boils at 67.5oC and freezes at -60oC. It for determining protein in such materials as
readily forms explosive mixtures with air. distillers dried grains. Kjeldahl protein, or
Inhalation of vapors may cause narcosis and crude protein, is Kjeldahl N x 6.25.
unconsciousness.
Kluyveromyces fragilis (or marxianus) A
lactose-fermenting yeast used in the
production of ethanol from cheese whey.
J
Kubierschky process The first patented process
Jet cooker An apparatus for the continuous for the continuous dehydration of ethanol with
cooking of grain mashes in which the mash is benzene. With relatively minor variations, the
pumped past a jet of steam that instantly heats process developed in 1914 based on Young’s
the mash to gelatinize the starch. earlier batch process is still used today in fuel
ethanol plants.
422 The Alcohol Alphabet
hydrolyzed by α-amylase (or occasionally by saccharify the barley starch and other
an acid) to give soluble dextrins. This converts additional starch in a mash to yield
the starch mash into a free-flowing liquid, fermentable sugars. (In Scotland, the drying
cutting viscosity. may be done by exposing the malt to a flow
of peat smoke to impart a smoky odor to the
Liqueurs and cordials Defined by the BATF as malt). Malt is used in whisky production,
‘products obtained by mixing or redistilling where it also contributes to product flavor. In
distilled spirits with or over fruits, flowers, fuel ethanol production the necessary
plants or pure juices therefrom, or any other saccharifying enzymes are normally derived
natural flavoring materials, or with extracts from microbial sources.
derived from infusions, percolation or
maceration of such materials, and containing Malt whisky In the US, malt whisky is defined
sugar, dextrose or levulose, or a combination by the BATF as a whisky produced at less than
thereof, in an amount not less than 2.5% by 1600 proof from a fermented mash containing
weight of the finished product’. at least 51% malted barley, and stored at under
1250 proof in charred, new oak barrels. In
Liter A metric measurement of volume defined Scotland, malt whiskies are made from a
as the equivalent of 1000 cubic centimeters 100% malted barley mash and may be aged
or 0.2642 US gallons. in previously-used oak barrels. Malt whiskies
Logarithmic phase Applied to yeast may be mixed with grain whiskies to impart
propagation, it refers to the period in which much of the characteristic flavor of blended
cell numbers increase at an exponential rate Scotch whisky.
after the initial lag phase. Maltase An enzyme capable of hydrolyzing
Low boilers In reference to ethanol distillation, maltose sugar molecules into their two
the term is applied to the congeners, or component glucose units. Occasionally the
fermentation by-products, which boil at a name is loosely applied to the
lower temperature than ethanol. More amyloglucosidase enzyme, which hydrolyzes
commonly referred to as heads, these polysaccharides to glucose.
compounds are principally aldehydes and Maltose A fermentable sugar which is a dimer
methanol. (or disaccharide) of glucose in that it is
LPA Abbreviation for liters of pure alcohol. comprised of two linked glucose units. It is
the normal end product of starch
saccharification by the ß-amylase enzyme in
malt.
M
Manioc See cassava.
Macromolecule A giant molecule in which there
is a large number of one or more relatively Mannose A fermentable 6-carbon sugar (or
simple structural units or monomers. hexose) which is an isomer of glucose.
Chemical formula C6H12O6. It is a constituent
Malt Barley grains that have been steeped in of hemicellulose.
water and then allowed to germinate. The
germination is normally halted by drying the Mash A mixture of milled grain or other
grains when the sprouts are about the same fermentable carbohydrate in water used in
length as the grains. At this stage, the malt (or the production of ethanol. The term may be
‘malted barley’) contains considerable used at any stage from the initial mixing of
amounts of α- and ß-amylase enzymes that the feedstock in water prior to any cooking
424 The Alcohol Alphabet
Microorganism A collective term for feedstock when prices are favorable, and has
microscopic organisms including bacteria, the advantage over blackstrap of causing less
yeasts, viruses, algae and protozoa. distillation column scaling. However, it
requires more nutrients for fermentation. (See
Milo (or millet, or grain sorghum) Much also citrus molasses).
smaller than corn, this cereal grain has a
similar starch content and yields almost the Mole A gram mole is the quantity of a chemical
same amount of ethanol per bushel on compound or element equal to its molecular
fermentation. Milo is more drought-resistant weight in grams. By this definition, moles of
than corn and is frequently grown in areas different compounds or elements contain the
unsuited to corn. Milo fermentations may tend same number of molecules. Thus in the vapor
to foam more than corn fermentations and state, as ideal gases, moles of different
will produce a characteristic surface crust if compounds occupy the same volume at the
not agitated. same temperature and pressure. This leads
distillation equipment designers to base their
Molar solution A solution of salts or other calculations on the mole percentages of
substances in which one molecular weight of different compounds in a mixture to be
the substance in grams (one mole) is dissolved separated, such as ethanol and water, rather
in enough solvent to make up to one liter. than the percentages by weight or volume.
Molarity (or molar concentration) is a measure
of moles per liter of a solution and is indicated Molecule The smallest unit into which a pure
by the letter M as in 1M, 10M, etc. substance can be divided and still retain the
composition and chemical properties of the
Molasses The thick liquid remaining after substance. For example the formula for water
sucrose has been removed from the mother is H2O; and the smallest unit recognizable as
liquor (of clarified concentrated cane or beet water is the single molecule of two atoms of
juice), in sugar manufacture. Blackstrap hydrogen linked to one atom of oxygen.
molasses is the syrup from which no more
sugar may be removed economically. It has Molecular sieve A microporous substance
usually been subjected to at least three composed of materials such as crystalline
evaporating and centrifuging cycles to aluminosilicates belonging to a class known
remove the crystalline sucrose. Its analysis as zeolites. The size of the pores in the
varies considerably, depending on many substance may vary with its chemical
factors including sugar mill equipment and structure, being generally in the range of 3 to
operational efficiency; but it may contain 10 Angstrom units in diameter. With material
approximately 45-60% fermentable sugars by having a very precise pore size, it is possible
weight and approximately 10% ash (or to separate smaller molecules from larger ones
minerals). It is commonly used as an ethanol by a sieving action. For example, in ethanol
feedstock when prices are favorable. High test dehydration with a potassium aluminosilicate
molasses (HTM) is not a true molasses as it is material prepared with pores of a diameter of
the mother liquor from which no crystalline 3 Å units, water molecules with a diameter of
sugar has been removed by centrifugation but 2.5 Å may be retained by adsorption within
which has been treated with acid to reduce the pores while ethanol molecules of a
crystallization. It may contain approximately diameter of 4 Å cannot enter and therefore
80% sugars by weight and is very low in ash. flow around the material.
Typically HTM is only produced in years
when the sugar price does not justify its The term molecular sieve is frequently used
recovery. It may be used as an ethanol loosely to describe the entire ethanol
dehydration apparatus holding the beads of
426 The Alcohol Alphabet
sieve material and includes the equipment and rating of an automobile fuel. (See Octane
controls necessary to regenerate them when rating).
saturated with water.
MSW Abbreviation for municipal solid waste.
Molecular weight The sum of the atomic
weights of the atoms in a molecule. For MTBE Abbreviation for methyl tertiary butyl
example, water molecules are composed of ether.
two atoms of hydrogen (atomic weight of one) Multiple effect evaporator A system
and one atom of oxygen (atomic weight of comprising a series of interlinked evaporator
16) to give a total molecular weight of 18. vessels (or ‘effects’) operating under
MON Abbreviation for motor octane number. increasing degrees of vacuum. As liquids boil
at lower temperatures with increasing
Monomer A single molecule of a substance of vacuum, it is possible to use the latent heat
relatively low molecular weight and simple of the vapors from one vessel to heat the next
structure, which is capable of conversion to (and so on through a series of up to about six
polymers by combination with other identical vessels under increasing vacuums) with steam
or similar molecules. For example, glucose is or other external source of heat only being
a monomer, which by interlinking of introduced to the first vessel. This allows
molecules can be built into large polymers considerable economy in energy consumption.
such as the amylose and amylopectin Multiple effect evaporators are commonly
contained in starch. used for concentrating thin stillage solids into
a syrup.
Monosaccharide A sugar monomer. Examples
are glucose, fructose and galactose. These are Municipal solid waste (MSW) Regular city
the simplest forms of sugars, which are more trash or garbage. There have been numerous
readily fermented by yeasts than their proposals and some pilot-scale attempts to
polymers (or polysaccharides). produce ethanol from MSW by hydrolyzing
and fermenting the cellulosics contained in the
Mother yeasting A system of yeast propagation material. Such efforts have, however,
frequently used for molasses fermentations in generally run into problems due to the normal
which the propagator is not emptied entirely wide variability of MSW.
when inoculating a fermentor and the portion
retained is used for starting another yeast MVR Abbreviation for mechanical vapor
propagation cycle. By adding acid to maintain recompression.
a low pH in the propagator it may be possible
to repeat the process for numerous cycles Mycotoxin A toxic substance produced by
without excessive bacterial contamination. molds.
analysis of incoming grains, profiling (or Nox) where x represents any proportion
fermentors, distillate analysis, dry house of oxygen to nitrogen.
operations, and finished products.
Normal solution A solution containing one
Net energy balance The amount of energy equivalent weight of a dissolved substance
available from a fuel by combustion, less the per liter. One volume of a normal acid will
amount of energy taken for its production. In neutralize an equal volume of a normal alkali
the early 1980s a lot of emphasis was placed (or vice versa). This principle is used in
on the need for a net energy balance in the measuring the acidity in fermentations by
production of fuel ethanol to be used as a titration with a standardized alkali. The
gasoline extender. The emphasis was later ‘normality’ of solutions is indicated by the
reduced when it became appreciated that a) letter N, as in 0.1N, 2N, etc.
much of the energy used in ethanol
production may come from low quality
sources such as coal, wood or bunker-C oil,
which cannot be used as automobile fuels,
O
and b) that ethanol has a value as an octane Occupational Safety and Health
enhancer and oxygenate and not just as a Administration (OSHA) An agency of the
gasoline extender. In recent work the net US Department of Labor with the mission of
energy balance in corn to ethanol conversion developing and promulgating occupational
is 1.3 t. safety and health standards, developing and
issuing regulations and conducting
Neutral spirit Defined by the BATF as ‘distilled
inspections and investigations to ensure their
spirits produced from any material at or above
compliance.
190 0 proof’. In practice, neutral spirit is
purified, odorless, tasteless and colorless Octane A flammable liquid hydrocarbon of
ethanol produced by distillation and chemical formula C8H18 found in petroleum.
rectification techniques that remove any One of the eighteen isomers of octane, 2,2,4-
significant amount of congeners. It is used in trimethylpentane is used as a standard in
the production of beverages such as vodka, assessing the octane rating of fuels.
gin, cordials and cream liqueurs.
Octane enhancer Any substance such as
Nitrogen oxides Air-polluting gases contained ethanol, methanol MTBE, ETBE, benzene,
in automobile emissions, which are regulated toluene, xylene, etc., that will raise the octane
by the EPA. They comprise colorless nitrous rating when blended with gasoline.
oxide (N 2O) (otherwise know as dinitrogen
monoxide or as the anaesthetic ‘laughing Octane rating (or octane number) A laboratory
gas’), colorless nitric oxide (NO) and the assessment of a fuel’s ability to resist self-
reddish-brown nitrogen dioxide (NO2). Nitric ignition or ‘knock’ during combustion in a
oxide is very unstable and on exposure to air spark-ignition engine. A standardized-design,
it is readily converted to nitrogen dioxide, single-cylinder, four-stroke engine with a
which has an irritating odor and is very variable compression ratio is used to compare
poisonous. It contributes to the brownish layer the knock resistance of a given fuel with that
in the atmospheric pollution over some metro- of reference fuels composed of varying
politan areas. Other nitrogen oxides of less proportions of two pure hydrocarbons. One
significance are nitrogen tetroxide (N2O4) and component is the octane isomer 2,2,4-
nitrogen pentoxide (N 2O 5). Nitrogen oxides trimethylpentane (sometimes referred to as
are sometimes collectively referred to as NOx isooctane), which has a high resistance to
428 The Alcohol Alphabet
knock and is given the arbitrary rating of 100. pressure across a restriction (or ‘orifice plate’)
The other component, heptane, has a very low placed in the flow stream.
knock resistance and is given the base rating
of zero. For fuels with a rating higher than OSHA Abbreviation for Occupational Safety
100 octane, the rating is obtained by and Health Administration.
determining how much tetraethyl lead needs Overhead vapors The vapors emerging from
to be added to pure isooctane to match its the top of a distillation column that are
knock resistance. conducted to a condenser system.
The engine knock tests are performed under Oxygenated fuels Literally meaning any fuel
two sets of operating conditions, the so-called substance containing oxygen, the term is
‘motor’ or M method, and the ‘research’ or R commonly taken to cover gasoline-based fuels
method. The average of the two results is containing such oxygen-bearing compounds
taken to be a good indicator of a fuel’s as ethanol, methanol, MTBE, ETBE, etc.
performance in a typical automobile on the Oxygenated fuel tends to give a more
road. Hence, the use of the (R+M)÷2 octane complete combustion of its carbon to carbon
rating on labels on service pumps. dioxide (rather than monoxide) and thereby
OFA Abbreviation for Oxygenated Fuels to reduce air pollution from exhaust emissions.
Association.
Oligomer A macromolecule formed by the
P
chemical union of two, three or four identical
units known as monomers. The oligomers of Packed distillation column A column filled with
two units are dimers, three units are trimers a packing of ceramic, metal or other material
and four units are tetramers. (The prefix oligo- designed to increase the surface area for
means ‘few’). contact between liquids and vapors.
Oligosaccharide Short-chain polymers of PADD Abbreviation for Petroleum Administration
simple sugars (or monosaccharides) for Defense District.
generally considered to cover the range of 2-
8 units. Short dextrins produced by hydrolysis Pasteurization A method devised by Pasteur
of starch are included in this category. to partially sterilize a fluid by heating to a
specific temperature for a specific length of
OPIS Abbreviation for Oil Price Information time. Pasteurization does not greatly change
Service. chemical composition, but destroys
undesirable bacterial contaminants. The
Organic Adjective for chemical compounds
practice may be applied to molasses or other
containing carbon and hydrogen with or
liquid fermentation feedstocks to reduce the
without oxygen, nitrogen or other elements.
initial level of bacterial contamination.
Organoleptic testing The quality control
Patent A certificate or grant by a government
process of checking samples of alcoholic
of an exclusive right with respect to an
products on the basis of odor and taste. It is
invention for a limited period of time. In the
normally performed by comparing samples
US the effective period is normally 17 years
of new production with older samples of
from the date of granting. Patents may cover
acceptable quality that have been designated
processes, machines or other apparatus,
as standards.
methods of manufacture, composition of
Orifice meter An instrument used to measure materials and designs. Patents are published
fluid flow by recording the differential and freely available and are required to
The Alcohol Alphabet 429
Plate (or tray) A contacting device placed closed off without incorporating a pressure
horizontally at intervals within a distillation relief or recycle system. Positive displacement
column. Plates may be simple perforated discs pumps are frequently used for moving
with or without downcomers as in the sieve molasses and stillage syrups, yeast slurries and
plate and the dual flow plate, or they may fuel oil. They are also effective for metering
have bubble caps, tunnel caps or various since every rotation or stroke moves a precise
types of floating valves to improve the contact volume from the suction side to the discharge
between the rising vapor and descending side regardless of the pressure required.
liquid. Sieve plates and tunnel caps are the
most common form of plates used in ethanol Potassium aluminosilicate A zeolite. (See
production facilities. molecular sieve).
Plate (or tray) distillation column A distillation Pot still A simple batch distillation unit used
column with horizontally arranged contacting for the production of heavily-flavored
plates located at intervals up the column; as distillates for beverage use. It consists of a
compared to a packed column, which is filled tank (heated either by an internal steam coil
randomly with contacting devices. or by an external fire) and an overhead vapor
pipe leading to a condenser. It may be used
Polymer A macromolecule formed by the in the production of heavily-flavored rums
chemical union of identical combining units and whiskies.
known as monomers. While generally
referring to a combination of at least five Pre-fermentor See yeast propagator.
monomers, in many cases the number of Proof A measure of the absolute ethanol content
monomers is quite large, such as the average of a distillate containing ethanol and water.
of 3,500 glucose monomers in cellulose. In the US system, each degree of proof is
Polysaccharide A polymer composed of equal to 0.5% ethanol by volume, so that
numerous sugar monomers or mono- absolute ethanol is 2000 proof. In the Imperial
saccharides. Examples are cellulose and the system 100 proof is equal to 57.06% ethanol
amylose and amylopectin in starch. For by volume, or 48.24% by weight, while
fermentation purposes, polysaccharides must absolute ethanol is 75.25 over proof, or
be subjected to hydrolysis to yield their 175.25 proof.
component fermentable monosaccharides. Proof gallon (PG) The volume of a liquid that
Positive displacement pump A pump in which contains the equivalent of one gallon of
the displacement is accomplished by ethanol at 1000 proof.
mechanically moving a volume of fluid from Proof tables Books of tables compiled by the
the pump suction to the discharge with no BATF and other organizations for the
fluid slippage within the pump - no matter calculation of proof of ethanol-containing
how high the discharge pressure. With every liquids with adjustments for varying
stroke or rotation the same volume of fluid is temperatures.
pumped. Positive displacement pumps may
take various forms such as reciprocating Propagation The process of increasing numbers
piston pumps, gear pumps, lobe pumps and of organisms by natural reproduction. In the
diaphragm pumps. Generally, they are used case of yeast, propagation occurs by budding
to move relatively low volumes of fluid at to form new cells in both aerobic and
high pressures. As they force a positive anaerobic conditions. Yeast companies grow
displacement, they may not be used in yeast under aerobic conditions with high
systems where the output may be throttled or
The Alcohol Alphabet 431
yield. Fuel alcohol propagation of yeast is Recoopered barrels Used wooden barrels that
normally under anaerobic conditions (even have been repaired to replace any cracked
with air supplied) due to the presence of high staves or heads to be suitable for reuse in the
amounts of sugar. (See yeast propagator). aging of alcoholic beverages.
Propanol (or propyl alcohol) A minor Rectification The process of concentrating and
constituent of fusel oil. Chemical formula purifying ethanol or other materials in a
C 3H7OH. It exists as either of two isomers. rectifying column. The process may be
Both are colorless, toxic, flammable liquids operated simultaneously with beer distillation
with odors similar to that of ethanol. with the rectifying column directly connected
to a beer stripping column. In beverage
Protein Any of a class of high molecular weight ethanol production the rectification process
polymer compounds composed of a variety may involve concentration and removal of
of linked amino acids. Proteins are an essential fusel oil, esters and heads. In fuel ethanol
ingredient in the diets of animals and humans. production the process commonly only
Yeast dry matter is approximately 50% involves concentration and the removal of
protein. Corn DDG generally contains over fusel oil. In some instances, where benzene
27% crude protein. is to be used as an entrainer in the subsequent
PSI Abbreviation for pounds per square inch. dehydration process, the heads may also be
Atmospheric pressure is 14.7 psi, 1 bar is 14.5 removed in the rectification process.
psi. Rectifying column (rectifier, rectification
PSIG Abbreviation for pounds per square inch column or rectifying section) The portion of
gauge. a distillation column above the feed tray in
which rising vapor is enriched by interaction
with a countercurrent descending stream of
condensed vapor. In the case of ethanol
R production, the rectifying column may have
valves at draw-off points on various trays to
Rack price Used in reference to gasoline and
allow the removal of accumulated fusel oil and
fuel ethanol, it is the wholesale price at the
other congeners. In some instances the
tank-truck loading terminal exclusive of any
ethanol product is withdrawn from the reflux
federal, state or local taxes.
of the overhead vapor condensate, but in
Reboiler A device for supplying heat to a others the product may be drawn from one of
distillation column without introducing live the upper trays to permit the heads or low
steam. It generally consists of a shell-and-tube boilers to be purged from the reflux. (See
heat exchanger connected to the base of the Concentrating column).
column with liquid from the column entering
Reflux The portion of the condensed overhead
inside the tubes to be heated indirectly by
vapors returned to a distillation column to
steam on the shell side.
maintain the liquid-vapor equilibrium.
Receiver A tank into which new distillates flow
Reflux ratio The ratio of the amount of
from the still for verifying proof, quantity and
condensate refluxed to the amount withdrawn
quality before transfer to shipping or storage
as product. Generally, the higher the reflux
tanks. Where the tank is sealed or has locks
ratio the greater the degree of separation of
on the valves to meet government excise
the components in a distillation system.
regulations for preventing unlawful access, it
may be referred to as a ‘closed receiver’. Refractometer An instrument used to measure
432 The Alcohol Alphabet
the refractive index of liquids and liquid the grain, and they may turn at differing speeds
solutions. Refractometers can be calibrated in to increase the abrasion. Roller mills are
concentration of a solute instead of refractive suitable for small grains such as wheat, but
index. In the ethanol industry refractometers do not perform as well as a hammer mill on
are calibrated as degrees Brix or degrees corn.
Balling.
RON Abbreviation for research octane number.
Reid vapor pressure (Rvp) A measure of the
vapor pressure in pounds per square inch of Rotameter A flow meter with a float and a
a sample of gasoline at 100 o F. It is an variable area flow tube.
indication of the volatility of a gasoline. The Rum Defined by the BATF as ‘an alcoholic
blending of ethanol with gasoline tends to distillate from the fermented juice of
increase Rvp while the blending of MTBE, and sugarcane, sugarcane syrup, sugarcane
more particularly ETBE, tends to reduce Rvp. molasses, or other sugarcane by-products,
Renewable Fuels Association (RFA) The produced at less than 1900 proof, in such a
Washington DC-based trade association for manner that the distillate possesses the taste,
the US fuel ethanol industry. Its membership aroma and characteristics generally attributed
includes companies involved in the to rum’. Unlike the specifications for whiskies,
production, blending and marketing of the BATF does not require that rum be aged
ethanol-blended fuels. in oak barrels. British regulations specify that
rum be produced ‘from sugarcane products
Research octane number (RON) See Octane in sugarcane-growing countries’.
rating.
Rvp Abbreviation for Reid vapor pressure.
Reverse osmosis A technique used in water
purification and wastewater and stillage Rye whisky Defined by the BATF as whisky
treatment in which pressure is applied to the produced at not more than 1600 proof, from a
liquid in a suitable apparatus to force pure fermented mash of not less that 51% rye and
water through a membrane that does not allow stored at not more than 1250 proof in charred,
the passage of dissolved ions. new oak barrels.
Saccharomyces A genus of unicellular yeasts top of a stack of sieves with increasingly fine
of the family Saccharomycetaceae mesh sizes descending downwards. The sieve
distinguished by the general absence of stack is vibrated for a standard time period
mycelium and by their facility to reproduce and the weight percentage retained on each
asexually by budding. This genus includes screen is determined. With hammer mills, the
the species Saccharomyces cerevisiae, which sieve analysis will generally show that the
is the yeast most commonly used by bakers, meal gradually becomes more coarse as the
brewers, distillers and wine producers. hammers wear and need turning or
replacement.
Sanitizing In processing systems, sanitizing is
reducing the population of viable organisms Sieve plate (or sieve tray) A distillation column
in the equipment to a level that is acceptable plate with perforations of precise number and
for the process. size such that the ascending vapor passes
through the plate vertically to mix with the
Scale A deposit formed on the surface of descending liquid held on the plate. A sieve
equipment due to formation of insoluble plate differs from a dual flow plate in that it is
inorganic compounds. designed for the descending liquid to
Scale inhibitor See Antiscalant. overflow down a downcomer, instead of
weeping through the perforations.
Scaling The precipitation of salts in a distillation
column or heat exchanger which if Simultaneous saccharification and
uncontrolled may reduce capacity and fermentation (SSF) A procedure in which
eventually block the equipment and make it saccharification of a cooked (gelatinized and
unusable. Scaling occurs because some salts liquefied) starch mash occurs in the fermentor
such as calcium sulfate and oxalate are less (by addition of glucoamylase) simultaneously
soluble at high temperatures and/or in the with the commencement of fermentation (by
presence of ethanol. Scaling problems are yeast). This procedure is replacing the
more common in molasses beer distillation, traditional process taken from the whisky
but may occur in grain beer distillation when industry in which there is a specific holding
conducted under pressure (i.e., at high stage for saccharification with malt or
temperatures). microbial enzymes (in a sacc’ tank) before the
mash goes to a fermentor.
Scrubber A device for the removal of entrained
liquid droplets in a gas stream. Scrubbers may Slurry tank The vessel in which grain meal is
be used to recover ethanol from the carbon mixed into a slurry with water and enzyme
dioxide vented from fermentors. before being pumped through a continuous
cooking system.
SDA Abbreviation for specially denatured
alcohol. Small Business Administration (SBA) A US
federal government agency charged with
SG Abbreviation for specific gravity. encouraging and assisting the development
of small businesses. It offers financial
Shooked barrel (or shook) A used bourbon
assistance such as direct loans, loan
whisky barrel that has been dismantled to
guarantees and direct participation loans for
reduce the space requirements for
virtually any legitimate small business
transportation.
development. The SBA has assisted in the
Sieve analysis A laboratory test made on grain financing of many small fuel ethanol facilities.
meal to check that the milling process is being
Soil Any unwanted material that is left on a
conducted correctly. The meal is added to the
surface that needs to be clean.
434 The Alcohol Alphabet
Solute One or more substances dissolved in (amylose and amylopectin), both of which are
another substance (a solvent) to form a composed of glucose monomers linked by
solution. The solute is uniformly dispersed in glycosidic bonds. Starch is a principal energy
the solvent in the form of molecules (as with storage product of photosynthesis and is
sugars) or ions (as with salts). found in most roots, tubers and cereal grains.
Starch may be subjected to hydrolysis
Solution A uniformly dispersed mixture at the (saccharification) to yield dextrins and
molecular or ionic level of one or more glucose.
substances (the solute) in one or more other
substances (the solvent). Ethanol and water Sterilization The complete destruction of all
form a liquid-liquid solution, while sugar and organisms including viruses and spores.
water form a solid-liquid solution.
Still An apparatus used in distillation
Solvent A substance capable of dissolving comprising a vessel in which the liquid is
another substance (a solute) to form a vaporized by heat and a cooling device in
uniformly-dispersed mixture (solution) at the which the vapor is condensed. It may take
molecular or ionic level. the form of a batch or continuously operable
unit.
Specially denatured alcohol (SDA) The term
used to describe ethanol denatured with any Stillage The mixture of non-fermentable (or
formulation of compounds selected from a list non-fermented) dissolved solids, insoluble
approved by the BATF. The denaturant renders grain fines proteins, dead yeasts and water,
the ethanol unfit for beverage purposes which are the residues after removal of
without impairing its usefulness for other ethanol from a fermented beer by distillation.
applications. Stillage may be dried to recover the solid
material (as DDG in the case of grain
Specific gravity (SG) The ratio of the density feedstocks).
of a material to the density of a standard
reference material such as water at a specific Stoichiometric yield The theoretical yield of a
temperature. For example, the specific gravity product of a chemical reaction as calculated
of ethanol is quoted as 0.7893 at 20/4oC. This from the chemical reaction equation. For
means that a given volume of ethanol at 20oC example, in the Gay Lussac equation for the
weighs 0.7893 times the weight of the same fermentation of glucose to ethanol by yeast,
volume of water at 4oC. C6H12O6 → 2CO2 + 2C2H5OH, 100 parts (by
weight) glucose should yield 48.89 parts
Spent sulfite liquor (SSL) See sulfite waste carbon dioxide and 51.11 parts ethanol. Due
liquor. to a variety of factors, this yield is not
Spirit whisky Defined by the BATF as a mixture achieved in practice.
of neutral spirits with at least 5% on a proof Stoichiometry The branch of chemistry that
gallon basis of whisky or straight whisky. deals with the quantities of substances that
SSF Abbreviation for simultaneous sacchar- enter into and are produced by chemical
ification and fermentation. In the enzyme reactions.
industry SSF also is used to indicate solid- Stover The dried stalks and leaves remaining
state fermentation. from a crop, particularly corn, after the grain
SSL Abbreviation for spent sulfite liquor. (See has been harvested. It is of interest as a
sulfite waste liquor). potential source of cellulose feedstock for
ethanol production.
Starch A mixture of two carbohydrate polymers
The Alcohol Alphabet 435
Straight whisky Defined by the US Bureau of in some paper mills. It partly consists of a
Alcohol, Tobacco and Firearms as a product dilute solution of sugars produced by the acid
conforming to the requirements for the whisky hydrolysis of cellulose. It may be used as a
designation, which has been stored in feedstock for the production of ethanol by
charred, new oak barrels for a period of at fermentation using selected yeast strains after
least two years. The straight whisky definition stripping out the sulfite (or sulfur dioxide) with
may include mixtures of straight whiskies of steam.
the same grain type, produced at the same
distillery, all of which are not less than four Supplementary column See Demethylizing
years old. column.
Strain See Yeast strain. Sweet sorghum A plant of the species Sorghum
bicolor, closely related to milo or grain
Stripping column (or stripping section) The sorghum, which is cultivated primarily for the
portion of a distillation column below the feed sweet juice in its stem. It is widely used for
tray in which the descending liquid is cattle fodder and silage. The juice contains
progressively depleted of its volatile sucrose, which may be extracted by crushing
components by the introduction of heat at the with modified sugar mill equipment. The juice
base. may be used in the fermentative production
of ethanol. The plant has an advantage over
Sub-octane blending The practice of blending sugarcane in that it can be grown over a much
ethanol with selected gasoline components wider range of climatic regions.
that have octane ratings below the
commercially or legally acceptable level. It SWL Abbreviation for sulfite waste liquor.
takes advantage of the fact that ethanol can
enhance the octane rating of the mixture to Synthetic ethanol Ethanol produced by any of
bring it to an acceptable level. several synthetic processes such as the
catalytic hydration of ethylene, the sulfuric
Sucrose Common table sugar, derived from beet acid hydration of ethylene and the Fischer-
or cane sources. It is a disaccharide Tropsch process, in which it is a major
comprising a monomer of glucose linked to a by-product of the synthesis of methanol by
monomer of fructose. It has a chemical catalytically reacting carbon dioxide and
formula C12H22O11. It is directly fermentable hydrogen. Synthetic ethanol is chemically
by common yeasts to produce ethanol. identical to fermentation ethanol, but does not
qualify for US federal or state incentives for
Sugar Any of a class of water-soluble, simple blending with gasoline and may not be used
carbohydrate, crystalline compounds that in the production of alcoholic beverages.
vary widely in sweetness including the
monosaccharides and lower oligosaccharides.
Sugars may be chemically reducing or non-
reducing compounds and are typically T
optically active. Examples include the
Tafia An unaged Caribbean or South American
monosaccharides glucose, fructose, mannose
alcoholic beverage produced by batch
and xylose, and the disaccharides sucrose,
distillation of beers obtained by the
maltose, lactose, and the trisaccharides
fermentation of sugarcane juice or molasses.
raffinose and maltotriose.
It is similar to aguardiente and rum.
Sulfite waste liquor (SWL) An effluent
Tails See high boilers.
produced in the sulfite pulping process used
436 The Alcohol Alphabet
TBA Abbreviation for tertiary butyl alcohol. (See Theoretical yield See stoichiometric yield.
butanol).
Thermal efficiency The ratio of the energy
Tequila Defined by the BATF as ‘an alcoholic output of a process to the energy input.
distillate from a fermented mash derived
principally from Agave tequilana Weber Thermal vapor recompression A method using
(‘blue’ variety) with or without additional the same recompression and recirculation
fermentable substances, distilled in such a principles as in mechanical vapor
manner that the distillate possesses the taste, recompression (to enhance the efficiency of
aroma and characteristics normally attributed evaporation) except that recompression is
to tequila. It is a distinctive product of performed by passing steam through a venturi
Mexico, manufactured in that country in jet on the vapor line.
compliance with the laws regulating the Thermophilic Literally meaning ‘heat loving’,
manufacture of tequila for consumption in that it refers to microorganisms such as bacteria
country’. Tequila may be matured in wooden that thrive at warm temperatures, generally
containers to acquire a light gold color. in the range of 35-60oC. Lactobacillus is an
Ternary azeotrope An azeotrope or constant example of a thermophilic bacterium that
boiling mixture made up of three components. multiplies rapidly at around 55oC. Appreciation
For example, a mixture of 74% volume of this fact has been a reason why the use of
benzene, 18.5% ethanol and 7.5% of water sacc’ tanks has fallen into disfavor.
forms an azeotrope boiling at 64.9oC. Thermosacc TM Stress-tolerant (heat and
Tetramer A macromolecule produced by ethanol) strain of Saccharomyces cerevisiae
linking four identical units known as used in ethanol production.
monomers. (See oligomer). Thin stillage The liquid portion of stillage
Tetraethyl lead (or lead tetraethyl) An organo- separated from the solids by screening or
metallic compound used as an octane centrifuging. It contains some yeast,
enhancer in gasoline. Its use is regulated by suspended fine particles and dissolved
the EPA to control air pollution. If added to material. It is normally sent to an evaporator
gasoline on its own, tetraethyl lead would to be concentrated to a thick syrup and then
leave combustion residues of lead and lead dried with the solids portion to give DDGS.
oxide on engine cylinder walls. To prevent Titration A method of volumetrically
accumulation of these deposits, ethylene determining the concentration of a substance
dibromide and dichloride are also added to in solution by adding a standard solution of
convert the lead to volatile halides before known volume and strength until the reaction
release into the atmosphere. between the two is completed, usually as
Theoretical plate (or tray) A distillation indicated by a change in color due to an
column plate or tray that produces perfect added chemical indicator. For instance, it is
distillation i.e. it would yield the same common practice to perform a titration with
difference in composition between liquid and 0.1 Normal sodium hydroxide solution using
vapor as that normally existing between the phenol-phthalein as indicator to determine the
liquid and the vapor when in equilibrium. A acidity of a mash or beer.
distillation column giving the same separation Toluene An aromatic hydrocarbon compound
as 10 simple theoretical distillations is said to that can be used as an octane enhancer in
have 10 theoretical plates. gasoline. It is a colorless, flammable liquid
with a benzene-like odor. Chemical formula
The Alcohol Alphabet 437
C6H5CH3. It boils at 110.7oC and freezes at - with the mission to improve and maintain farm
94.5 o C. It is derived from the catalytic income, to develop and expand markets for
reforming of gasoline or the distillation of coal agricultural products, etc.
tar light oil. It is known to be carcinogenic.
Urethane See Ethyl carbamate.
Total sugars as invert (TSAI) A simple crude
analytical measure of reducing sugars in USDA Abbreviation for United States
molasses. Department of Agriculture.
Transglucosidase An enzyme with the opposite US gallon A measure of 231 cubic inches liquid
effect of amyloglucosidase in that it brings at 60oF. It is the equivalent of 3.785 liters in
about a polymerization of glucose into the metric system or 5/6 of an Imperial gallon.
polysaccharides. It may be present as an
impurity in commercial enzyme preparations.
V
Trimer A macromolecule produced by the
linking three identical units known as Vacuum distillation A process that takes
monomers. (See oligomer). advantage of the fact that liquids boil at lower
temperature under reduced pressure. Thus,
Tray See plate.
distillation under vacuum may be used for
TSAI Abbreviation for total sugars as invert. substances that otherwise have a relatively
high boiling point such as ethylene glycol
Tunnel cap A contacting device used (which may be used in the dehydration of
occasionally on distillation plates. It is ethanol). The process may also be used in the
essentially an elongated bubble cap, distillation of ethanol-water mixtures, as the
consisting of a chimney portion and a domed azeotrope is formed at a higher proof under
cover that deflects the vapors rising through lower pressures. Vacuum distillation may be
the chimney, causing them to pass down used in the production of rum or ethanol from
through the liquid layer on the plate. Tunnel molasses to reduce the incidence of scaling
caps may be made from interlocking metal in the column that occurs at high
channels to form an integral part of the plate. temperatures.
Vacuum fermentation A process of operating
a fermentation under vacuum, so that the
U ethanol or other product is vaporized and
Upgrading A term used loosely to mean the removed as it is formed to avoid its
dehydration of hydrous ethanol. concentration becoming inhibitory to the
yeast. In a patented variation known as the
Ultrafiltration A process for the separation of Vacuferm process, instead of maintaining the
colloidal or very fine solid materials, or large entire fermentor under vacuum, the
dissolved molecules, by filtration through fermenting beer is circulated through a
microporous or semi-permeable membranes. vacuum chamber to flash off the ethanol
The process may be used for removal of before returning the beer to the fermentor.
protein from cheese whey prior to
fermentation. Vapor A dispersion in air of molecules of a
substance that is liquid or solid in its normal
United States Department of Agriculture state at standard temperature and pressure. An
(USDA) A federal government department example is water vapor or steam.
438 The Alcohol Alphabet
Vaporization (or Volatilization) The Wet milling A process in which corn is first
conversion of a chemical substance from a steeped (or soaked) in water containing sulfur
liquid or solid state to a vapor or gaseous state dioxide. This softens the kernels and loosens
by the application of heat, by the reduction the hulls. (The liquid when separated is
of pressure or by a combination of these known as corn steep liquor). The grain is
processes. then degermed by abrasion and liquid
separation. Oil is extracted from the germ,
Vapor pressure The saturation pressure exerted while the remainder of the kernel is ground
by vapors when in equilibrium with their in impact fiber mills and passed through
liquid or solid forms. stationary screens to separate the starch and
Vent condenser The final condenser in a series gluten from the fibrous portion. The heavier
of two or more connected to the overhead starch is then separated from the gluten by
vapor line of a distillation column. As the final use of centrifuges. The starch portion may
condenser, it is the one from which non- then be processed into commercial starch, or
condensed (or non-condensable) gases and it may be used as a feedstock for production
vapors are vented to the atmosphere or to a of ethanol or HFCS.
scrubber system.
Vinasse The term sometimes applied to the Weeping The condition when droplets of liquid
stillage of molasses, grape juice or other liquid fall through the holes of a sieve plate in a
ethanol feedstocks. distillation column. It may be caused by a)
Vodka Currently defined by the BATF as steam flow that is too low, or b) too low a
‘neutral spirit so distilled, or so treated after liquid flow to maintain a level on the plate, or
distillation with charcoal or other materials c) having a tilted plate such that liquid depth
as to be without distinctive character, aroma, is uneven.
taste or color’. It should be noted that the Whey The serum or watery part of milk
charcoal treatment is now optional, whereas separated from the curd in the process of
in earlier BATF regulations it was mandatory making cheese. It contains lactose and may
and the minimum time of treatment and be used as a feedstock for ethanol production.
amounts of fresh charcoal to be used were
specified. Whisky Defined by the BATF as ‘an alcoholic
distillate from a mash of grain, produced at
Volatile Adjective to describe a solid or liquid less than 190° proof, in such as manner that
that readily converts to the vapor state. the distillate possesses the taste, aroma and
Volatility The tendency of a solid or liquid to characteristics generally attributed to whisky’.
pass into the vapor state at a given With the exception of corn whisky, it should
temperature. With automotive fuels, the be stored in oak barrels.
volatility is determined by measuring the Reid Whole stillage The entire stillage emerging from
vapor pressure (Rvp). a distillation unit before any removal of solids
Volatilization See Vaporization. by screening or centrifuging.
Wine gallon A US gallon of liquid measure as
distinct from a proof gallon.
W
Wood alcohol See methanol.
Wash A British synonym for distillers beer.
Wort A brewery term used occasionally in
The Alcohol Alphabet 439
Index
JACQUES
KELSALL
EDITION
LYONS
TH
TH
2003
A
A reference
reference for
for the
the beverage,
beverage, fuel
fuel and
and industrial
industrial alcohol
alcohol industries
industries