Professional Documents
Culture Documents
Beer Haze 1999
Beer Haze 1999
Beer Haze
Charles W. Bamforth
To cite this article: Charles W. Bamforth (1999) Beer Haze, Journal of the American Society of
Brewing Chemists, 57:3, 81-90
Article views: 2
ABSTRACT These authors are not the first to describe bits in beer (103), and
it is certainly known that unsightly particles may arise through the
J. Am. Soc. Brew. Chem. 57(3):81-90, 1999 interaction of different stabilizing agents in beer, such as a cross-
linking of papain with propylene glycol alginate (PGA) occurring
Several substances can cause haze in beer, but the most frequently en- during pasteurization (52). Indeed, PGA can cross-link with other
countered problem is due to a cross-linking of polyphenol and protein.
These materials probably exist in equilibrium in beer and manifest them- proteins in beer, including residual isinglass finings, to throw ge-
selves as a haze when the polyphenol polymerizes. The proteins particu- latinous precipitates when beer encounters high temperatures.
larly involved in haze are rich in proline. A range of stabilization treat- These precipitates break up into a snowstorm of bits when the
ments—of relatively low cost— is available, but the avoidance of haze beer is poured.
problems (which seem to be occurring with increased frequency) de- At the other end of the particle size range are the so-called
mands a holistic approach from the brewer in which all stages from raw “invisible” hazes, which are more accurately (if rather less fre-
material selection to final processing are geared to achieving prolonged quently) called “pseudo-hazes.” These are caused by very small
shelf life. particles (<0.1 µm) which cause high levels of light scatter when
Keywords: Haze model, Polyphenol, Polypeptide, Prediction, Stabi-
measured at 90° to incident. At least one cause was found to be
lizer
unmodified regions of the starchy endosperm of barley (42), per-
haps manifesting themselves in the form of retrograded α-glucan
RESUMEN
(103).
Existen diversas substancias capaces de enturbiar la cerveza, pero el From time to time, visible hazes in beer have been reported as
problema que se encuentra más frecuentemente se debe a un entre- being due to residual starch (50), pentosans from wheat-derived
cruzamiento de polifenoles y proteínas. Probablemente estos consti- adjunct (17), oxalate from calcium-deficient worts (34), β-glucan
tuyentes se encuentran en equilibrio en la cerveza y se manifiestan en from inadequately modified malt (30), carbohydrate and protein
forma de turbiedad cuando los polifenoles polimerizan. Las proteínas from damaged yeast (51), lubricants from can lids (88), and dead
específicamente implicadas en la formación del turbio son ricas en bacteria from malt (4,99). For the purposes of this article, the
prolina. Se dispone de una gama de tratamientos para la estabilización de author is not considering the risk of biological hazes arising from
la cerveza, todos ellos de un costo relativamente pequeño, pero el
soslayar los problemas de turbiedad (que parecen presentarse cada vez de
living microbes.
forma más frecuente) exige que el cervecero aborde el problema Articles by Glenister (31,32) on how to identify hazes have long
globalmente, de forma que todas las etapas, desde la selección de since become bibles for haze classifiers. Buckee (7) has also pub-
materias primas hasta las fases finales del proceso, estén encaminadas a lished a description of haze identification procedures, including a
conseguir la mayor vida útil del producto en el mercado. comprehensive listing of the materials that may cause hazes in
beer. Being able to establish the composition of a haze
The great unsolved mystery in the world of brewing is, surely, (remembering that the underlying cause of a haze may be at the
the cause of (and solution to) flavor instability. Most other aspects nucleus of the particle rather than in the material which resides on
of beer quality, including foam stability, color, and initial beer the surface) is key to rectifying the problem.
flavor are, at least for the most part, under control. For many However, by far and away the largest clarity risk in beer is the
brewers, haze problems, too, are a thing of the past. Remarkably, haze produced from protein and polyphenol. The vast majority of
though, some brewers have encountered difficulties in recent publications on haze center on this problem and it is likely that a
years, despite the proven worth of several palliative treatments great many of the recent haze difficulties have been caused by it,
allied to good brewing practice. despite this extensive study. This article will focus primarily on
There are at least three possible explanations for this situation. this type of haze.
The first is that some companies are forgetting the basics. The
second is that they are trying to take short cuts. The third is that Protein-Polyphenol Hazes
colloidal instability problems of a different type are being en- When a low molecular weight polyphenol cross-links with a
countered. protein through weak interactions such as hydrogen bonds, a “chill
haze” (a haze which forms when beer is chilled to 0°C, but which
Types of Haze returns to solution when the beer is returned to 20°C) is produced.
In connection with the third explanation, one very real change Particle sizes range from 0.1 to 1.0 µm. Polyphenols, however, are
in recent years has been the distances over which beers are trans- prone to polymerization and, when they interact with protein after
ported. Walters et al (102) described the formation of “bits” in this process, they form a “permanent haze” (a haze which is present
beers which had been subjected to a degree of agitation mimick- in beer even at 20°C), with particles between 1 and 10 µm. Both are
ing that which a beer will encounter when it is shipped and han- obviously a concern; however, in reality, it is chill haze which the
dled. These bits, which contain protein and perhaps pentosans, are brewer must counter, because it is this which will arise first and
thought to arise as the skins around foam generated within the which will go on to develop into permanent haze.
package. Polymerization of polyphenols is promoted by oxidation. In
part, this may be catalyzed by enzymes such as polyphenol oxi-
1 Corresponding author. Tel: 530 752-1467; Fax: 530 752-4759;
dase, sometimes known as laccase, (44) and peroxidase (15), but
E-mail: cwbamforthgucdavis.edu polyphenol can react extensively with oxygen during wort boiling,
presumably during the heat-up phase as oxygen concentration at
Publication no. J-1999-0608-01R. boiling point will be sensibly zero (76), during which no enzyme
© 1999 American Society of Brewing Chemists, Inc. activity can occur. Furthermore, malt enzymes do not survive into
81
82 / Bamforth, C. W.
finished beer, in which the majority of the take-up of any oxygen the significance of hydrophobic amino acid residues; he believes
is into polyphenols (76). Kaneda et al (45) remind us that ground- that haze particles grow through interactions between these resi-
state oxygen is comparatively unreactive and they invoke reduced dues. Hydrophobic character is also key to the foaming properties
radical forms of oxygen, notably hydroxyl, as being responsible of beer polypeptides (94), which might argue for a crossover be-
for haze formation owing to the efficiency with which it oxidizes tween haze- and foam-forming materials. In addition, isinglass
polyphenols in beer. finings (gelatin) contain high proportions of proline residues;
Delcour et al (20) observe that most beers today are packaged normally, finings are considered to be protein precipitants, but it is
with very low levels of oxygen, yet nonetheless can throw a haze. very likely that they also adsorb polyphenols.
They suggest that aldehydes, including acetaldehyde (in their Recently, Ishibashi (40) raised antibodies to both foaming
model studies, they used atypically high concentrations ranging polypeptides and haze. Whereas the antibody raised against
from 25 to 400 mg/L), can react with polyphenols to form species foaming polypeptide was specific to that type of material, the an-
which cross-link with proteins. tibody raised against haze cross-reacted both with haze and
Clearly, any strategy for dealing with this type of haze must ad- foaming polypeptide, suggesting that a broad span of polypeptides
dress the elimination of haze-active polyphenol or haze-active can enter into haze. Ishibashi used the antibodies to show that
protein during the brewing process, or a proportion of both, be- both foam and haze polypeptides were lost throughout the brew-
cause there is a school of thought that maintains that classes of the ing process but preferentially during boiling, fermentation, and
polyphenols are beneficial for mouthfeel (48), as antioxidants filtration. This would not augur well for the prospects of being
countering staling (100), or for their nutritional value (33). Fur- able to preferentially eliminate haze proteins. Furthermore, he
thermore, as long as there is doubt about whether the haze-form- claimed a good correlation between the shelf life of a beer and the
ing and foam-forming proteins are different, there will be concern level of haze protein measured by an enzyme-linked immunosorb-
about the removal of too much protein. ent assay test incorporating the anti-haze antibody.
In summary, at this time there must be some doubt about there
Haze-Active Proteins being a clear differentiation between haze- and foam-potentiating
As little as 2 mg/L of protein is sufficient to induce a haze of 1 polypeptides in beer. Indeed, it has been claimed that polyphenols
EBC unit (12). Haze intensity is defined by an EBC method (1) can cross-react with foaming polypeptide to improve head reten-
which involves the measurement of light scattering at an angle of tion (18), and this would support cross-functionality in the
90° to the incident beam, calibrated with a formazin standard. On polypeptide species.
the EBC scale, a value of <0.5 is “brilliant,” 0.5–1 is “almost When considering haze, however, it should be reiterated that
brilliant,” 1–2 is “very slightly hazy,” 2–4 is “slightly hazy,” and materials may accumulate adventitiously on the surfaces of parti-
≥4 is “hazy.” In view of the fact that most beers contain in excess cles, without in themselves being causative agents of haze forma-
of 2 g/L of protein, it is apparent that there is vastly more protein tion.
in beer than is needed to form a haze.
Asano and colleagues (2,3) have made the most detailed study Haze-Active Polyphenols
of haze-active proteins. They isolated a series of polypeptide frac- There are few more intimidating topics in brewing science than
tions from beer with haze-forming tendencies and concluded that that of polyphenols; the complexity is immense. Even leaving to
they were for the most part derived from hordeins and were rela- one side the myriad of formulae, we are confronted with a con-
tively rich in proline. However, there was not an absolute distinc- fusing plethora of terminology. Phenolics, as the name indicates,
tion. Albumin- and globulin-derived polypeptides also came out are based on a backbone of phenol (monohydroxylated benzene).
of solution as chill haze, but only after the hordein-derived spe- Phenolic acid includes a carboxyl group (e.g., ferulic acid).
cies. Indeed, foam-forming polypeptides progressively came out “Polyphenol” includes all molecules containing two or more phe-
of solution as haze, with a commensurate decay in foaming ability nol rings. Thus, the following are all examples of polyphenols:
of beer with time. This observation flies in the face of theories flavanol, a monomeric species with structures of the type dis-
that predict a clear separation between haze-forming and foaming played by catechin; flavonol, monomeric species with structures
polypeptides. Adherents to the dogma that haze-active proteins of the type displayed by quercetin, but usually present in hops as
originate in hordein and foam-active polypeptides in albumins glycosides (i.e., joined to sugar); flavanoid, oligomers of flavanols
should also note recent observations that antibodies raised to hor- (e.g., prodelphinidin B3 and procyanidin B3); proanthocyanidin,
deins will cross-react with foaming polypeptides (68). molecules cleavable by acid to form substances which polymerize
McMurrough et al (58) isolated four high molecular weight in the presence of oxygen to pigments called anthocyanidin
polypeptide fractions on DEAE-cellulose and found that they dif- (examples include the flavanoids mentioned above); anthocyano-
fered greatly in their response to different polyphenols. Generally gen, an outmoded terminology for proanthocyanidin; tannoid,
speaking, the polyphenols derived from hops caused more pre- polymers of flavanoids which are intermediates on the “growth
cipitation than did those from malt. route” to tannins; and tannin, polymers of flavanoids of a size
Matsuzawa (67) claims that polypeptides with an isoelectric sufficient to precipitate proteins. An illustrative set of formulae
point between 3 and 5 are particularly haze-active, and Mussche, for the various classes is shown in Figure 1.
too, claims that haze-potentiating proteins can be clearly differen- Bright beer is usually devoid of tannins, but does contain tan-
tiated from foaming proteins on the basis of isoelectric point, with noids (12). These species, of relatively low molecular weight, co-
the acid proteins being haze formers and the foaming polypeptides exist in beer alongside protein, and Chapon claims that there is a
being basic (70). He furthermore claims that the higher molecular readily reversible equilibrium between the polyphenols and the
weight proteins are more prone to throwing hazes. proteins in their free and soluble but bound forms. He observes
Outtrup (83,84) cites the high level of proline residues in haze- that proanthocyanidins are the most active low molecular weight
forming protein, accepting that it is not simply the amount of polyphenols.
proline in a protein that matters, but also its distribution. He notes The ability of catechin to form a haze is enhanced by its oxida-
the similarity of the polypeptide polyproline to the stabilizer poly- tion (82). McMurrough and O’Rourke (65) demonstrated that the
vinylpolypyrrolidone (PVPP), which makes it unsurprising that level of flavanoids declines during beer storage, through conver-
polyphenols recognize both PVPP and proteins containing rela- sion to tannoids. McMurrough et al (63) showed that monomeric
tively high levels of exposed proline. Furthermore, Outtrup notes polyphenols potentiate haze once they have been oxidatively poly-
Beer Haze / 83
merized. This occurs readily if beer is “punished” by heat (e.g., in of polyphenol extraction during wort separation is a reflection of
a forced-aging regime), even in the absence of oxygen. the varying extents to which oxidation has taken place in mashing
McMurrough concludes that 75% of the haze in untreated beer is and lautering.
formed from already complexed polyphenols, which surely should The polyphenols precipitated during wort boiling are believed
focus more concern on what happens during the brewing process to find their way into hot break by way of adventitious association
than on whether or not excessive levels of monomeric polyphe- with coagulated protein rather than by taking part in the insolubi-
nols survive into beer. lization process (56). In large part, this conclusion hangs on the
McManus (55) claims that, for a polyphenol to participate in the observation that hot break production is unaffected by the mode of
reactions leading to haze formation, at least two hydroxyl groups “hopping” (e.g., tannin-free kettle extracts give the same hot break
on an aromatic ring are required. Flavanoids are readily oxidized formation as “standard” extracts) (23). By contrast, polyphenols
through dihydroxy and trihydroxy aromatic rings to form tanning do have a role to play (probably via their oxidation) in the forma-
complexes (21,29).
Polyphenols in beer arise from both malt and hops. Some
authors have claimed that those from hops have an insignificant
role in protein precipitation (95). Chapon (12) says that this is true
in the context of the quantitative removal of total nitrogen from
beer, but he maintains that the contribution of hop polyphenols is
substantial when judged in protein-sensitivity terms. Delcour et al
(22) claim that malt and hop polyphenols are identical in their
haze potentiating capability. Srogl et al (95) state that hop
polyphenols have little influence during wort boiling as regards
the precipitation of proteins, but rather remain into beer, where
presumably they potentiate haze formation. They observe that
aroma hops deliver more polyphenols pro rata than do bitter hops.
Between 2 and 4% of the dry weight of hops is accounted for by
polyphenols (98). Srogl et al (95) claim that barley polyphenols
are much less extractable into buffer of pH 5.5 than are the hop
polyphenols. As for the barley-derived polyphenols, Bellmer et al
(5) confirm that winter varieties have higher levels than do spring
cultivars.
Hops contain flavonols (57), usually in the form of glycosides,
and they survive into beer (59). McMurrough and Delcour say that
there is no convincing evidence that they have any effect on beer
quality (56). Rather, they observe that it is the monomeric and
oligomeric flavanols that impact quality. Among these are cate-
chin and epicatechin, which exist as monomers and in polymer-
ized forms known as proanthocyanidins. In acidic conditions in
the presence of oxygen, these materials yield anthocyanidins, with
their strong red coloration. These tannin polymers also cross-react
with proteins to form haze.
McMurrough and Delcour (56) briefly summarize the literature
concerning the levels and types of the monomeric and polymeric
flavanols in barley and malt, and conclude that catechin, two
dimers, and four trimers are present. Monomers and dimers, at
least, survive into beer.
Malting probably has an insignificant effect on the levels of the
flavanols. Catechin is more readily extracted into mashes than are
the dimers (56). McMurrough claims that extraction of mono-
meric, dimeric, and trimeric flavanols is greater from single tem-
perature infusion mashes than from programmed mashes (56).
During mashing, too, heat can promote the co-precipitation of
proteins and polyphenols and, at least as importantly, any oxygen
present in the mash can be involved directly (13,15) or indirectly
(69) in the oxidation of polyphenols, primarily through the inter-
mediacy of peroxidases from malt. The effect is an increase in the
polymerization index of the polyphenols, an attendant develop-
ment of red color, and an increased precipitation (16) due to the
conversion of flavanol monomers and dimers to tannins.
McMurrough and Delcour (56) have amply described how the
extraction of monomeric and polymeric flavanols during wort
separation is complex and variable. It appears that total polyphe-
nol extraction mirrors that of other wort constituents (43).
McMurrough et al (60) observed that the highest concentrations of
all flavanols are to be found in the first runnings from a mash.
Conversely, it has been found that tannins are eluted maximally at Fig. 1. A series of phenolics of increasing complexity found in wort and
the end of runoff (47). Perhaps the variation observed in the extent beer.
84 / Bamforth, C. W.
tion of cold break (18). McMurrough et al found that 70% of malt especially manifest themselves as visible particles when there are
polyphenols survive boiling and remain in solution, whereas only roughly equal quantities of haze-active protein and haze-active
20% of hop-derived polyphenols avoid precipitation as break (58). polyphenol. Naturally, the levels of both will need to be suffi-
McMurrough’s lab has been instrumental in identifying the ciently high to manifest a visible haze when they associate. Light
types of flavanol found in wort and beer (60). In sweet wort, there scattering is less intense when the levels of protein and polyphe-
is a preponderance of dimers, and some trimers, from malt. After nol are disproportionate.
boiling, there is proportionately more catechin and epicatechin Such a model has major implications for the stabilization of
(which is extracted from hops and by transformation of catechin) beer. For instance, “single ended” polyphenols would be expected
and also of complexed flavanols. Very little change to this distri- to block haze formation by competing for proline residues in pro-
bution occurs during fermentation, though the more polymerized teins and preventing cross-linking. Indeed, an excess of either
materials do seem to be lost on cold conditioning (56). The com- haze-active protein or haze-active polyphenol would be expected
plexed flavanols formed the bulk of the polyphenols in beer, but to counter haze development, suggesting that a beer should be
did not automatically correlate with haze formation. McMurrough “over-dosed” on one or other, but certainly not both. Also relevant
concludes that it is the low levels of such materials as dimeric is Chapon’s model (12), which states that the following equilibria
flavanols that are the problem regarding haze (56). exist in beer: P + T ←→ P – T (soluble) → P–T (insoluble), where
P = haze-active protein and T = tannoid.
Model for Protein-Polyphenol Interactions This model implies that protein-tannoid complexes exist in beer
Siebert’s team (92,93) has described a model for protein- in equilibrium with the monomeric species and that this equilib-
polyphenol interactions (Fig. 2). The model assumes that it is only rium can be shifted towards the monomers by removing either
proline-containing proteins that interact to form chill haze, with a protein or tannoid.
fixed number of polyphenol binding sites in toto. Furthermore, it
is assumed that a polyphenol has two (or more) “ends” which can Stabilization Treatments
specifically interact with these binding sites on proteins, thereby It is clear that both raw materials and process have a major
allowing a single polyphenol molecule to bridge between protein bearing on the level of haze precursors entering into beer, sug-
molecules. If there is an excess of haze-active protein over haze- gesting that good practice can be recommended in respect to col-
active polyphenol—as Siebert claims for beer (89)—then most loidal stabilization of beer. However, when considering how to
polyphenols are involved in bridging two proteins together, with lengthen the shelf life of a product, most people automatically
insufficient polyphenol to bridge dimers and form larger particles. gravitate to considerations of stabilizer treatments. Stabilizers can
If the haze-active polyphenol is in excess of protein (as, for in- undoubtedly have a major role to play, but should only be applied
stance, occurs in ciders), then there will be a shortage of free as part of a more holistic approach to the achievement of colloidal
proline sites able to enter into cross-linking of protein molecules. robustness. Furthermore, some caution needs to be applied when
The conditions exist for the formation of large networks that will surveying the literature discussing these agents, because much of
it has been generated by suppliers of one or the other of the stabi-
lizing treatments, who are obviously going to stress the merits of
their own type of product.
Essentially, there are three potential strategies: remove protein,
remove polyphenol, or remove a proportion of each. Many would
argue, in the context of uncertainties concerning the value of some
polyphenol and some protein in benefiting other aspects of prod-
uct quality, that the third of these strategies is the most valuable.
Whichever is applied, it must be borne in mind that the binding
forces involved in holding polymeric materials to adsorbents are
relatively weak; therefore, low temperatures are a pre-requisite for
colloidal stabilization.
One other key factor, which is sometimes neglected in descrip-
tions of beer stabilization, is beer clarity. Quite apart from the
influence which clarity may have on the efficiency of processes
involving PVPP and silica hydrogels, for instance, less-than-bright
beer will have a greater tendency to develop a haze through parti-
cle growth. If there are many small particles in beer (perhaps
those which are responsible for invisible or pseudo-hazes), they
can form nucleation sites for haze development. The first priority
must be to ensure that brewery operations are configured to mini-
mize the load on the filter (e.g., possible use of kettle, isinglass,
and auxiliary finings; efficient particle settling regimes in cold
conditioning; and so on) and that the filtration regime is properly
set up in respect of filter aid selection, dosing rate, lowest possible
temperature, minimum oxygen pick-up, and other procedures, not
forgetting that filter aid will itself adsorb haze materials. Only
then is it appropriate to establish the appropriate stabilization
treatment for the product in question.
Polyinylpolypyrrolidone
Polyinylpolypyrrolidone (PVPP) adsorbs both flavanoids and
tannins (82). Some people consider it an inappropriate treatment
Fig. 2. Siebert’s model for protein-polyphenol interaction in beer. because of its removal of potential antioxidants such as catechin.
Beer Haze / 85
However, as Walters et al have demonstrated (101), catechin and, An additional claim is that PVPP is beneficial to foam stability,
by inference, other flavanols surviving into beer have a value only perhaps by taking out polyphenol that would otherwise remove
as protectants against new oxidation in beer and not the inherent foaming polypeptide (78).
staling potential built in during the process. Therefore, the litera-
ture from suppliers of PVPP claims that beers treated with this Silica Hydrogels and Xerogels
agent are of identical flavor stability to other beers. It has even The manufacturing process for silica hydrogels and xerogels
been claimed (78)—but disputed by McMurrough (64)—that has been described by Fernyhough et al (27). Xerogels are pro-
PVPP preferentially adsorbs highly hydroxylated flavanoids that duced from hydrogels by drying before the milling stage that is
promote staling by stimulating the production of oxygen free radi- used to derive the preferred particle sizes. The critical features
cals (i.e., PVPP is beneficial to all types of shelf life). Further- include the pore size of the particles and the surface area pre-
more, O’Rourke (82) and others have inverted the arguments that sented by them. McKeown and Nock (54) say that the most effec-
polyphenols are needed for beer body with their claims that PVPP tive pore size for a hydrogel is 3–12 nm and that to use anything
improves taste quality by removing harsh and astringent con- larger is to risk head retention. A reduction in particle size (giving
tributors. This whole issue of the contribution of polyphenols to an increase in surface area) increases the adsorption rate. This is
taste and mouth feel needs urgent rationalization. At this point in of particular significance for the mode of use of hydrogels. If they
time, the conclusion seems to be that the levels of polyphenols are dosed into the storage tank, then time will allow equilibrium to
present in most beers are too low to make a material contribution be established. Conversely, if they are to be dosed in-line, then
(6). adsorption rate is an especially important parameter. The latter
Two types of PVPP are presently in use (78). The first, single- approach has been increasingly used, with 50 g/hL of an appropri-
use PVPP, comprises a micronized white powder of high sur- ate grade having been successfully substituted for kieselguhr body
face/weight ratio. This type of material readily adsorbs polyphe- feed, allowing savings on the latter and longer filter runs (27).
nols on its surface and can therefore be incorporated into a filter Silica hydrogels with low permeability (filterability) afford
aid bodyfeed dosing regime. The second type, regenerable PVPP, better stability (54). To achieve beers of prolonged shelf life, the
can either be impregnated into sheets or used within devoted hori- brewer should employ either low-permeability gels in storage
zontal leaf pressure vessels. These are regenerated by 1–2% caus- tanks, with a commensurate decrease in throughput and increased
tic with subsequent rinsing with hot and cold water and neu- filter aid usage, or else use larger quantities of high-permeability
tralization by CO2 or 0.3% nitric acid. These materials are only gels. Newer generations of silica hydrogels have been developed
used after powder-based filtration (35), and a trap filter of cotton with very high stabilization efficiency (36) and it is claimed that
or cellulose candles may be necessary after a powder-based sys- such preparations allow shelf lives as long as two years (27). They
tem in order to retain any very fine PVPP. have reduced permeability and should be added to the storage
Chapon (12) reminds us that treatment of beer with PVPP must tank. Guzman and Nock (37) describe the use of combined filtra-
be homogeneous; otherwise, there may be substantial differences tion and stabilization with silica hydrogel, describing savings in
between the level of polyphenols in different packages. raw materials, waste disposal costs, total process costs, filter life,
The rate of usage of PVPP differs between beers (e.g., in par- and reduced risk of iron pick-up. Fernyhough et al (27) say that
ticular, depending on the level of polyphenols emerging from raw the extent of silica hydrogel usage depends on the intrinsic stabil-
materials) and the level required needs to be identified empirically ity of the beer being treated, the shelf life required, and the extent
for each beer (78). O’Reilly claims that reduction of a typical to which other stabilizers are being used. They report the devel-
flavanoid level of 30–40 mg/hL in a 20% adjunct lager by 25– opment of models for individual beers. For one product, hydrogel
30% will allow a 12-month shelf life. To achieve this would de- at 75 g/hL increased shelf life by 3–6 months, whereas doubling
mand a dose rate of 15–20 g/hL for single-use PVPP and 35– this rate also doubled colloidal lifetime. O’Rourke cites an
40g/hL for the regenerable grades, which have lower surface area. equivalence of 1 g/hL PVPP to 7 g/hL silica hydrogel (79).
These low doses of PVPP preferentially remove flavanoid dimers The earth bentonite (which is now rarely, if ever, used in the
and greater, with doses as high as 100g/L needed to remove large brewing industry) is non-specific in its protein binding, removing
quantities of the monomers such as catechin (66), which should both haze and foaming polypeptide; in contrast, silica hydrogel
reassure those who worry about the removal of a strong antioxi- removes haze-forming protein preferentially (91). This is because
dant from beer. silica hydrogel recognizes and interacts with the same sites on
McMurrough (61) cites a binding capacity of PVPP for haze-active polypeptides, as do the polyphenols. A question still
polyphenols of ≈90 mg of polyphenol per 1 g of PVPP. O’Reilly not wholly answered is whether a competition can exist in beer, in
(78) acknowledges the role of haze-sensitive protein and points which polyphenols and hydrogels vie for the polypeptides. If this
out that the majority of single-pass PVPP is used alongside silica is the state of affairs, then it may be that high levels of polyphe-
hydrogel or xerogel, whereas the regenerable material is fre- nols interfere with stabilization efficiency by silicas (as is cer-
quently used alone. tainly the case with ciders, possessed of far higher polyphenol
McMurrough et al (61) demonstrated a convincing relationship levels than beers) and that a co-treatment of beer with PVPP and
between haze development and the product (sensitive proteins × silica hydrogel would be best. Narziss mentions a pairing of 20–
proanthocyanidins), with the proteins measured by tannic acid 50 g/hL of PVPP with a similar quantity of silica hydrogel incor-
precipitation and the latter as the sum of the dimeric flavanoids porated into the body feed (73).
prodelphinidin B3 and procyanidin B3 measured by HPLC. A
much poorer correlation was demonstrated when the proanthocya- Tannic Acid
nidin measurement was replaced by total or oxidizable polyphe- Tannic acid possesses a molecular structure closely similar to
nols. Therefore, models were developed which calculate how that of other polyphenolic species; therefore, it is hardly surprising
much a given beer needs to be treated with PVPP, silica hydrogel, that tannic acid is claimed to be a relatively specific precipitant of
or both to achieve a given shelf life. McMurrough is adamant that haze-active proteins in beer (72). Just as the co-use of PVPP and
the dimeric species should be of particular concern to the brewer silica hydrogel has been advocated, so too did Mussche (72) ex-
and should be measured with a view to eradication. He accepts plain the benefits of joint use of tannic acid and PVPP, in order
that they need to be oxidized into a polymerized form before they that both haze-active proteins and haze-active polyphenols should
acquire a tanning capability. be removed.
86 / Bamforth, C. W.
Tannic acid is a precipitant, as opposed to an adsorbent like which they make process additions. If papain (as haze stabilizer)
PVPP or silica hydrogel; therefore, it throws a sizeable precipitate is added as a solution with ascorbic acid (antioxidant) and copper
when added to a cold conditioning tank. This necessitates either sulfate (sulfur scavenger), then at least one of those agents won't
cautious transfer of beer from sedimented material in the tank or be functional.
the use of a polisher centrifuge followed by membrane filtration
(71). Alternatively, because the reaction time of the new genera- Future Possibility
tion of gallotannins is so rapid, they can be dosed on-line to the Katzke et al (46) described the development of regenerable ion
powder filter. exchangers for removing haze-forming proteins from beer, materi-
Tannic acid is normally added from a 1–5% solution made up in als which are at the trial stage.
deaerated water at room temperature in the brewery. Sedimenta-
tion time following dosing on transfer from fermenter to storage Relative Cost of Stabilizing Treatments
tank is at least one day, depending on temperature, yeast count, A recent comparison of stabilizer costs (80) is given in Table I.
vessel geometry, and related factors. When tank bottoms are in-
creased, the claim is for 25% longer filter runs. Predictive Methods
If the gallotannin is dosed as beer flows to filter, then 5–10 min There is great interest in the availability of reliable methods that
of contact time at 0 to –1°C is necessary and the filter aid must be will enable the brewer to predict how long a beer will last in the
adjusted to a much coarser grade of kieselguhr or a blend with a trade before it throws a haze. The very fact that a number of
high (90%+) proportion of perlite. methods exist is a sure sign that none of them is perfect. It is es-
Mussche claims that tannic acid is not detrimental to foam sential for any of these tests that the brewer convinces himself or
when used at recommended dose rates, but that it can remove herself that there is an unalterable relationship between actual
foaming polypeptides when used at levels as high as 200 mg/L performance of a beer in trade and the results of one or a combi-
(72). Ishibashi, however, using antibodies, detected a loss of nation of tests.
foaming polypeptides caused by tannic acid (40). O’Neill (77) defines colloidal shelf life as the length of time be-
fore beer displays a haze value of 2.5 EBC at 0°C. For many
Papain brewers, this (or a similar figure) would be the yardstick against
Papain (26), from Carica papaya (paw paw), was the first haze- which they would judge breakdown in a predictive test. There are
preventative employed in the brewing industry. It retains some two main types of predictive tests: hot-cold cycling tests and pre-
usage, particularly by brewers taking a “belt and braces” approach cipitation tests.
for beers destined for particularly challenging conditions. How-
ever, few brewers would deny the disadvantage of proteolytic Hot-Cold Cycling
enzymes as stabilizing agents, because of their tendency to lessen There seem to be almost as many variants of this test as there
foam quality. Papain has a preference for peptide bonds formed are brewers. Some tests don’t even consider the cold stage. For
from hydrophobic amino acids; therefore, it is perhaps no surprise instance, in the Guinness test (61), beer is held at 37°C; years of
that it hydrolyses foaming polypeptides, which are those with a experience with their products convince them that one week at
high hydrophobic character. Should foam quality be a secondary this temperature is the equivalent of one month of “normal” stor-
consideration in a given marketplace, then papain is a perfectly age (18°C). O’Neill (77) cites a few examples, illustrating the
acceptable stabilizer, one which has the particular attraction of not length of time needed for a result on a beer by focussing on a
requiring any special equipment to enable its use. variant calling for 60°C for 2 days followed by –2°C for one day,
Papain is added on transfer to maturation or during maturation in which one complete cycle is said to be the equivalent of six
itself. It progressively loses its activity during storage and especially weeks of normal storage. A four-week test is needed to fully pre-
during pasteurization. Notwithstanding, it will continue to act for a dict whether a beer will last a full year in the trade. Few brewers
limited time in a tunnel pasteurizer, and for a short period will effect have the luxury of being able to hold packaged beer for long
proteolysis approximately 100-fold more rapidly than during cold enough to allow such a test and, once the beer has left for the
conditioning. In sterile-filtered beers, any papain would survive into marketplace, there is very little value in going through with such a
the package to progressively lower foam stability. test, other than as a warning that there may be some beer to re-
Some argue that the mode of action of papain has as much to do cover from trade in a few months’ time.
with a clotting effect as one of hydrolysis. Militating against this Such tests are useful in the study of trial brews; for example,
is the observation by Fukal and Kas (28) that the action of papain the evaluation of new palliative procedures.
is blocked by a combination of ascorbic acid, copper, and oxygen.
Such a mixture generates hydroxyl radicals which inactivate the Precipitation Tests
enzymatic activity. This observation has practical implications in The most famous such procedure—one which has almost be-
that some brewers are less than circumspect about the manner by come an industry norm—is the alcohol-chilling test of Chapon
TABLE I
Comparison of Stabilizer Costs
Point of Addition
Process Aid To Kettle To Maturation To Filter Cost Treated ($/hL)
PVPP single usea 20 g/hL* b 12–20 g/hL 15–20 g/hL 0.40 at 20 g/hL
Tannic acid 5 g/hL 5–7 g/hL 2–5 g/hL 0.32 at 6 g/hL
Silica hydrogel 30 g/hL* 50–100 g/hL 30–80 g/hL 0.16 at 80 g/hL
Proteolytic enzyme nil 2–5 mls/hL 2–5 mls/hL 0.14 at 5 mls/hL
PVPP regeneration c 15 g/hL* nil 20–30 g/hL 0.05 at 1% loss
a PVPP = polyvinylpolypyrrolidone.
b * = not usual point of application but has been or is being tried.
c The regeneration grade price is based on material costs with 1% loss and takes no account of capital or running cost of the filter.
Beer Haze / 87
(11). Alcohol is added in the lowering of the temperature of a beer importance of homogeneous modification to the elimination of
to –8°C and chill haze is forced out within a total test time of 40 troublesome β-glucans and, indeed, to the subsequent efficient
min. McKeown and Nock (54) observe that the method is depend- release and degradation of starch in mashing is significant. Re-
ent on the starting alcohol content of a beer (i.e., it is “beer de- placement of barley malt with syrup adjuncts, or with rice and
pendent”), but they report a good correlation between the values maize grits or flakes, will dilute all types of haze precursor.
obtained and the product of sensitive protein and tannoid which, Barley- and wheat-based adjuncts, however, will increase risks
as we have seen earlier, McMurrough has shown to relate closely from haze-forming proteins, polyphenols, glucans, and, in the case
to shelf life. Chapon himself, however, accepts that the test pre- of wheat at least, pentosans.
dicts only chill haze. Although this is the main problem in the Barley varieties vary considerably in their content of polyphe-
lifetime of most beers, it is not the only potential colloidal prob- nols; for example, winter varieties having relatively high levels
lem in beer. (49). This is not an argument against the use of such varieties,
A further test—one of those incorporated into the Tannometer because it has been shown that the behavior of specific flavanoids
(a commercial instrument which combines several predictive tests is important, rather than the absolute level of polyphenol. The
for haze life) (8,86)—involves the precipitation of haze-active majority of the polyphenols are located in the husk. Alkaline
proteins by tannic acid, the “protein-sensitivity” of a beer. In a steeping of barley removes polyphenol from the husk (19). How-
comparable test, the saturated ammonium sulfate precipitation ever, in order to lower the total polyphenol contribution from any
limit (SASPL) test, gallotannin is replaced by a solution of satu- variety, it is possible (if perhaps tedious) to remove the husk from
rated ammonium sulfate. In the former, the amount of light scatter malt prior to brewing (41). The first low-anthocyanogen barley
caused by a standard addition of tannic acid is measured. In the was bred several years ago (25) and varieties with improved agro-
latter, the number of milliliters of (NH4)2SO4 which need to be nomic qualities have been developed since then (85). These varie-
added to cause a measurable increase in turbidity is recorded, on ties are not yet in widespread use, apparently because of short-
the basis that the more salt is needed to bring out protein, the less comings other than in the brewery; however, their efficacy in
such precipitable protein is in solution. This method is useful for terms of providing colloidally stable beer is unquestioned. The
comparing samples of a single brand of beer, but not for compar- brewer should never lose sight of the fact that hops also provide
ing performance across brands. Tannoids can be quantified in a polyphenols; therefore, a stabilizer-free brewing regime must in-
method analogous to the sensitive-protein test by using poly- volve polyphenol-free hop preparations as well as proanthocya-
vinylpyrrolidone monomer (PVP) as precipitant. nidin-free barley.
Finally, there are several procedures available for measuring the Aroma varieties of hops tend to contain higher levels of
redox state of beer, either reducing power (14) or redox coefficient polyphenols; however, it is the nature of these substances which
per se (9). Although it is unequivocally the case that oxidation is matters, not the absolute level. Various polyphenol-free bitterness
an integral aspect of haze production, this type of measurement is preparations are available (39). The major concern for the brewer
no better an indicator of colloidal stability than of flavor stability. should be to use good-quality hops with the appropriate α-acid
In each case, the chemistry is complicated. content for the purpose in question. The author is aware of one
In one of the few authoritative comparisons of predictive tests, instance in which a brewery was left with a substantial quanity of
McMurrough et al (62) used a selection of procedures to evaluate relatively low-α hops that had deteriorated so badly on storage
PVPP treatments of beers. Clearly, values of simple flavanoids that the residual α-acid was extremely low. Undaunted, the brewer
(rather than those which had polymerized into tannoids) and the simply increased the hop charge in the kettle, without dealing with
cooling test are the best predictors of instability. the substantially higher levels of polyphenol that this practice
unavoidably introduced into the process. A severe haze problem
Holistic Approach to Achieving Haze-Stable Beer ensued.
No brewer should expect to deal simply and efficiently with In summary, concerning raw materials (apart from the low-
colloidal stability by the application of one or a couple of down- polyphenol barleys and hop preparations), we might conclude that,
stream stabilization techniques. As much as any other aspect of without knowledge of the varietal contribution to specifically
beer quality, the achievement of a colloidally robust beer depends troublesome flavanoids or a proven link between use of a given
on attention to detail from selection of raw materials all the way variety and haze problems, there is little potential for doing more
through to packaging, storage, and distribution. In this section, I than specifying a maximum nitrogen content in malt (or consid-
summarize the relevance of raw materials and the various process ering dilution of the malt charge with low nitrogen adjuncts) and a
stages for haze stability, and address preventative measures rele- minimum α-acid content for hops.
vant to hazes additional to protein-polyphenol hazes.
Brewhouse
Raw Materials Sweet wort production is best viewed as a protein
The author is unaware of any claim that a given barley variety “precipitation” rather than a protein “hydrolysis” stage because of
is particularly low (or high) in its level of haze-active protein. It is claims that proteolysis during mashing is limited (53). However,
to be expected that there will be a pro rata higher level of poten- the limitations of the Coomassie Blue method used by Lewis et al
tially troublesome material in high-nitrogen batches. However, it have been highlighted (38,90). Low-temperature mashing-in, often
would be wise to be dubious about any claims regarding nitrogen called a “proteolytic stand,” could be re-named a “β-glucanolytic
modification of malt in respect to colloidal stability of proteins stand.” It will certainly be necessary to deal with high-glucan
until there is a clear demonstration that high molecular weight grists, unless exogenous β-glucanases are used. Concerning the
polypeptides have a greater propensity to haze formation than do polyphenols, it is generally accepted that lower pHs are commen-
low molecular weight ones. Based on Siebert’s model, it could be surate with polyphenol precipitation and that prolonged run-off to
argued that relatively low molecular weight polypeptides with extract weaker worts will deliver proportionately more polyphenol
proline residues would enter into a more convoluted (and pre- into the wort if pH is allowed to drift too high (water pH of 6.5–
sumably larger-sized) network than a few very high molecular 7.0). The application of weak wort recycling can hardly be in
weight proteins. On the other hand, the proven ability of papain to keeping with attempts to maximize shelf life. Furthermore, any
enhance colloidal stability is accepted, although its efficacy brewhouse technique which promotes wort turbidity (e.g., exces-
against polypeptides of different sizes is not fully appreciated. The sive use of rakes in lautering) will be disadvantageous in this re-
88 / Bamforth, C. W.
gard, unless the high-particle charge is dealt with in some other when blended back to the mainstream. Temperature and oxygen
way. It is important that there is sufficient calcium in the grist to control are critical; otherwise, the dangers of returning haze pre-
ensure precipitation of oxalate (10). cursors to beer are huge.
Of particular significance for colloidal stability is the kettle Brewers employing PGA as a foam-stabilizer should do so
boil, with a well-formed hot break readily removed in hop back or strictly according to manufacturers’ instructions. Experience
whirlpool representing nothing less than precipitable material that shows that it is worth paying slightly more for PGA already made
would otherwise survive the process to destabilize beer. Apart into solution rather than trying to produce PGA solutions from
from the levels in sweet wort of polyphenols and proteins and the scratch.
extent to which polyphenolics are extracted from the preferred
hopping material, factors influencing the formation and removal Filtration and Packaging
of hot break include the extent of agitation and turbulence (a Beer should be kept as cold as possible en route to and through
“rolling” boil affords enhanced opportunity for precipitation of the filter; otherwise, material precipitated in cold storage will tend
materials at localized surfaces) and the use of coagulants. O’Neill to return to solution and pass through into bright beer. Rigorous
(77) points out that proteins barely in solution at 101°C will be- minimization of oxygen ingress is essential, as is the avoidance of
come insoluble as very fine particles (diameter <2 µm) as wort is any opportunity for heavy metal ions, such as iron and copper, to
cooled to pitching temperature. These are the materials removed enter the beer, because they will activate oxygen to the radical
by agents such as carrageenan, the use of which (as is true for any forms in which it is active. Precautions here include the use only
fining agent) needs to be evaluated empirically by site trials. of additions with low metal loadings, avoidance of recycling on
It is not known whether a recurrence of colloidal stability filters, especially during filter loading when high iron charges can
problems in recent years has mirrored the development of be leached off filter aid and, of course, use of water supplies with
brewhouse operations in which much more attention is paid to low metal content, perhaps achieved by treatment. Filter aid se-
minimizing oxygen ingress. It is claimed that wort oxidation is lection is also critical if there is to be efficient removal of the dead
substantially to the detriment of flavor stability (75) and that, as a bacteria that originate in malt, survive the brewing process, and
consequence, every attempt should be taken to minimize the con- enter into beer where they cause dullness.
tact of wort with air (apart from that required to support fermenta- Attention should be paid to getting the gas contents correctly
tion). However scrutiny of a series of papers (74,76,81,82,87,96) within specification the first time, in order to avoid the need for
reveals that the benefits are by no means dramatic. In contrast, gas washing with its risks of foaming and “skin” formation.
long-established observation reveals that oxidation of polyphenols Cans should be properly screened to ensure the absence of lu-
during wort production is demonstrably to the benefit of haze life bricant residues capable of generating haze.
because polymerization to tannins and their removal as trub is
promoted. It is only downstream that oxidation is to be avoided, if ADDENDUM
the polymerization of monomeric flavanoids is promoted, with Since preparation of this manuscript, a further relevant review paper
such materials passing into beer as a destabilizing influence. has been published by K. J. Siebert: Effect of protein-polyphenol inter-
There seems to be an involvement of oxygen in the formation of actions on beverage haze, stabilization and analysis. J. Agric. Food
cold break (18), although it is felt by some that gas injection has a Chem. 47:353-362, 1999.
physical effect on wort cooling and that nitrogen serves perfectly
well as a replacement for oxygen to promote trub formation in ACKNOWLEDGMENTS
systems where yeast rather than wort is oxygenated.
I thank G. Stewart and Graphics Services at Heriot-Watt University for
the artwork.
Fermentation and Maturation
Other than consideration of cold-break removal (24), there are LITERATURE CITED
few adjustables in fermentation associated with colloidal stability.
It is likely that polyphenols are removed by adsorption onto the 1. Analytica-EBC, Brauerei und Getranke-Rundschau. Method 9.29,
Haze in beer: calibration of haze meters. Verlag Hans Carl, 5th edi-
yeast cell surface and that the extent of this removal will depend
tion, 1997.
on the freshness of the yeast and the extent of the growth. It also 2. Asano, K., and Hashimoto, N. Isolation and characterization of
appears that yeast in poor condition (large number of generations, foaming proteins in beer. J. Amer. Soc. Brew. Chem. 38:129-137,
physical abuse such as centrifugation, and adverse storage condi- 1980.
tions, among other factors) will release polysaccharide materials 3. Asano, K., Shinigawa K., and Hashimoto, N. Characterization of
from within and from its surface which can cause clarity prob- haze-forming proteins of beer and their roles in chill haze forma-
lems. However, the cold-conditioning stage is particularly critical. tion. J. Am. Soc. Brew. Chem. 40:147-154, 1982.
Beer should be chilled to as low a temperature as possible without 4. Bamforth, C. W. Processing and packaging and their effects on beer
freezing (high-gravity beers will stand lower temperatures; there- stability. Ferment 1(5):49-53, 1988.
5. Bellmer, H-G., Galensa, R., and Gromus, J. Bedeutung der
fore, beers produced using high-gravity brewing are potentially
Polyphenols fur die Bierkerstellung Ergebnisse. Brauwelt
more stable than those brewed at sales gravity). The target for a 135:1372-1379 and 1477-1496, 1995.
beer of 5% (v/v) alcohol should be –1°C. Three days at this tem- 6. Bradley, E. L., and Hughes, P. S. Elucidating the role of polyphe-
perature is rather more effective than three weeks at 0°C. Ac- nols in beer quality. Eur. Brew. Conv. Brew. Sci. Group Meet. Edin-
cording to O’Neill (77), “a filter room should show evidence of burgh, 175-191, 1998.
frosted mains.” 7. Buckee, G. K. Identification of hazes, turbidities and sediments.
Judicious use of finings will help sedimentation of particles at Proc. Eur. Brew. Conv.20th Congr. Helsinki, 467-474, 1985.
this stage. For many loss-conscious brewers, a primary considera- 8. Buckee, G. K. State-of-the-art instrumentation in the QC lab. Brew.
tion is the maximal recovery of beer from cold storage. Some will Distill. Int. 25(3):32-35, 1994.
9. Buckee, G. K., Mom, M., Nye, J. W. S., and Hammond, R. V.
blend back sediments on run-off from cold tank, removing them Measurement and significance of oxidation-reduction levels in beer.
by centrifugation. Others will recover beer from tank bottoms in Proc. Euro. Brew. Conv. 27th Congr. Maastricht, 607-614,1997.
various ways, including by cross-flow filtration. Although the 10. Burger, M, Glenister, P., and Becker, K. Oxalate studies in beer. II
economic desirability of these processes is accepted, there is an supplementing wort and observations on calcium-oxalate relation-
unavoidable problem of recovered beer presenting a stability risk ships. Proc. Am. Soc. Brew. Chem. 169, 1956.
Beer Haze / 89
11. Chapon, L. Nephelometry as a method for studying the relations 38. Hii, V., and Herwig, W. C. Determination of high molecular weight
between polyphenols and proteins. J. Inst. Brew. 99:49-56, 1993. proteins in beer using Coomassie Blue. J. Am. Soc. Brew. Chem.
12. Chapon, L. The mechanics of beer stabilization. Brew. Guard 40:46-50, 1982.
123(12):46-50, 1994. 39. Hughes, P. S., and Simpson, W. J. Production and composition of
13. Chapon, L., and Chemardin, M. The dissolving and oxidation of hop products. Tech. Q. Master Brew. Assoc. Am. 30:146-154, 1993.
malt tannoids on mashing-in. Proc. Am. Soc. Brew. Chem. 244-248, 40. Ishibashi, Y., Terano, Y., Fukui, N., Honbou, N., Kakui, T., Kawasaki,
1964. S., and Nakatani, K. Development of a new method for determining
14. Chapon, L., Louis, C., and Chapon, S. Estimates of the reducing beer foam and haze proteins by using the immunochemical method
power of beer using an iron-dipyridyl complex. Proc. Eur. Brew. ELISA. J. Am. Soc. Brew. Chem. 54:177-182, 1996.
Conv. 13th Congr. Estoril, 307-322, 1971. 41. Isoe, A., Kanagawa, K., Ono, M., Nakatani, K., and Nishigaki, M.
15. Clarkson, S. P., Large, P. J., and Bamforth, C. W. Oxygen scaveng- Evaluation of dehusked malt and its influence on the brewing proc-
ing enzymes in barley and malt and their effects during mashing. J ess and beer quality. Proc. Eur. Brew. Conv. 23rd Congr. Lisbon,
Inst. Brew. 98:111-115, 1992. 697-704, 1991.
16. Clarkson, S. P., Large, P. J., Hegarty, P. K., and Bamforth, C. W. 42. Jackson, G., and Bamforth, C. W. Anomalous haze readings due to
Oxygen radicals—their influence on process performance and beta glucans. J. Inst. Brew. 89:155-156, 1983.
product quality. Proc. Eur. Brew. Conv. 22nd Congr. Zurich, 267- 43. Jerumanis, J. Quantitative analysis of flavanoids in barley, hops and
274, 1989. beer by high performance liquid chromatography. J. Inst. Brew.
17. Coote, N., and Kirsop, B. H. A haze consisting largely of pentosan. 91:250-252, 1985.
J. Inst. Brew. 82:34, 1976. 44. Jerumanis, J., van Huynh, N., and Devreux, A. Determination and
18. Crompton, I. E., and Hegarty, P. K. The importance of polyphenols properties of barley and malt polyphenoloxidase. J. Am. Soc. Brew.
in cold break formation. Proc. Eur. Brew. Conv. 23rd Congr. Chem. 34:38-43, 1975.
Lisbon, 625-632, 1991. 45. Kaneda, H., Kano, Y., Osawa, T., Kawakishi, S., and Kamimura,
19. Dadic, M., van Gheluwe, J. E. A., and Valyi, Z. Alkaline steeping M. Effects of free radicals on haze formation in beer. J. Agric.
and the stability of beer. J. Inst. Brew. 82: 273-276, 1976. Food Chem. 38:1909-1912, 1990.
20. Delcour, J. A., Dondeyne, P., Trousdale, E. K., and Singleton, L. 46. Katzke, M., Nendze, R., and Oechsle, D. Die Bierstabilisierung mit
The reaction between polyphenols and aldehydes and the influence lonentauschem. Brauwelt 138:991-994, 1998.
of acetaldehyde on haze formation in beer. J. Inst. Brew. 88:234- 47. Kirby, W., Williams, P. M., Wheeler, R. E., and Jones, M. The im-
243, 1982. portance of the polyphenols of barley and malt in beer sediment and
21. Delcour, J. A., Schoeters, M. M., Meysman, E. W., Dondeyne, P., haze formation. Proc. Eur. Brew. Conv. 16th Congr. Amsterdam,
and Moerman, E. The intrinsic influence of catechins and procya- 415-427, 1977.
nidins on beer haze fon-nation. J. Inst. Brew. 90:381-384, 1984. 48. Kretschmer, K. F. Auswertung der Tannoideergebnisse. Brauwelt
22. Delcour, J. A., Schoeters, M. M., Meysman, E. W., Dondeyne, P., 132:1855-1858, 1992.
Schrevens, E. L., Wijnhoven, J., and Moerman, E. Flavor and haze 49. Krottenthaler, M. Praxistaug Lichkeit von Wintergersteninalz.
stability differences due to hop tannins in all-malt pilsner beers Brauindustrie 82:259, 1997.
brewed with proanthocyanidinfree malts. J. Inst. Brew. 91:88-92, 50. Letters, R. Origin of carbohydrates in beer sediments. J. Inst. Brew.
1985. 75:54-80, 1969.
23. Devreux, A. Wort boiling—compromise between quality and econ- 51. Lewis, M. J., and Poerwantaro, W. M. Release of haze material
omy of energy. Cerevisia 9:97-103, 1984. from the cell walls of agitated yeast. J. Am. Soc. Brew. Chem.
24. Eils, H-G., and Junemann, A. Einsatz eines Separators zur 49:43-46, 1991.
Kuhltrubentfemung. Proc. Euro. Brew. Conv. 27th Congr. 52. Lewis. M. J., Muhleman, D. J., and Krumland, S. C. Beer colloid:
Maastricht, 367-375, 1997. studies with model systems. J. Am. Soc. Brew. Chem. 37:61-65,
25. Erdal, K. Proanthocyanidin-free malt in malting and brewing— 1979.
centenary review. J. Inst. Brew. 92:220-224, 1986. 53. Lewis, M. J., Robertson, I. C., and Dankes, S. U. Proteolysis in the
26. Esnault, E. Beer stabilization with papain. Brew. Guard. 124(1):47- protein rest of mashing—an appraisal. Tech. Q. Master Brew.
49, 1995. Assoc. Am. 29:117-121, 1992.
27. Fernyhough, R., McKeown, I., and McMurrough, I. Beer stabiliza- 54. McKeown, I. P., and Nock, A. Improved beer stabilization using
tion with silica gel. Brew. Guard. 123(10):44-50, 1994. silica gel. Brauwelt. Int. 14:151-155, 1996.
28. Fukal, L., and Kas, J. The role of active and inactivated papain in 55. McManus, J. P., Davis, K. G., Beart, J. E., Gaffney, S. H., Lilley, T.,
beer chillproofing. J. Inst. Brew. 90:247-249, 1984. and Haslam, E. Polyphenol interactions. 1. Introduction; Some ob-
29. Gardner, R. J., and McGuinness, J. D. Complex phenols in brew- servations on the reversible complexation of polyphenols with pro-
ing—a critical survey. Tech. Q. Master Brew. Assoc. Am. 14:250- teins and polysaccharides. J. Chem. Soc. Perkin Trans. 2 11:1429-
261, 1977. 1438, 1985.
30. Gjertsen, P. Beta-glucans in malting and brewing. I Influence of 56. McMurrough, I., and Delcour, J. A. Wort polyphenols. Ferment
beta-glucans on the filtration of strong beers. Proc. Am. Soc. Brew. 3:175-182, 1994.
Chem. 113-120, 1966. 57. McMurrough, I., and Hennigan, G. P. Influence of organic modifi-
31. Glenister, P. R. Application of optical staining in the examination of ers and the isocratic reverse phase high performance liquid chro-
beer sediment. Proc. Am. Soc. Brew. Chem. 192-199, 1971. matography of flavonol glycosides from brewers hops. J.
32. Glenister, P. R. Some useful techniques for the study of beer sedi- Chromatogr. 258:103-109, 1983.
ments. II. Characterizing particles of absorbent materials.: Particle 58. McMurrough, I., Hennigan, G. P., and Cleary, K. Interaction of
counting. Proc. Am. Soc. Brew. Chem. 11-12, 1974. proteoses and polyphenols in worts, bees and model systems. J.
33. Glidewell, S. M., Deighton, N., Goodman, B. A., Troup, G. J., Inst. Brew. 91:93-100, 1985.
Hutton, D. R., Hewitt, D. G., and Hunter, C. R. Free radical scav- 59. McMurrough, I., Hennigan, G. P., and Loughrey, M. J. Quantitative
enging abilities of beverages. Int. J. Food Sci. Tech. 30:535, 1995. analysis of hop flavonols using high performance liquid chromatog-
34. Greif, P., and Schildbach, R. Investigations on the oxalic acid raphy. J. Agric. Food Chem. 30:1102-1106, 1982.
problem in the brewery. Monatsschr. Brauwiss. 31: 275-280, 285, 60. McMurrough, I., Hennigan, G. P., and Loughrey, M. J. Contents of
1978. simple, polymeric and completed flavanol in worts and beers and
35. Grosswiler, O., and Meier, J. Betriebserfahrungen mit der PVPP- their relationship to haze formation. J. Inst. Brew. 89:15-23, 1983.
Bierstabilisierung. Brau. Rundsch. 91:185-191, 1980. 61. McMurrough, I., Kelly, R., Byme, J., and O’Brien, M. Effect of the
36. Guzman, J., and Nock, A. Improved beer stabilization using silica removal of sensitive proteins and proanthocyanidins on the colloi-
adsorbents. Tech. Q. Master Brew. Assoc. Am. 33:185-186, 1996. dal stability of lager beer. J. Am. Soc. Brew. Chem. 50:67-76.
37. Guzman, J., and Nock, A. Use of filtration grade silicas to provide 62. McMurrough, I., Madigan, D., and Kelly, R. J. Evaluation of rapid
combined filtration and beer stabilization benefits. Tech. Q. Master colloidal stabilization with polyvinylpolypyrrolidone (PVPP). J.
Brew. Assoc. Am. 34:272-273, 1997. Am. Soc. Brew. Chem. 55:38-43, 1997.
90 / Bamforth, C. W.
63. McMurrough, I. Madigan, D., Kelly, R. J., and Smyth, M. R. The 84. Outtrup, H. Haze active peptides in beer. Proc. Eur. Brew. Conv.
role of flavanoid polyphenols in beer stability. J. Am. Soc. Brew. 22nd Congr. Zurich, 609-616, 1989.
Chem. 54:141-148, 1996. 85. Ponton, I. D. Proanthocyanidin-free malt. Ferment. 1(1):33-39,
64. McMurrough, I., Madigan, D., and Smyth, M. R. Adsorption by 1988.
polyvinylpolypyrrolidone of catechins and proanthocyanidins from 86. Schneider, J., Raske, W., Drost, M., Meier, H.-U., and Pfenninger,
beer. J. Agric. Food Chem. 43:2687-2691, 1995. H. Prufung der Wirkung von Bierstabilisierungsmitteln im Labor.
65. McMurrough, I., and O’Rourke, T. New insight into the mechanism Brau. Rundsch. 108:227-235, 1997.
of achieving colloidal stability. Tech. Q. Master Brew. Assoc. Am. 87. Schur, F. Effect of oxygen on wort production—full-scale brewing
34:271-277, 1997. trials. Brau. Rundsch. 97:117-120, 1986.
66. McMurrough, I., Roche, G. P., and Hennigan, G. P. Effects of PVPP 88. Sharpe, F. R., and Channon, P. J. Beer haze caused by can lid lubri-
dosage on the flavanoid contents of beer and the consequences for cant. Proc. Eur. Brew. Conv. 21st Congr. Madrid, 599-606, 1987.
beer quality. Brew. Dig. 59(10):28-31, 1984. 89. Siebert, K. J., Carrasco, A., and Lynn, P. Y. Formation of protein-
67. Matsuzawa, K., and Nagashima, T. A new hydrated silica gel for polypehnol haze in beverages. J. Agric. Food Chem. 44:1997-2005,
stabilization of beer. Tech. Q. Master Brew. Assoc. Am. 27:66-72, 1996.
1990. 90. Siebert, K. J., and Knudson, E. J. The relationship of beer high
68. Mills, E. N. C., Kauffman, J. A., Morgan, M. R. A., Field, J. M., molecular weight protein and foam. Tech. Q. Master Brew. Assoc.
Hejgaard, J., Proudlove, M. O., and Onishi, A. Immunological Am. 26:139-146, 1989.
study of hydrophobic polypeptides in beer. J. Agric. Food Chem. 91. Siebert, K. J., and Lynn, P. Y. Mechanisms of beer colloidal stabili-
46:4475-4483, 1998. zation. J Am. Soc. Brew. Chem. 55:73-78, 1997.
69. Muller, R. E. The formation of hydrogen peroxide during oxidation 92. Siebert, K. J., and Lynn, P. Y. Comparison of polyphenol interac-
of thiol-containing proteins. J. Inst. Brew. 104:307-310, 1997. tions with polyvinylpolypyrrolidone and haze-active protein. J. Am.
70. Mussche, R. Physico-chemical stabilization of beer using new gen- Soc. Brew. Chem. 56:24-31, 1998.
eration gallotannins. Proc. 21st Conv. Inst. Brew. (Aust. NZ Sect). 93. Siebert, K. J., Troukhanova, N. V., and Lynn, P. Y. Nature of
136-140, 1990. polyphenol protein interactions. J. Agric. Food Chem. 44:80-85,
71. Mussche, R. Beer stabilization with gallotannin. Brew. Guard. 1996.
123(11):44-49, 1994. 94. Slack, P. T., and Bamforth, C. W. The fractionation of polypeptides
72. Mussche, R., and de Pauw, C. Total stabilization of beer in a single from barley and beer by hydrophobic interaction chromatography:
operation. Proc. 25th Conv. Inst. Brew. (Asia Pacif Sect.) 125-130, the influence of their hydrophobicity on foam stability. J. Inst.
1998. Brew. 89:397-401, 1983.
73. Narziss, L. Filtration and stabilization of beer. Eur. Brew. Conv. 95. Srogl, J., Kosar, K., Mikyska, A., and Bousova, P. Changes occur-
Monogr. XIX. 101-116, 1992. ring in polyphenol substances content during wort production. Proc.
74. Narziss, L., Reicheneder, E., Farber, W., and Freudenstein, L. Oxy- Eur. Brew. Conv. 26th Congr. Maastricht, 275-282, 1997.
gen uptake during mashing: experiments on controlled aeration on 96. Taylor, D. G., Bamber, P., Brown, J. W., and Murray, J. P. Uses of
the minature and semi-technical scale; lautering, wort and beer nitrogen in brewing. Tech. Q. Master Brew. Assoc. Am. 29:137-142,
properties. Brauwelt Int. 5:20-30, 1987. 1992.
75. Narziss, L., Reicheneder, E., and Lustig, S. Oxygen optimization in 97. van Gheluwe, J. E. A., Dadic, M., and Weaver, R. L. Hop and malt
wort preparation. New findings in pilot-plant and commercial scale. phenolics in lager brewing. J. Inst. Brew. 85:23-25, 1979.
Brauwelt Int. 7:238-250, 1989. 98. Verzele, M. 100 years of hop chemistry and its relevance to brew-
76. Ohtsu, K., Hashimoto, N., Inoue, K., and Miyaki, S. Flavor stability ing. Centenary Review. J. Inst. Brew. 92:32-48, 1986.
of packaged beer in relation to the oxidation of wort. Brew. Dig. 99. Walker, M. D., Boume, D. T., and Wenn, R. V. The influence of
61(6):18-23, 1986. malt-derived bacteria on the haze and filterability of wort and beer.
77. O’Neill, M. Advances in beer stabilization. Brew. Guard. Proc. Eur. Brew. Conv. 26th Congr. Maastricht, 191-198, 1997.
125(2):54-58, 1996. 100. Walters, M. T. Natural antioxidants and flavor stability. Ferment.
78. O'Reilly, J. P. The use and function of PVPP in beer stabilization. 10:111-119, 1997.
Brew. Guard. 123(9):32-36, 1994 101. Walters, M. T., Heasman, A. P., and Hughes, P. S. Comparison of
79. O’Rourke, T. The requirements of beer stabilization. Brew. Guard. catechin and ferulic acid as natural antioxidants and their impact on
123(8):30-33, 1994. beer flavor stability. Part 2: extended storage trials. J. Am. Soc.
80. O’Rourke, T. Personal communication, 1998. Brew. Chem. 55:91-98, 1997.
81. O’Rourke, T., Brown, J. W., Theaker, P. D., and Archibald, H. The 102. Walters, M. T., Seefeld, R., Hawthorne, D. B., and Kavanagh, T. E.
effects of oxidation during wort production on the processing and Composition and kinetics of particle formation in beer post pack-
flavor of beer. Proc. 22nd Conv. Inst. Brew. (Aust. NZ. Sect.) 52-58, aging. J. Am. Soc. Brew. Chem. 54:57-61, 1996.
1992. 103. Wenn, R. V., Wheeler, R. E., and Webb, D. J. The prediction of
82. O’Rourke, T., Ianniello, R., McMurrough, I., and Springle, A. The high haze levels in freshly filtered lager beers responsible for the
role of tannoids in the colloidal stabilization of beer. Proc. 25th generation of 'invisible' hazes and of 'bits' in fresh bottled lager.
Conv. Inst. Brew. (Asia Pacif. Sect.) 143-146, 1998. Proc. Eur. Brew. Conv. 22nd Congr. Zurich, 617-624, 1989.
83. Outtrup, H., Fogh, R., and Schaumburg, K. The interaction between
proanthocyanidins and peptides. Proc. Eur. Brew. Conv. 21st Congr.
Madrid, 583-590, 1987. [Received March 26, 1999. Accepted April 26, 1999.]