Water Stress in Biological, Chemical, Pharmaceutical and Food Systems.2015
Water Stress in Biological, Chemical, Pharmaceutical and Food Systems.2015
Water Stress
in Biological,
Chemical,
Pharmaceutical
and Food Systems
Food Engineering Series
Series Editor
Gustavo V. Barbosa-Cánovas, Washington State University, USA
Advisory Board
José Miguel Aguilera, Catholic University, Chile
Kezban Candoǧan, Ankara University, Turkey
J. Peter Clark, Clark Consulting, USA
Richard W. Hartel, University of Wisconsin, USA
Albert Ibarz, University of Lleida, Spain
Jozef Kokini, Purdue University, USA
Michael McCarthy, University of California, USA
Keshavan Niranjan, University of Reading, United Kingdom
Micha Peleg, University of Massachusetts, USA
Shafiur Rahman, Sultan Qaboos University, Oman
M. Anandha Rao, Cornell University, USA
Yrj€o Roos, University College Cork, Ireland
Jorge Welti-Chanes, Monterrey Institute of Technology, Mexico
ISSN 1571-0297
Food Engineering Series
ISBN 978-1-4939-2577-3 ISBN 978-1-4939-2578-0 (eBook)
DOI 10.1007/978-1-4939-2578-0
Springer Science+Business Media LLC New York is part of Springer Science+Business Media
(www.springer.com)
To our families, institutions, fellow
colleagues, and students.
Preface
This book is based on scientific presentations made at the 11th Symposium on the
Properties of Water (ISOPOW 11) held in Querétaro, Mexico, September 5–9,
2010. This publication received decisive and integrated support from ISOPOW
Central Committee and two Mexican academic institutions: Instituto Politécnico
Nacional and Tecnol ogico de Monterrey.
A comprehensive review on novel aspects related to water stress in biological,
chemical, food, and pharmaceutical systems is presented in this book. The chapters
have been written by well-established authorities in their respective fields coming
from research institutions and industry as well as by very promising young scien-
tists making strides on these ever-evolving topics around the properties of water.
One of the premises of this book is that a thorough update requires an in-depth
knowledge of each subject under consideration where a number of disciplines are
effectively integrated. This approach facilitates to appropriately formulate the latest
on fundamental aspects; present the bases for innovative applications; and identify
upcoming developments, as well as research and development needs. The under-
standing of water stresses in complex systems is a true challenge for scientists and
practitioners. The mechanisms involved in the interaction of water with a number of
compounds imply to follow a multi-interdisciplinary approach. The ISOPOW
community is well acquainted with the challenges ahead and the approaches to be
taken whenever dealing with water as part of complex systems. This implies,
among others, to identify ongoing and upcoming efforts towards the use and
inclusion of a diversity of intertwined disciplines, novel tools, and concepts to
successfully integrate existing knowledge with potentially useful new
developments.
ISOPOW is a nonprofit scientific organization which aims to advance the
understanding of the properties of water in chemical, pharmaceutical, biological,
and food systems and on the exploitation of findings on water properties in
improved raw materials, intermediate and final products, as well as processes.
One of the objectives of ISOPOW is to promote high level scientific symposia in
different and targeted places around the world. Another objective is to generate a
vii
viii Preface
publication from each Symposium containing the scientific highlights of this event,
including invited speaker presentations and the results of investigations made by
selected participants.
The first ISOPOW took place in Glasgow, Scotland, in 1974, following the
pioneering initiative of the late Ron B. Duckworth and of Louis B. Rockland. Since
then, all ISOPOWs are recognized for their quality and relevance to the field. Prior
to the Symposium in Querétaro, ten ISOPOWs and two ISOPOW-PRACTICUMs
were held in different parts of the world, marking the global dimension and impact
of these meetings, as well as the interdisciplinary nature of the addressed subjects. It
is worth mentioning that all the ISOPOWs were endorsed and partially sponsored,
among others, by the International Union of Food Science and Technology
(IUFoST). At ISOPOW 11, the latest insights on the scientific aspects of water
properties in foods and other biological systems were presented, keeping the
tradition of exploring and discussing the many roles of water from a multidis-
ciplinary perspective. This is the complete list of ISOPOWs held before the one in
Querétaro, Mexico.
ISOPOW 1 Glasgow, U.K., 1974
ISOPOW 2 Osaka, Japan, 1978
ISOPOW 3 Beaune, France, 1983
ISOPOW 4 Banff, Canada, 1987
ISOPOW Practicum 1 Penang, Malaysia, 1987
ISOPOW 5 Pe~
nı́scola, Spain, 1992
ISOPOW Practicum 2, Puebla, Mexico, 1994
ISOPOW 6 Santa Rosa, USA, 1996
ISOPOW 7 Helsinki, Finland, 1998
ISOPOW 8 Zichron Yaakov, Israel, 2000
ISOPOW 9 Mar del Plata, Argentina, 2004
ISOPOW 10 Bangkok, Thailand, 2007
A round table, chaired by David Reid, was held on the last day of ISOPOW
11 aimed to present the state of the art and to identify research needs on water stress
in biological systems. David Reid, Brad Anderson, David Lechuga, Micha Peleg,
Yrj€o Roos, David Reid, Pilar Buera, and Gustavo Gutiérrez-Lopez contributed a list
of the topics which received the most attention during the entire scientific program.
This promoted a vivid interaction from all the attendees facilitating the formulation
of future endeavors in the topics under consideration. At the end of the meeting,
Gustavo Barbosa-Cánovas presented “Future Research and Suggested Activities on
the Properties of Water.” Some of the highlighted topics in need of attention were as
follows: to the frequent misleading concepts on the concept of freezable and frozen
water; modeling and predicting water behavior under a variety of stresses; dilution
and dissolution in water matrices; plasticizing effect of water; crystallinity or
amorphicity as evaluated by X-rays. A very interesting debate took place on the
extrapolation of equations developed for a number of systems in conditions far from
those at which they are frequently used in water science as well as on exploration
and exploitation of the relationships among aw–water mobility–polymer science
Preface ix
The editors wish to express their gratitude to the following institutions, companies,
and individuals who contributed to making this book possible:
To the Instituto Politécnico Nacional (IPN), Mexico; the Escuela Nacional de
Ciencias Biol ogicas (ENCB-IPN), Mexico; the Tecnol ogico de Monterrey
(ITESEM), México; the Universidad de Buenos Aires (Argentina); Washington
State University, Pullman, USA; and the International Union of Food Science and
Technology (IUFoST)
To our sponsors: IPN, ENCB-IPN, ITESM, IUFoST, CICATA-IPN,
UAQ-México, Kasetsart University (Thailand), Government of Querétero Anton
Paar-Mexico, Coca-Cola de Mexico and Asociaci on de Asociaci on de
Embotelladoras Mexicanas de Coca-Cola, Yakult (Mexico), and Decagon Devices
(Mexico).
To The ISOPOW former Central Committee for trusting Mexico as a convenient
venue for carrying out ISOPOW 11 and for their constant support and encourage-
ment; to the ISOPOW 11 Organizing Committee team for their efficient and most
intensive work and support towards the preparation of this book. Thanks to Melissa
Danae Velarde for image editing and processing; Jeannie C. Bagby, Washington
State University, Pullman, USA, for editing process; to Amor Monroy, Darı́o
Téllez, Raquel González, and Leticia Pe~na for helping the editors prepare the
final version of this book by revising references, formatting the manuscript, and
incorporating in the text the editorial comments.
xi
Contents
xiii
xiv Contents
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 649
Contributors
xix
xx Contributors
B.D. Anderson
Abbreviations
DOPC 1,2-Dioleoyl-sn-glycero-3-phosphocholine
IMC Indomethacin
MD Molecular dynamics
PVAc Polyvinylacetate
PVP Polyvinylpyrrolidone
RH Relative humidity
1 Introduction
received greater emphasis during the selection of new drug candidates, the lead
compounds emerging from these selection processes tend to be more lipophilic and
less water soluble. Consequently, a variety of amorphous or lipid-based delivery
systems such as self-emulsifying or self-microemulsifying lipid dispersions often
administered in soft gelatin capsule form for oral delivery are now being considered
for commercialization. Similarly, various types of colloidal or nanoparticle formu-
lations (e.g., nanosuspensions, solid-lipid nanoparticles, liposomes, micelles, poly-
meric micelles) in which the drug is suspended as a nano-sized particle or
incorporated into a lipid particle have become increasingly attractive as possible
vehicles for enhancing the bioavailability of orally administered drugs having poor
water solubility, controlling drug release, enabling intravenous injection of nearly
insoluble compounds, reducing side effects after intravenous administration, and in
some cases providing enhanced permeability and retention in tumor tissue for the
treatment of cancer.
In all of the aforementioned solid formulations, lipid dispersions, and nanopar-
ticle or colloidal systems, the drug molecules may reside largely in a nonaqueous
environment. Nevertheless, water is ever present in such environments and may
have an impact on the formulation properties that is disproportionate in terms of the
wt% of water present. Removing all water from a product and preventing atmo-
spheric moisture from diffusing into a product are virtually impossible. The small
size of a water molecule enhances its diffusivity through packaging materials and
the drug formulation matrix, while the low molecular weight of water means that
even trace quantities of water may represent a significant molar percentage in
relation to the active pharmaceutical ingredient.
After administration of one of the above solid or lipid-based drug delivery
systems to a patient, the uptake of water into the formulation matrix and its
interactions with various components of the formulation remain of interest but for
quite different reasons. In such cases, water is no longer a contaminant to be
avoided. Rather, it becomes essential in facilitating dispersion of the delivery
system and ultimately governs the kinetics of drug release and overall bioavailabil-
ity. Many lipid dispersions and amorphous solid dispersions have recently been
designed with the intent of producing supersaturated aqueous solutions of the drug
after their administration in order to promote oral bioavailability. For these systems,
understanding how to optimize performance extends beyond knowledge of their
equilibrium properties and will ultimately require a deeper understanding of the
dynamics of water uptake and the effect of water on other processes that influence
delivery system performance.
Our laboratory has been employing both experiments and molecular dynamics
(MD) computer simulations to address a number of questions relating to the state of
water and its impact on the performance of lipid-based delivery systems and
amorphous solid matrices useful for pharmaceutical formulations (Anderson
and Marra 1999; Cao et al. 2004; Xiang and Anderson 2004, 2005, 2013; Rane and
Anderson 2008a, b; Anderson et al. 2010). Some of these questions relate to the state
of water in lipid-based formulations and amorphous solids. For example: (1) What
factors govern the extent and rate of H2O uptake?; (2) Where are water molecules
The State of Water and Its Impact on Pharmaceutical Systems. . . 3
located and how is the water organized?; (3) Do sorbed water molecules alter the
local structure of the matrix?; and (4) Does the presence of water alter the solvent
characteristics of the amorphous or lipid matrix, possibly leading to either enhanced
solubility or perhaps unwanted phase transformations?
For a number of years, we have also had interest in various properties of lipid
bilayer membranes, such as the influence of lipid composition and permeant
structure on permeant binding to bilayer membranes (Xiang and Anderson 1994a,
1999; Xiang et al. 2006; Tejwani et al. 2010, 2011) and membrane transport (Xiang
et al. 1992; Xiang and Anderson 1993, 1994b, c, 1995a, b, 1997, 1998a, b, 2002,
2006; Mayer 2001; Mayer and Anderson 2002; Mayer et al. 2003) and the design of
liposomal drug delivery systems with optimal loading and release characteristics
(Joguparthi and Anderson 2008a, b; Joguparthi et al. 2008a, b; Modi et al. 2012).
Particularly relevant to the present discussion are these questions: (1) Where does
water reside in lipid bilayers?; (2) How does the presence of water in the lipid
bilayer influence drug binding and permeability? These questions will be addressed
in this review through a combination of experiments and molecular dynamics
computer simulations, which can shed light on the molecular phenomena underly-
ing the experimental observations.
X kijb 2 X k θ
ijk
2
U¼ r ij r ijeq þ eq
θijk θijk
bonds
2 angles
2
" #
X X Aij Bij q i q j
þ kd ½1 þ cos ðnðϕ ϕeq ÞÞ þ 6 þ
dihedrals i< j
r 12
ij r ij 4πε0 r ij
The terms in the above equation refer to contributions due to covalent bond
energies, bond angles, dihedrals, and intermolecular van der Waals and electrostatic
interactions. Numerous references available can provide further details on the
potential function or molecular dynamics simulations in general (see, for example,
Allen 2004 and Tieleman et al. 1997).
4 B.D. Anderson
In several recent publications, we have addressed drug solubility, water uptake, and
the state of water in various mono-, di-, or triglycerides or their mixtures that have
been popular formulation components for use in lipid-based drug delivery systems
(Anderson and Marra 1999; Cao et al. 2004; Rane et al. 2008; Rane and Anderson
2008a, b; Anderson et al. 2010). Typical lipid-based delivery vehicles contain
triglycerides of varying chain length combined with surfactants/emulsifiers such
as long-chain carboxylic acids, mono- and diglycerides, phospholipids, or
pegylated versions of these compounds. Water-soluble organic solvents such as
polyethylene glycol or glycerol may also be included. Shown in Fig. 1 are the
structures of a medium-chain triglyceride (tricaprylin) and two medium-chain
monoglycerides (monocaprylin and monocaprin) often used in lipid-based formu-
lations. These lipids were employed in experimental studies in our laboratory,
aimed at understanding the specific intermolecular interactions that govern water
uptake into these lipid solvents. As shown by the phase diagram in Fig. 1, Friberg
and Mandell (1970) demonstrated that mixtures containing monocaprylin and
tricaprylin form microemulsions upon dilution with water. Similar behavior
would be expected in the stomach after administration of a pre-concentrate to
Fig. 1 Molecular structures of tricaprylin, monocaprylin, and monocaprin used as model lipids
for studies of water uptake and solubility. The phase diagram for systems containing varying
percentages of tricaprylin, monocaprylin, and water from Friberg and Mandell (1970) illustrates
the microemulsion formation that occurs when mixtures of tricaprylin and monocaprylin contain
small percentages of water. Right panel reprinted with kind permission from Springer Science
+ Business Media: J Am Oil Chem Soc, Phase equilibria and their influence on the properties of
emulsions, 47 (5), 1970, 149–152, Friberg S, Mandell L, figure number 2
The State of Water and Its Impact on Pharmaceutical Systems. . . 5
patients. Thus, the dynamics of water uptake into these solvent mixtures leading to
the formation of stable microemulsion particles and the thermodynamics associated
with their stability and solvent characteristics are important features underlying
their use in enhancing oral drug bioavailability.
Given the diversity and complexity of such lipid mixtures, simple quantitative
principles that govern water uptake, its organization, its effects on organization of
the lipids, and its effects on the solubility of drugs in lipid formulations would be
desirable to elucidate. We have been interested, in particular, in the important role
of hydrogen bonding, both in determining water uptake and its organization in lipid
mixtures. Figure 2 (left panel) illustrates the water uptake as a function of mono-
glyceride concentration in triglyceride/monoglyceride mixtures at 37 C and 100 %
relative humidity, either in the presence of an added solute (benzamide) at its
saturation solubility or in the absence of solute. Triglycerides consisting of
tricaprylin and monoglycerides consisting of either monocaprylin (C8) or
monocaprin (C10) were employed. Water content increased approximately linearly
with the relative humidity (Rane et al. 2008) and, as evidenced by the profiles, was
similar at the same molar concentration of either monoglyceride, indicating that the
most important variable is the molar concentration of hydroxyl groups in the lipid
mixture. The effects of monoglyceride concentration and water uptake on the
solubility of benzamide are shown in the right panel of Fig. 2. An increase in the
concentration of monoglyceride in the lipid mixture resulted in corresponding
increases in benzamide solubility. As benzamide is an amide-containing solute,
this enhanced solubility is likely due to the formation of hydrogen bonds between
the amide carbonyl and hydroxyl groups in the lipid mixtures. As shown by the
10 0.7
C8 + benzamide C8
C10 + benzamide C10
C8 (no solute) 0.6
8 C10 (no solute)
0.5
% wt/wt water
mol/Lsolute
100% RH
6 0.4
4 0.3
0.2
2 6% RH
0.1
0 0
0 0.5 1 1.5 2 0 0.5 1 1.5 2
mol/L monoglyceride mol/L monoglyceride
Fig. 3 Left panel: Snapshot from a molecular dynamics simulation showing the distribution of
water molecules in a water saturated 60 % tricaprylin/40 % monocaprylin lipid mixture at 37 C.
Water molecules are rarely found to be hydrogen bonded to the single molecule of benzamide.
Right panel: Monocaprylin chains hydrogen bonded to benzamide
The State of Water and Its Impact on Pharmaceutical Systems. . . 7
Unlike moisture uptake in crystals, where stepwise changes in water content occur
with increasing relative humidity due to formation of stoichiometric hydrates, water
uptake in amorphous solids increases in a continuous fashion with relative humid-
ity, though the increases are not typically linear. Studies of moisture uptake in
polyvinylpyrrolidone (PVP), polyvinylacetate (PVAc), and their copolymers (Tay-
lor et al. 2001) demonstrated an important role of the functional group composition
of the polymer in determining water affinity. PVP exhibits substantially greater
water affinity than PVAc, presumably due to the greater hydrogen-bond accepting
ability of the amide carbonyl in PVP. The water uptake obtained in PVP/PVAc
copolymers was close to that predicted assuming additivity of the functional group
contributions from the individual PVP and PVAc chains, as expected if hydrogen
bond formation were the dominant factor governing water uptake.
Molecular interactions such as hydrogen bonding may be significantly affected,
however, by the degree of metastability of a given amorphous glass. Dawson
et al. (2009) recently found substantial differences in the water uptake profiles for
an ordinary indomethacin (IMC) glass prepared by cooling, an annealed glass, and a
more stable glass prepared by vapor deposition, with the more stable glass
exhibiting a fivefold lower moisture uptake than an ordinary glass. Dramatic
differences in the rates of water uptake were also noted. The authors speculated
8 B.D. Anderson
that water sorption into indomethacin is driven by hydrogen bonding and that in
stable IMC glasses, more hydrogen bonds are formed between IMC molecules
themselves, leaving fewer hydrogen-bonding sites available for water molecules.
This competition would account for the reduced water uptake in more stable IMC
glasses. Physical aging of trehalose for 120 h at 373 K resulted in a decrease in
equilibrium moisture uptake from 1.4 to 0.8 % at 10 % relative humidity (Surana
et al. 2004), again suggesting that water uptake in amorphous solids depends not
only on the functional groups present in the amorphous material but also on their
availability for interaction with water.
We have recently conducted an MD simulation of an amorphous indomethacin
glass to investigate various structural and dynamic properties, including the inter-
actions of water with IMC. The assembly consisted of 105 indomethacin molecules
and 12 water molecules, corresponding to a water content of 0.6 % w/w (Xiang and
Anderson 2012). At this relatively low water content, >90 % of the water molecules
at any given time formed 1–3 hydrogen bonds with indomethacin with probabilities
of 0.36 0.13, 0.28 0.07, and 0.29 0.10, respectively. Figure 4 (left panel) is a
snapshot of the distribution of water molecules in IMC indicating that, even at
0.6 % w/w water content, some water dimers co-exist with monomers. The right
panel in Fig. 4 shows the probability distribution for either the IMC carboxylic acid
group or water to donate a hydrogen to various atoms. Water molecules acting as
hydrogen donors have a similar tendency to form hydrogen bonds with the IMC
benzoyl C ¼ O (39 8 %) and the carboxylic acid C ¼ O (37 12 %) with a
smaller percentage forming hydrogen bonds at the OCH3 group (12 6 %) or
with other water molecules (12 9 %).
Fig. 4 Left panel: A representative snapshot showing the water distribution in an amorphous IMC
glass at 298 K. Right panel: Probability distributions for different HB acceptor sites to form
hydrogen bonds with either the IMC –COOH (black) or the water H–O groups (white) in the
simulated IMC glass at 298 K
The State of Water and Its Impact on Pharmaceutical Systems. . . 9
Several investigators have examined the state of water in amorphous PVP, with
the following general conclusions. Lebedeva et al. (1999) determined from FTIR
spectra of sorbed water in PVP at lower water contents that water molecules were
predominantly hydrogen bonded to PVP carbonyl oxygen atoms. At higher water
concentrations, experimental data suggest that water self-associates in PVP. For
example, Taylor et al. concluded from peak shifts in the Raman spectra for the PVP
carbonyl and water at various relative humidities that some water molecules were
hydrogen bonded directly to the polymer, while additional water molecules were
hydrogen bonded to each other (Taylor et al. 2001).
Our group has also conducted molecular dynamics simulations to explore the
distribution and plasticization effects of water, along with other phenomena in
PVP polymer assemblies containing six PVP chains (each having 40 monomers)
with 8–167 water molecules, corresponding to approximately 0.5 % and 10 % w/w
water, respectively (Xiang and Anderson 2004, 2005). Consistent with the exper-
imental evidence, we found that at higher water content, water self-associates to
form hydrogen-bonded strands or clusters, similar to the pattern previously
observed in lipid mixtures. As shown in Fig. 5, this tendency for water to
Fig. 5 Water clustering with increasing water content as demonstrated in molecular dynamics
simulations in various polymers. Left panel reprinted with kind permission from Springer Science +
Business Media: Pharm Res, Distribution and Effect of Water Content on Molecular Mobility in
Poly(vinylpyrrolidone) Glasses: A Molecular Dynamics Simulation, 22 (8), 2005, 1205–1214,
Xiang T-X, Anderson BD, figure number 4. Middle panel reprinted with kind permission from
American Chemical Society: Macromolecules, Molecular Dynamics Simulation Study of Water in
Amorphous Kapton, 41 (9), 2008, 3349–3362, Marque G, Neyertz S, Verdu J, Prunier V and
Brown D, figure number 16. Right panel reprinted with kind permission from Elsevier: Polymer,
Effect of absorbed water on oxygen transport in EVOH matrices: A molecular dynamics study,
45 (10), 2004, 3555–3564, Kucukpinar E, Doruker P, figure number 5
10 B.D. Anderson
Numerous experimental (Xiang et al. 1992; Xiang and Anderson 1993, 1994b, c,
1995a, b, 1997, 1998a, b; Mayer 2001; Mayer and Anderson 2002; Mayer
et al. 2003) and computational (Xiang and Anderson 2002, 2006) studies in our
laboratories and others have established that the barrier domain for small drug-like
molecular permeability across lipid bilayer membranes is the hydrocarbon-like acyl
chain interior. This region is nearly devoid of water, as illustrated by the results of a
recent MD simulation in 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC) bila-
yers, where the center of the bilayer is indicated at 0.0 Å (Fig. 6, left panel). Given
that most drug molecules have one or more polar functional groups, the dislocation
1.0
Water 4-Ethylphenol
Phosphate (x6) Phenylethylamine
Carbonyl (x6) Tyramine
Nitrogen (x6)
0.8
-2
0.4
-4
0.2
0.0
-6
of drug molecules from the interfacial region of the bilayer where the polar groups
are hydrogen bonded to water or phospholipid head groups into the hydrocarbon
interior is energetically highly unfavorable.
We recently explored the membrane partitioning and permeabilities of a set of
structurally related permeants, as shown in Fig. 7 (i.e., phenethylamine,
4-ethylphenol, and tyramine), in an effort to couple experimental results to results
from MD simulations (Tejwani et al. 2010, 2011). In particular, the effects of the
hydroxyl and amine substituents, either alone or when both were present on the
same molecule (as in tyramine) on the free energy profiles within DOPC bilayers,
were examined, as shown in the right panel of Fig. 6. Two prominent features were
noted in the free energy profiles. The water/bilayer interface was the preferred
binding region for all solutes, while the highest free energy region (least favorable
for partitioning) was the center of the bilayer. The barrier domain region in the
hydrocarbon chain interior was found to be highly selective to the number of polar
residues on the permeant, while the preferred binding region at the interface was
relatively insensitive to such differences in chemical structure.
These differences in chemical selectivity between the interfacial region and the
hydrocarbon interior reflect the abilities of permeants to adopt orientations that
favor hydrogen bonding with water or polar head groups in the interfacial region,
while in the hydrocarbon chain interior all hydrogen bonds are lost, unless the
molecule is sufficiently large to span a significant fraction of the bilayer thickness.
This is shown by the number of hydrogen bond interactions that each of the three
12 B.D. Anderson
solutes formed at various locations within the bilayer, as illustrated in Fig. 7. As the
permeants move away from the bilayer interface, a substantial fraction of the
hydrogen bonds are formed with water molecules that have partially penetrated
the bilayer.
In some snapshots from the MD simulations, we were able to observe that
removal of a solute from the interface was facilitated by hydrogen bonding to
water molecules, which were in turn linked to either another water molecule closer
to the interface or one of the polar atoms in the phospholipid head group.
Water wires involving multiple hydrogen-bonded water molecule chains accompa-
nied by head-group migration appear to be particularly important in facilitating
anion and cation transport across bilayers (Xiang and Anderson 2006; Tepper and
Voth 2006; MacCallum et al. 2008), since the free energy penalty for removing
water of solvation from a charged residue is extremely large.
6 Conclusions
Acknowledgements The author would like to acknowledge the contributions of several col-
leagues or former colleagues who were largely responsible for the results in the publications from
our laboratory cited in this chapter, including Drs. Tian-xiang Xiang, Michelle Marra, Yichen Cao,
Sagar Rane, Peter Mayer, Riku Niemi, Ravi Tejwani, Vijay Joguparthi, and Sweta Modi.
References
Cao Y, Marra M, Anderson BD (2004) Predictive relationships for the effects of triglyceride ester
concentration and water uptake on solubility and partitioning of small molecules into lipid
vehicles. J Pharm Sci 93:2768–2779
Dawson KJ, Kearns KL, Ediger MD, Sacchetti MJ, Zografi GD (2009) Highly stable indomethacin
glasses resist uptake of water vapor. J Phys Chem B 113:2422–2427
Friberg S, Mandell L (1970) Phase equilibria and their influence on the properties of emulsions.
J Am Oil Chem Soc 47:149–152
Joguparthi V, Anderson BD (2008a) Effect of cyclodextrin complexation on the liposome perme-
ability of a model hydrophobic weak acid. Pharm Res 25:2505–2515
Joguparthi V, Anderson BD (2008b) Liposomal delivery of hydrophobic weak acids: enhancement
of drug retention using a high intraliposomal pH. J Pharm Sci 97:433–454
Joguparthi V, Feng S, Anderson BD (2008a) Determination of intraliposomal pH and its effect on
membrane partitioning and passive loading of a hydrophobic camptothecin, DB-67. Int J
Pharm 352:17–28
Joguparthi V, Xiang T-X, Anderson BD (2008b) Liposome transport of hydrophobic drugs: gel
phase lipid bilayer permeability and partitioning of the lactone form of a hydrophobic
camptothecin, DB-67. J Pharm Sci 97:400–420
Kucukpinar E, Doruker P (2004) Effect of absorbed water on oxygen transport in EVOH matrices.
A molecular dynamics study. Polymer 45:3555–3564
Lebedeva TL, Feldstein M, Kuptsov SA, Plate NA (1999) The products of water H-bonding to
poly(N-vinyl pyrrolidone) in solid state. In: Spectroscopy of biological molecules: new
directions. 8th European conference on the spectroscopy of biological molecules, Enschede,
Netherlands, pp 581–582
MacCallum JL, Bennett WF, Tieleman DP (2008) Distribution of amino acids in a lipid bilayer
from computer simulations. Biophys J 94:3393–3404
Marque G, Neyertz S, Verdu J, Prunier V, Brown D (2008) Molecular dynamics simulation study
of water in amorphous kapton. Macromolecules 41:3349–3362
Mayer PT (2001) Functional group contributions to peptide permeability in lipid bilayer mem-
branes: the role of peptide conformation and length. University of Utah, Salt Lake City, p 227
Mayer PT, Anderson BD (2002) Transport across 1,9-decadiene precisely mimics the chemical
selectivity of the barrier domain in egg lecithin bilayers. J Pharm Sci 91:640–646
Mayer PT, Xiang T-X, Niemi R, Anderson BD (2003) A hydrophobicity scale for the lipid bilayer
barrier domain from peptide permeabilities: nonadditivities in residue contributions. Biochem-
istry 42:1624–1636
Modi S, Xiang T-X, Anderson BD (2012) Enhanced active liposomal loading of a poorly soluble
ionizable drug using supersaturated drug solutions. J Control Release 162:330–339
Rane SS, Anderson BD (2008a) Molecular dynamics simulations of functional group effects
on solvation thermodynamics of model solutes in decane and tricaprylin. Mol Pharm 5
(6):1023–1036
Rane SS, Anderson BD (2008b) What determines drug solubility in lipid vehicles: is it predict-
able? Adv Drug Deliv Rev 60:638–656
Rane S, Cao Y, Anderson BD (2008) Quantitative solubility relationships and the effect of water
uptake in triglyceride/monoglyceride microemulsions. Pharm Res 25:1158–1174
Surana R, Pyne A, Suryanarayanan R (2004) Effect of preparation method on physical properties
of amorphous trehalose. Pharm Res 21:1167–1176
Taylor LS, Langkilde FW, Zografi G (2001) Fourier transform Raman spectroscopic study of the
interaction of water vapor with amorphous polymers. J Pharm Sci 90:888–901
Tejwani RW, Davis ME, Anderson BD, Stouch TR (2010) Functional group dependence of solute
partitioning to various locations in a DOPC bilayer: I. A comparison of molecular dynamics
simulations with experiment. J Pharm Sci 100:2636–2646
Tejwani RW, Davis ME, Anderson BD, Stouch TR (2011) An atomic and molecular view of the
depth dependence of the free energies of solute transfer from water into lipid bilayers. Mol
Pharm 8:2204–2215
14 B.D. Anderson
Tepper HL, Voth GA (2006) Mechanisms of passive ion permeation through lipid bilayers:
insights from simulations. J Phys Chem B 110:21327–21337
Tieleman DP, Marrink SJ, Berendsen HJ (1997) A computer perspective of membranes: molecular
dynamics studies of lipid bilayer systems. Biochim Biophys Acta 1331(3):235–270
Xiang T-X, Anderson BD (1993) Diffusion of ionizable solutes across planar lipid bilayer
membranes: boundary layer pH gradients and the effects of buffers. Pharm Res 10:1654–1661
Xiang T-X, Anderson BD (1994a) Molecular distributions in interphases: a statistical mechanical
theory combined with molecular dynamics simulation. Biophys J 66:561–573
Xiang T-X, Anderson BD (1994b) Substituent contributions to the transport of substituted p-toluic
acids in lipid bilayer membranes. J Pharm Sci 83:1511–1518
Xiang T-X, Anderson BD (1994c) The relationship between permeant size and permeability in
lipid bilayer membranes. J Membr Biol 140:111–121
Xiang T-X, Anderson BD (1995a) Development of a combined paramagnetic ion-induced line
broadening/dynamic light scattering method for permeability measurements across lipid
bilayer membranes. J Pharm Sci 84:1308–1315
Xiang T-X, Anderson BD (1995b) Phospholipid surface density determines the partitioning and
permeability behavior of acetic acid in DMPC:cholesterol bilayers. J Membr Biol 148:157–167
Xiang T-X, Anderson BD (1997) Permeability of acetic acid across gel and liquid-crystalline lipid
bilayers conforms to free-surface-area theory. Biophys J 72:223–237
Xiang T-X, Anderson BD (1998a) Influence of chain ordering on the selectivity of dipalmitoylpho-
sphatidylcholine bilayer membranes for permeant size and shape. Biophys J 75:2658–2671
Xiang T-X, Anderson BD (1998b) Phase structures of binary lipid bilayers as revealed by
permeability of small molecules. Biochim Biophys Acta 1370:64–76
Xiang T-X, Anderson BD (1999) Molecular dissolution processes in lipid bilayers: a molecular
dynamics simulation. J Chem Phys 110:1807–1818
Xiang T-X, Anderson BD (2002) A computer simulation of functional group contributions to free
energy in water and a DPPC lipid bilayer. Biophys J 82:2052–2066
Xiang T-X, Anderson BD (2004) A molecular dynamics simulation of reactant mobility in an
amorphous formulation of a peptide in poly(vinylpyrrolidone). J Pharm Sci 93:855–876
Xiang T-X, Anderson BD (2005) Distribution and effect of water content on molecular mobility in
poly(vinylpyrrolidone) glasses: a molecular dynamics simulation. Pharm Res 22:1205–1214
Xiang T-X, Anderson BD (2006) Liposomal drug transport: a molecular perspective from molec-
ular dynamics simulations in lipid bilayers. Adv Drug Deliv Rev 58:1357–1378
Xiang TX, Anderson BD (2013) Molecular dynamics simulation of amorphous indomethacin. Mol
Pharm 10(1):102–14
Xiang T-X, Chen X, Anderson BD (1992) Transport methods for probing the barrier domain of
lipid bilayer membranes. Biophys J 63:78–88
Xiang T-X, Jiang Z-Q, Anderson BD (2006) Molecular dynamics simulations and experimental
studies of binding and mobility of 7-t-butyldimethylsilyl-10-hydroxycamptothecin (DB-67)
and its 20(S)-4-aminobutyrate ester in DMPC membranes. Mol Pharm 3:589–600
Food Preservation by Nanostructures-Water
Interactions Control
Abbreviations
aw Water activity
FD Fractal dimension
K Boltzmann’s constant
N Entire system
n Cooperative rearrangement
Nn Regions whose state does not allow such a transition
RH Relative humidity
Sc Configurational entropy of the system
sc Configurational entropy of the subsystem with z molecules
T Absolute temperature
T1 Spin-lattice relaxation time
T2 s Spin-spin relaxation time
1 Introduction
Initially, the water content was used as a control parameter for predicting the
stability of food. However, it was insufficient to determine the chemical and
microbiological stability since they do not depend only on the water content, but
on how the water is bound to the food components (Rahman 2010). Under the same
conditions of environmental relative humidity (RH %), different foods may have
different equilibrium moisture content, while at the same moisture content, these
foods may differ completely in their physicochemical and microbiological stability
(Vermeulen et al. 2014; Kalichevsky-Dong 2000).
Typically, to define the conditions under which food must be preserved, espe-
cially dehydrated foods, the water activity (aw) and glass transition temperature (Tg)
are employed (Mrad et al. 2013; Roos 1993). This is a combination of the use of
water-matrix energetic interactions (aw) and low molecular mobility (T < Tg) in
order to allow for the product quality to be maintained longer. However, many real
storage conditions do not match the values predicted by these parameters.
In recent years, the effect of microstructure on the chemical and microbiological
stability of foods has been widely recognized (Kong and Singh 2011; Creed 2010;
Azuara-Nieto and Beristain-Guevara 2007); Azuara and Beristain 2006; Hills
et al. 2001) and the control of the food structure has been performed by manipu-
lating the operation conditions in different processes such as convective drying,
freeze drying and freezing, among others, so allowing to evaluate the properties of
the final product at macroscopic level (Anwar and Kunz 2011; Khraisheh
et al. 2004). Nonetheless, there is scarce research on the microstructural changes
in the food solid matrix due to processing and its relation to the behavior of water at
nanometric scale.
Nanoscience deals with the study of phenomena on a scale of 0.1–100 nm and
the manipulation of materials at the atomic, molecular, and macromolecular scales,
whose properties may differ completely from those at the macroscopic level.
Nanotechnology, on the other hand, deals with the design, characterization, pro-
duction, and application of structures such as devices, systems, and materials by
controlling the shape and size at the nanoscale (Chellaram et al. 2014; Knowles
et al. 2006; Salerno et al. 2008).
Studying the behavior of fluids on surfaces and microstructures has been impor-
tant in some areas of science and engineering, including food science, since the
properties of nanoconfined fluids markedly differ from those non-nanoconfined
(Ortony et al. 2014; Biggin and Sansom 2001). The aim of this chapter is to
examine how nanostructuration can modify food energy barriers and steric inter-
actions between water and food matrix and how this can be helpful to modify in a
controlled fashion the conditions to preserve products.
Food Preservation by Nanostructures-Water Interactions Control 17
2 Parameters of Stability
The glass transition temperature (Tg) is defined as the temperature where the
amorphous polymers undergo a transition between glassy and rubbery states. The
amorphous and partially crystalline structures are formed during some food
processing operations, such as extrusion, baking, concentration, and drying (Roos
et al. 1996). The properties of amorphous polymers can be related to their glass
transition temperature which is important to the physical stability of foods. The Tg
of a material depends on the moisture content, degree of crystallinity of the sample,
and the rate of temperature change (Kayacier and Singh 2002). It has been proposed
that the glass transition affects those chemical and enzymatic reactions which are
controlled by diffusion, since changes in the molecular mobility occur as the
temperature decreases to the vicinity of the Tg (Slade and Levine 1991). These
reactions in foods are associated with changes in moisture content, aw, temperature,
pH, reactant concentration, and structure (Bell and Touma 1996). In the glassy state
(T < Tg), the mobility of the reactive molecules is reduced due to inter- and
intramolecular arrangements of the food matrix, which increases the viscosity of
the medium and reduces the deterioration rate of the diffusion limited processes. In
contrast, at temperatures above the glass transition temperature (T > Tg), the mate-
rial is in a rubbery state due to a decrease in viscosity and an increase in molecular
mobility and the probability of deteriorating reactions (Roos 1993; Kalichevsky-
Dong 2000; Ludescher et al. 2001). Regarding microbial stability, Slade and Levine
(1991) proposed to replace the use of aw with the relationship Tg:water content.
Also, Gould and Christian (1998) indicated that the very viscous states (glassy
state) of foods can interfere with microbial growth due to restriction of the diffusion
of nutrients. However, Chirife and Buera (1994) demonstrated that microbial
18 E. Flores-Andrade et al.
growth in the foods can occur in both, glassy (T < Tg) and rubbery, states (T > Tg)
and indicated that the microbial stability primarily depends on the aw.
3 Thermodynamic Parameters
Beristain and Azuara (1990) developed a method which allows recommending the
most appropriate storage conditions for dry products if the moisture level at which
the minimum integral entropy of water molecules occurs is known. This point is
considered as the one with the maximum order of water molecules adsorbed on the
food microstructure and sometimes does not match neither the monolayer value nor
the predicted Tg for preserving foodstuffs (Beristain et al. 2002). Furthermore, it has
been proven that the relationship between the thermodynamic parameters of
enthalpy and entropy allows determining the mechanism that controls the interac-
tion of water molecules with food. The enthalpic control relates with the energy
aspects characteristic of the chemical composition of food, whereas the entropic
control refers to microstructural factors whose interaction with water molecules is
independent of ionic or polar affinity between the adsorbate and adsorbent mole-
cules (Beristain et al. 1996; Azuara and Beristain 2006).
Carbon nanotubes have been used as ideal models for studying the behavior of fluids
in channels of biological materials (cells, membranes, surface proteins,
microcapillaries, and aquaporins) and micro- and mesoporous solid matrices
(Huang et al. 2005; Liu et al. 2013; Peponi et al. 2014). The behavior of water
confined in nanometric spaces differs from that of the water at the macroscopic level.
Several studies with multiwalled carbon nanotubes whose diameters are about
5 nm have shown that the behavior of water is not the same as the one in nanotubes
with larger diameters (Sealy 2004). Moreover, in single-walled nanotubes with
diameter lower than 1.5 nm, it occurs spontaneous and continuous filling of water
molecules arranged in a linear fashion (in the middle of the nanotube) and tetrahe-
dral order (water molecules in contact with the wall of the nanotube). This property
allows for exploitation of nanotubes as channels for water and protons. Although
the modification of the hydrophilic and hydrophobic properties of the nanotube wall
does not affect the structure of water molecules, this has an effect on the transport
through the nanotubes. Another characteristic of water molecules in nanometric
spaces is that they orderly occupy spaces when are near to the walls of the
nanotubes, preferably orienting their dipole moment perpendicularly to such
walls (Hummer et al. 2001; Mara~non and Mara~non 2003).
Food Preservation by Nanostructures-Water Interactions Control 19
Adam and Gibbs (1965) developed a theory that relates the structure relaxation time
(kinetic parameter) with the configurational entropy (thermodynamic parameter) of
the system. This theory is based on the molecular mobility of fluids at temperatures
approaching the Tg in terms of molecular cooperatively rearranging regions.
According to this theory, the molecular movement occurs in regions of critical
size z* (atoms, molecules, etc.), and the probability of configurational modifications
20 E. Flores-Andrade et al.
Regarding microbial stability, it has been demonstrated that, at the same aw,
microorganism survival depends on microstructural aspects of the support medium
used for microbial growth (Hills et al. 2001). Probably due to entropic and energy
mechanisms already considered, the water contained in the micropores is practi-
cally inaccessible to the microorganism, which makes difficult its growth. There-
fore, the distribution of water in a solid matrix microstructure depends on its nature
and the nature of the microorganism, so that food nanostructuration may allow for a
better control against microbial growth.
The term “undefined nanostructuration” could be applied to the non-ordered
microstructure produced by some kind of processing which allows obtaining
nanometric spaces, such as micropores (with <20 Å pore diameter), in the final
product and influences the setting of stability conditions.
Fractal geometry is a branch of mathematics that is based upon the use of algo-
rithms and fractional dimensions to describe the morphology of objects which are
not adequately described by Euclidean geometry (Garcı́a-Armenta et al. 2014;
Mandelbrot and Riedi 1997). A fractal is a geometric figure composed of fragments
in an infinite variety of sizes, so that each of them is a reduced copy of the whole.
The latter property of fractals is called self-similarity, i.e., fractals are scale
independent and generally characterized by the presence of infinite detail, infinite
length, and the lack of smoothness or differentiability.
A natural fractal is a figure that shows self-similarity from a statistical point of
view, within a range of magnification at which the profile is analyzed (Barletta and
Barbosa-Cánovas 1993). One way to assess the heterogeneity of a solid (a food, for
instance) is in terms of the fractal dimension (FD), which may be obtained by
various methods such as sorption isotherms (Fripiat et al. 1986; Wang and Li 1997;
Calzetta-Resio et al. 1999) and image analysis. In the latter, FD is calculated from
the boundary (Alamilla-Beltrán et al. 2005) or the surface (Chanona et al. 2003) of
the material.
The modifications in the food morphology during processing can be described in
terms of the fractal dimension. Chanona et al. (2003) studied the change in the
heterogeneity of the surface of a model food system during convective drying and
concluded that the FD may be a quantitative index of the heterogeneity of the
material. Thereby, the use of fractal theory may help to characterize nanostructured
cavities in foods.
Food Preservation by Nanostructures-Water Interactions Control 23
8 Conclusions
References
Chui MM, Phillips RJ, McCarthy MJ (1995) Measurement of the porous microstructure of
hydrogels by nuclear magnetic resonance. J Colloid Interface Sci 174(2):336–344
Corezzi S, Fioretto D, Casalini R, Rolla PA (2002) Glass transition of an epoxy resin induced by
temperature, pressure and chemical conversion: a rationale based on configurational entropy.
J Non-Cryst Solids 307–310:281–287
Creed PG (2010) Chemical deterioration and physical instability in ready-to-eat meals and catered
foods. In: Skibsted LH, Risbo J, Andersen ML (eds) Chemical deterioration and physical
instability of food and beverages. Woodhead Publishing, Cambridge, pp 608–662
Djendoubi Mrad N, Bonazzi C, Courtois F, Kechaou N, Mihoubi NB (2013) Moisture desorption
isotherms and glass transition temperatures of osmo-dehydrated apple and pear. Food
Bioproducts Processing 91(2):121–128
Flores-Andrade E, Beristain CI, Vernon-Carter EJ, Gutiérrez-L opez GF, Azuara E (2009)
Enthalpy-entropy compensation and water transfer mechanism in osmotically dehydrated
agar gel. Drying Technol 27(6):999–1009
Fripiat JJ, Gatineau L, Van Damme H (1986) Multilayer physical adsorption on fractal surfaces.
Langmuir 2(5):562–567
Garcı́a-Armenta E, Téllez-Medina DI, Alamilla-Beltrán L, Arana-Errasquı́n R, Hernández-
Sánchez H, Gutiérrez-Lopez GF (2014) Multifractal breakage patterns of thick maltodextrin
agglomerates. Powder Technol 266:440–446
Gould GW, Christian JHB (1998) Characterization of the state of water in foods - biological
aspects. In: Seow C (ed) Food preservation by moisture control. Elsevier Applied Science,
London, pp 43–56
Hills BP, Arnould L, Bossu C, Ridge YP (2001) Microstructural factors controlling the survival of
food-borne pathogens in porous media. Int J Food Microbiol 66(3):163–173
Huang H, Liu CH, Wu Y, Fan S (2005) Aligned carbon nanotube composite films for thermal
management. Adv Mater 17(13):1652–1656
Hummer G, Rasaiah JC, Noworyta JP (2001) Water conduction through the hydrophobic channel
of a carbon nanotube. Nature 414:188–190
Jani AMMD, Losic D, Voelcker NH (2013) Nanoporous anodic aluminium oxide: advances in
surface engineering and emerging applications. Prog Mater Sci 58(5):636–704
Johari GP (2000) On the origin of the heat capacity feature of annealed ices and ice clathrates, and
interpreting water’s diffusivity in terms of the entropy. Chem Phys 258(2–3):277–290
Kalichevsky-Dong MT (2000) The glass transition and microbial stability. In: Kilcast D,
Subramaniam P (eds) The stability and shelf-life of food. Woodhead Publishing, Cambridge,
pp 25–54
Kayacier A, Singh RK (2002) Glass transition studies of baked tortilla chips using dynamic
mechanical thermal analysis. LWT Food Sci Technol 35(1):34–37
Khraisheh MAM, McMinn WAM, Magee TRA (2004) Quality and structural changes in starchy
foods during microwave and convective drying. Food Res Int 37(5):497–503
Knowles MRH, Rutterford G, Karnakis D, Dobrev T, Petkov P, Dimov S (2006) Laser micro-
milling of ceramics, dielectrics and metals using nanosecond and picosecond lasers. Second
international conference on multi-material micro manufacture, pp 131–134
Kong F, Singh RP (2011) Chemical deterioration and physical instability of foods and beverages.
In: Kilcast D, Subramaniam P (eds) Food and beverage stability and shelf life. Woodhead
Publishing, Cambridge, pp 29–62
Kurosaki S (1954) The dielectric behavior of sorbed water on silica gel. J Phys Chem 58
(4):320–324
Labuza TP, McNally L, Gallagher D, Hawkes J, Hurtado F (1972) Stability of intermediate
moisture foods. 1. Lipid Oxidation. J Food Sci 37(1):154–159
Li XY, Zhao QL, Xu TT, Huang J, Wei LH, Ma Z (2014) Highly ordered microporous polysty-
rene-b-poly(acrylic acid) films: study on the influencing factors in their fabrication via a static
breath-figure method. Eur Polym J 50:135–141
Food Preservation by Nanostructures-Water Interactions Control 25
Liu CX, Lang WZ, Shi BB, Guo YJ (2013) Fabrication of ordered honeycomb porous polyvinyl
chloride (PVC) films by breath figures method. Mater Lett 107:53–55
Ludescher RD, Shah NK, McCaul CP, Simon KV (2001) Beyond Tg: optical luminescence
measurements of molecular mobility in amorphous solid foods. Food Hydrocoll 15
(4–6):331–339
Mandelbrot BB, Riedi RH (1997) Inverse measures, the inversion formula, and discontinuous
multifractals. Adv Appl Math 18(1):50–58
Mara~ non J, Mara~n
on J (2003) Confined water in nanotube. J Mol Struct (THEOCHEM) 623
(1–3):159–166
Nunes RV, Rotstein E (1991) Thermodynamics of the water-foodstuff equilibrium. Drying
Technology: An International Journal 9(1):113–137
Ortony JH, Choi SH, Spruell JM, Hunt JN, Lynd NA, Krogstad DV, Urban VS, Hawker CJ,
Kramer EJ, Han S (2014) Fluidity and water in nanoscale domains define coacervate hydrogels.
Chem Sci 5:58–67
Peponi L, Puglia D, Torre L, Valentini L, Kenny JM (2014) Processing of nanostructured polymers
and advanced polymeric based nanocomposites. Materials Science and Engineering: R:
Reports 85:1–46
Perez J (1994) Theories of liquid-glass transition. J Food Eng 22(1–4):89–114
Rahman MS (2010) Food stability determination by macro–micro region concept in the state
diagram and by defining a critical temperature. J Food Eng 99(4):402–416
Roos Y (1993) Melting and glass transitions of low molecular weight carbohydrates. Carbohydr
Res 238:39–48
Roos YH, Karel M, Kokini JL (1996) Glass transitions in low moisture and frozen foods: effects on
shelf life and quality. Food Technol 50(11):95–108
Ruan R, Long Z, Chen P, Huang V, Almaer S, Taub I (1999) Pulse NMR study of glass transition
in maltodextrin. J Food Sci 64(1):6–9
Salerno M, Landoni P, Verganti R (2008) Designing foresight studies for Nanoscience and
Nanotechnology (NST) future developments. Technol Forecast Soc Chang 75(8):1202–1223
Sealy C (2004) Water on the nanoscale: carbon nanotubes. Materials Today 7(12):12
Slade L, Levine H (1991) Beyond water activity: recent advances based on an alternative approach
to the assessment of food quality and safety. Crit Rev Food Sci Nutr 30(2–3):115–360
Sri Haryani Anwar SH, Kunz B (2011) The influence of drying methods on the stabilization of fish
oil microcapsules: comparison of spray granulation, spray drying, and freeze drying. J Food
Eng 105(2):367–378
Vermeulen A, Marvig CL, Daelman J, Xhaferi R, Nielsen DS, Devlieghere F (2014) Strategies to
increase the stability of intermediate moisture foods towards Zygosaccharomyces rouxii: the
effect of temperature, ethanol, pH and water activity, with or without the influence of organic
acids. Food Microbiol 45:119–125
Wang Y (2014) Nanogeochemistry: nanostructures, emergent properties and their control on
geochemical reactions and mass transfers. Chem Geol 378–379:1–23
Wang F, Li S (1997) Determination of the surface fractal dimension for porous media by capillary
condensation. Ind Eng Chem Res 36(5):1598–1602
Yamamoto T, Endo A, Inagi Y, Ohmori T, Nakaiwa M (2005) Evaluation of thermoporometry for
characterization of mesoporous materials. J Colloid Interface Sci 284(2):614–620
Zeller BL, Saleeb FZ (1996) Production of microporous sugars for adsorption of volatile flavors.
J Food Sci 61(4):749–752
Ying JY, Mehnert CP, Wong MS (1999) Synthesis and applications of supramolecular-templated
mesoporous materials. Angew Chem Int Ed 38(1):56–77
Zimmerman JR, Holmes BG, Lasater JA (1956) A study of adsorbed water on silica gel by nuclear
resonance techniques. J Phys Chem 60(9):1157–1161
Water and Food Appearance
Abbreviations
1 Introduction
Food appearance depends not only on its chromatic characteristics (hue, saturation)
and luminosity but also on the spatial distribution of light inside the food material,
which determines appearance aspects such as transparency and gloss.
The perception of food color is based on visual sensations, which depend on the
spectral distribution of light. This aspect has been widely analyzed in food science.
The modes of appearance produced by different spatial distributions of light
were denoted as “cesia” by the architect Cesar Janello in 1965. An atlas of cesia was
later developed, which found application mainly in relation to paintings, architec-
ture, and ambient illumination materials (Caivano 1991, 1994, 1996; Caivano
et al. 2004).
Because chromatic properties define the color of a given material by means of
three variables (hue, saturation and luminosity), the three variables proposed for
cesia are perceived permeability (or its opposite, opacity), darkness (or its opposite,
lightness or luminosity, which is shared with the color properties), and diffusivity
(or its opposite, regularity), related to the distinctness of image and referring to the
sharpness of images produced by reflection and by transmission (Caivano
et al. 2004).
Permeability ranges from opaque samples to transparent samples, passing
through samples with different degrees of perceived transparency. The variation
of darkness can be perceived by reflectance or transmittance, while the variation of
diffusivity involves perceptions from matte or translucent to glossy. Both reflection
and transmission may occur regularly (specularly) or diffusely, and any intermedi-
ate combination may also appear.
In foods, the defined cesia characteristics represent important physical aspects of
food quality which are perceived by human vision. Nevertheless, these character-
istics have received little attention by food researchers in comparison with chro-
matic attributes, possibly because of the difficulty of easily and adequately
measuring and specifying the stimuli that generate those phenomena.
During processing or storage of foods, many structural changes occur, which can
induce changes in color, gloss, transparency, and luminosity. While stimuli for
color can be produced by primary sources (objects that emit light) or by secondary
sources (objects that reflect or transmit the light coming from another source), the
variations of cesia only occur in secondary sources, that is to say, in objects that
produce changes in the spatial distribution of the light they receive. These changes
are mainly due to microtextural variations on the surface or in the volume of the
object. If these textural variations are of a rather small size, then visual texture
changes are not perceived, while the effect produced on light is noticed (Caivano
et al. 2004).
Translucent materials are those that both transmit and scatter light. The degree
to which an object is translucent depends on the extent to which the light entering
a sample is reflected, scattered, or absorbed. As a phenomenon, translucency
occurs between the extremes of transparency and opaqueness (Hutchings 1999).
Water and Food Appearance 29
The aim of the present work is to analyze the effect of water on changes in the
distribution of light inside food materials and their impact on the perception of the
chromatic properties.
While slices of fresh vegetables look translucent, the same samples behave like
opaque materials after freeze-drying (Agudelo-Laverde et al. 2012, 2013). Also, as
illustrated in Fig. 1a for fresh and dehydrated beetroot, color perception is different
when comparing fresh and dehydrated vegetables; luminosity values are higher and
the values of chromatic variables are lower in the dried vegetables. This effect has
been studied in food matrices such as tomato, kiwi, and strawberries in which these
became of dull appearance after drying (Talens et al. 2002; Lana et al. 2006;
Agudelo-Laverde et al. 2013).
It is interesting to note that the fresh vegetables contain a mass fraction of water
higher than 85 %, and this amount decreases to less than 2 % due to the drying
process, causing a concentration of all the food components, including pigments.
Thus, it would be expected that the dried material would have more intense
chromatic attributes than the fresh material. However, during freeze-drying, the
water present in the inter- and intracellular spaces of the fresh material is replaced
by air, and the samples become opaque due to the presence of air cavities created in
Fig. 1 Effects of water on food appearance: color enlightening on beetroot due to dehydration (a)
and translucence increase due to rehydration of apple discs (b)
30 M.P. Buera et al.
the dried material. Their reflectance measurements and visual appearance do not
reflect the actual pigment concentration. This phenomenon has been widely studied
in the paper industry and is attributed to the diffusion of light inside a material at
interphases of different refraction indexes (Saarela et al. 2008).
Upon rehydration by imbibition, water (of refractive index much closer to that of
the solids than air) occupies part of the pores generated in freeze-drying process
(Lana et al. 2006; Talens et al. 2002), and the samples recover transparency, as
shown in Fig. 1b for the imbibition of freeze-dried apple slices at different times.
Fig. 2 Color enlightening due to increasing concentration of light scattering particles: at constant
pigment concentration (a); effect of sugar crystallization in amorphous hard candies promoted by
increased mobility with increasing relative humidity (b); and opacity changes due to gelatinization
degree (c)
Water and Food Appearance 31
In the same way, hard candies stored above the glass transition temperature
become “whiter” as relative humidity (RH) increases, due to the higher amount of
sugar crystals, which increases with temperature-glass transition temperature
(TTg) thus causing increased light scattering (Fig. 2b).
Starch is a highly structured polymer; it has crystalline as well as amorphous
regions structured in layers inside starch granules. This structure produces the effect
of opacity in corn grits (Fig. 2c, left), which show opacity values close to 1 (calcu-
lation formula is detailed in Sect. 3.3). Samples in Fig. 2c were placed over a scale
of lines with increasing thickness and retro-illuminated. In this way, transparency
could be compared semiquantitatively by observing which lines remained visible
through the sample. Upon gelatinization in the excess of water during cooking of
grits in cornflakes production, the water interacts with glucose units in the starch
molecules, replacing glucose-glucose interchain interactions. Crystalline structure
of starch is lost, the material matrix becomes more homogeneous after cooking, and
therefore opacity decreases (transparency increases), as shown in Fig. 2c, right for
the cooked grits. It is to be noted that simple measurements such as transparency/
opacity, appropriately standardized, may be used to obtain relevant information and
control the cooking process.
3.1 Materials
The chromatic properties and luminosity of samples presented in this work were
obtained by a computer vision system (CVS) composed of a high resolution (10.1
megapixel) digital camera (EOS 40D, Canon Inc., Japan) with an EF-S 60 mm f2.8
macro lens (Canon Inc., Japan) operated through software. The samples were
placed in a standardized diffuse-illumination cabinet painted with Munsell 7 gray
color, using a D65 standard light source. The angle between the camera lens axis
and the lighting source axis was 45 , at which the diffuse reflection responsible for
the color occurs. CIELAB color coordinates (luminosity, L*, red/green a*, and
yellow/blue, b*) were calculated from the images using Adobe Photoshop software
(Yam and Papadakis 2004).
For browning degree determination at the different process stages of cornflakes,
the absorbance at 420 nm was measured in the protease hydrolyzed extracts and the
color CIE tristimulus values X, Y, and Z (D65 light source and 2 observer angle)
were calculated from a transmittance spectrum in a Shimadzu 1620 UV-VIS
spectrophotometer, from which the hue angle (h0 ab) and the browning index
(BI) were obtained as previously described (Buera and Resnik 1990):
x 0:31
BI ¼ ð1Þ
0:172
3.3 Opacity
Opacity degree of samples was calculated as the ratio of L* values when the
samples are measured with black (L*b) and white (L*w) backgrounds (Eq. (2)), at
a given sample thickness. An opacity value of 1 corresponds to an opaque sample,
while lower values correspond to different degrees of translucency. As sample
thickness increases, opacity increases. At a certain thickness, at which the measured
reflectance is independent on the background, the material behaves as opaque.
L*blackbackground
Opacity ¼ ð2Þ
L*whitebackground
3.4 Translucence
The Kubelka–Munk analysis was originally derived in 1931; this analysis found
extensive use in the industry of paints and inks (Džimbeg-Malčić et al. 2011). The
Kubelka–Munk analysis was also applied to food materials in which the color is
Water and Food Appearance 33
affected by the background, such as in the case of translucent samples (Law and
Norris 1973; Budiastra et al. 1998; MacDougall 2002; Talens et al. 2002; Rozycki
et al. 2010).
Reflectance measurements are used to calculate the variables of Kubelka–Munk
theory, as described by Judd and Wyszecki (1975), through the following
equations:
1 R0 R þ Rg
a¼ Rþ ð3Þ
2 R0 Rg
2 1=2
b¼ a 1 ð4Þ
R1 ¼ a b ð5Þ
2
ð 1 R1 Þ
K=S ¼ ð6Þ
2 R1
1 ð 1 a R0 Þ
S ¼ arctgh ð7Þ
b b R0
K ¼ Sð a 1 Þ ð8Þ
4 Translucence Changes
The K and S values were calculated for powdered melon and pear systems at several
RH at which the samples were stored. Figure 3 shows that the scattering coefficient,
S, in both materials decreased drastically at about 30 % RH and above. Since
S represents the amount of scattering particles in the samples, it can be concluded
that the scattering points inside the fruit powders decrease due to hydration, and
transparency consequently increased when water started to fill the pores of the dried
material.
34 M.P. Buera et al.
Fig. 3 Scattering coefficient (S) as a function of relative humidity for melon (a) and pear (b)
powders
Fig. 4 K/S as a function of luminosity (L*) from dehydrated powders of melon (a) and pear (b),
after humidification at relative humidities of 11 (filled square), 22 (filled triangle), 33 ( filled
inverted triangle), 43 (open circle), 52 ( filled circle), 75 (star), and 84 % (open square). The
dotted line indicates the limit between opaque and translucent samples
Besides the opacity changes, the samples also became darker. Figure 4 shows the
relationship between K/S and L* values for powdered samples of melon (a) and pear
(b) rehumidified at several RH. Both variables were related to the degree of
darkening promoted by nonenzymatic reactions. At low RH values (11–22 %
RH), the systems behave as opaque materials (according to the previous discussion
in relation to Fig. 3), and the variation of K/S values is smaller than that for the L*
values. However, with increasing RH the systems are more translucent, and K/S
becomes more sensitive to detect sample darkening than L* values. In this way, the
selected variable to detect color changes sensitively is also dependent on the
hydration degree. The simple measurement of L* is adequately sensitive when
the materials behave as opaque. However, when translucence increases, the K/S
variable from the Kubelka–Munk theory is a better alternative and requires mea-
surements under black and white backgrounds.
Water and Food Appearance 35
Fig. 5 Retro-illuminated images of flaked grits (a) and toasted cornflakes (b). The black lines in
the back are no longer seen in the toasted cornflakes due to the bubbles formation during toasting.
Luminosity (L*) and Opacity (O) values for each samples are also indicated. Images (c) and (d)
correspond to SEM micrographs of flaked and toasted samples, respectively
Figure 5 shows flaked and toasted corn grits obtained from intermediate stages of
cornflake processing. As described previously, transparency can be compared
semiquantitatively by observing which lines remain visible through samples when
they are placed over a scale of lines with increasing thickness and retro-illuminated.
As can be clearly observed, coincident to the measured opacity (also indicated in
Fig. 5), the flaked grits are quite transparent, but opacity increased after the toasting
step. The observation of sample microstructure (through a Philips XL 30 electron
microscope by environmental mode (ESEM) without previous sample treatment)
showed that the toasting process created a porous structure due to sudden water
evaporation, as shown in the scanning electron microscopy (SEM) micrographs in
the left side of Fig. 5. This change in structure had a major effect on appearance and
increased opacity.
36 M.P. Buera et al.
Fig. 6 Transparentization of flaked grits and toasted cornflakes through n matching, by imbibition
in air, water, and polyethyleneglycol (PEG) 400
Table 1 shows the correlation coefficients between the analyzed variables for the
complete process of cornflake production, including all stages. It is to be noted that
those variables derived from surface (reflectance) measurements (h0 ab, L*, and K/S)
correlated between them (shaded region in the upper part of Table 1). On the other
Water and Food Appearance 37
Table 1 Correlation coefficients (r2) for the variables taken from reflectance measurements: L*,
hue (h0 ab), and the ratio between the absorption and scattering Kubelka–Munk coefficients (K/S);
and from aqueous extracts after enzymatic treatment: absorbance at 420 nm (A420) and browning
index (BI), for the entire cornflake production process
L* h0ab K/S A420 BI
L* 1 0.91 0.97* 0.18 0.01
h0 ab 1 0.97* 0.17 0.04
K/S 1 0.09 0.01
A420 1 0.98
BI 1
*Indicates statistically significant correlation p < 0.05
Table 2 Correlation coefficients (r2) for the variables taken from reflectance measurements: L*,
hue (h0 ab), the ratio between the absorption and scattering Kubelka–Munk coefficients (K/S); and
from aqueous extracts after enzymatic treatment: absorbance at 420 nm (A420) and browning index
(BI), for cooking process only
L* h0ab K/S A420 BI
L* 1 0.94* 0.97* 0.82* 0.81*
h0 ab 1 0.94* 0.75* 0.68*
K/S 1 0.77* 0.76*
A420 1 0.97*
BI 1
*Indicates statistically significant correlation p < 0.05
hand, the chemical markers for the Maillard reaction in the aqueous extracts (A420
and BI) were highly correlated (shaded region in the underside of Table 1). How-
ever, due to shape and transparency changes, surface variables did not correlate
with the values obtained from solubilized extracts.
Experiments with model systems are shown in Table 2. This model only
included the cooking step of corn grits, avoiding the dramatic shape and structural
changes produced by flaking and toasting. During cooking of corn grits, slight
changes in shape and surface rugosity take place, and reflectance variables corre-
lated better with those chemical markers obtained after pigment solubilization in
aqueous extracts.
5 Conclusions
The chromatic perception of foods is affected not only by pigment development but
also by changes in optical properties that are caused by supramolecular transitions
(crystallization, gelatinization), formation of bubbles, and other structural changes
affecting sample microstructure, which depend on the degree of hydration.
38 M.P. Buera et al.
6 Future Work
The mathematical development of cesia solids, such as those available for the
chromatic aspects of appearance, will allow better analysis and prediction of the
integrated aspects affecting sample visual characteristics. Such development will
provide tools for innovative product development and for process control and will
necessarily require multidisciplinary interactions, involving concepts from physics,
mathematics, chemistry, engineering, photography, imaging, and architecture.
References
Abbreviations
aw Water activity
IMF Intermediate moisture food
ΔE* Color difference
k Reaction rate constants
r2 Coefficients of determination
C Values of index (color)
t Time
K Relation between the zero-order kinetic constant for the color
formation
k0 Zero-order kinetic constant for the color formation
k1 First-order kinetic constant for the color disappearance
t0 Induction time
ANOVA Analysis of variance
MRPs Maillard reaction products
1 Introduction
and colorless reaction products which range from flavor volatiles (low molecular
weight compounds) to melanoidins, a series of brown pigments with high molecular
weights (Carabasa-Giribet and Ibarz-Ribas 2000; Martins and Van-Boekel 2005);
these effects could be either desirable or undesirable. Browning and the formation
of aroma are desired in baking, roasting, or frying, while it is undesirable in the
foods which have a typical weak or other color of their own such as browning in the
products of condensed milk, white dried soups, tomato soup, etc. and generation of
off-flavors in food during storage. Besides, Maillard reaction can also have negative
effects on nutritional values such as the losses of essential amino acids, as well as
the formation of mutagenic compounds (Belitz et al. 2004).
The rate of Maillard reaction and the nature of the colored products are con-
trolled by its immediate chemical environment including temperature, heating time,
pH, and the composition of the system (nature, concentration, and proportion of the
reactants), as well as water activity (aw). The regulation of all these factors is one of
the ways to control the Maillard reaction progress and hence the browning com-
pounds formed in the system (Ames 1990; Laroque et al. 2008).
Water activity is believed to be one of the important factors in controlling the
Maillard reaction. The chemical reaction is controlled by the effect of moisture
content or aw on reactant dissolution, mobility, and concentration. An increase in aw
to a certain level can increase reactant solubility and mobility and thus the rate of
chemical reaction. However, the reaction rate will decrease when aw increases
further. This is because the moisture content tends to dilute the concentration of
reactants; thus, it limits the availability of mobile water lowering the mobility of the
system and thereby decreases the reaction rate. The maximum browning rate
generally had been observed in most foods at aw 0.65–0.75, and reaction will
cease at aw below 0.3 (Warmbier et al. 1976).
Glycerol is a liquid humectant which contains residual water of less than 0.5 %.
It is naturally present in many foods and is used as food additive, as well as a
conditioner, moisturizer, and solvent for flavor extracts (Cerny and Guntz-Dubini
2006). It also helps to lower the aw in the food; therefore, it increases the microbial
stability of the food and provides longer shelf life to the product especially in
intermediate moisture food (IMF). The glass transition temperature of the pure
glycerol is >80 C (He et al. 2006). Therefore, glycerol has a sufficient molecular
mobility at higher temperatures than 80 C to allow the chemical reactions to
occur.
Maillard reaction has been proved to occur even at low aw due to the presence of
glycerol. This was thought to be an effect of reduced aw and the different
physiochemical environments (Cerny and Guntz-Dubini 2006). This effect was
studied few decades ago (Warmbier et al. 1976; Eichner and Karel 1972; Labuza
and Saltmarch 1981; Mustapha et al. 1998; Sherwin and Labuza 2003). Eichner and
Karel (1972) studied the extent of Maillard browning in a liquid model system of
glycine-glucose in glycerol and found that browning rate increased with decreasing
water activity. Besides, Mustapha et al. (1998) noted high browning in xylose/
lysine in glycerol model system if no water was added, even though the reactants
were not fully solubilized. These authors added that the maximum browning
Maillard Reaction in Limited Moisture and Low Water Activity Environment 43
Chemicals used in this study included hexoses (D-glucose, D-fructose) and a disac-
charide (lactose), amino acid (glycine), and glycerol (99.5 % v/v). All the chemicals
used were of the analytical grade supplied by Ajax Finechem (Sydney, Australia).
The model systems used in the experiment consisted of 13.5 mg of reducing
sugar and 6.68 mg of glycine in 3.5 mL of glycerol at three added moisture levels:
0, 5, and 10 % (v/v). The calculated aw for the three moisture levels was as follows:
~0 (aw ¼ 0.025), 5 (aw ¼ 0.21), and 10 % (aw ¼ 0.36). These aw values were
predicted based on Couchman-Karasz model as illustrated by Katkov and Levine
(2004). All sugars and glycine were dried in vacuum oven at 65 C for 1–2 h in
order to obtain anhydrous condition in the system. The samples were considered to
be close to “zero” % moisture content (actual moisture contents were determined to
be 0.5 %, since pure glycerol contained 0.5 % (v/v) moisture). The residual
moisture in the dried sugars and amino acid can be considered negligible since
the amount of sugars and amino acid used in the experiment was very low.
0.1567 mL and 0.3308 mL of Milli-Q water was added to adjust the moisture
level to 5 and 10 % (v/v) in 3.5 mL glycerol. The samples were prepared in a Pyrex
glass test tube and heated for 1, 2, 3, 4, 10, 17, and 24 h at 100 C. The test tubes
containing samples with ~0 % moisture were uncapped to ensure minimum residual
moisture capture in the product, whereas the samples with 5 and 10 % moisture
44 C.W. Wong et al.
were capped to avoid moisture evaporation. Each sample was mixed with a vortex
mixer for 30 s, before and after heating. At different lengths of heating time up to
24 h, the samples were removed from the oven and cooled immediately to room
temperature before color measurement analyses.
As for control, a series of sample consisting only glycine and glycerol without
reducing sugars were prepared with the same concentration and treated at the same
heating conditions. Systems consisting of each reducing sugar only in glycerol were
also prepared and treated at the same heating temperature and time conditions to
ensure that there is no caramelization of sugars. This can otherwise interfere the
results. This step should not be overlooked. All treatments were performed in
duplicate as per the experimental design described in Sect. 2.4 below.
The data were subjected to regression analysis to determine the reaction order of the
browning reaction. Reaction rate constants (k) and their coefficients of determina-
tion (r2) were calculated. General reaction rate equation was used to evaluate the
kinetic reaction rate order, when:
Maillard Reaction in Limited Moisture and Low Water Activity Environment 45
where C is the values of index (color), n is the reaction order, k is the reaction rate
constant, and t is time. In some cases, the following combined equation (4)
(a pseudo-first-order kinetics) was tested according to Carabasa-Giribet and
Ibarz-Ribas (2000):
where K is the relation between the zero-order kinetic constant for the color formation
(k0) and the first-order kinetic constant for the color disappearance (k1)` and t0 is the
induction time.
The objective of this experiment was to study the effects of different independent
variables, namely, heating time and moisture content on the nonenzymatic brow-
ning reaction, prepared from various sugar-amino acid-glycerol model systems.
The dependent variables (experimental response) that were investigated in this
study were the extent of browning (absorbance) and the intensity of browning
(CIE L*, a*, and b* values). All data was collected from a completely randomized
experiment, and all samples were independent and freshly prepared. Experiments
were performed at least twice.
All experimental data were analyzed using the analysis of variance (ANOVA)
and linear regression data analysis tools in Minitab version 15 and Microsoft Excel
2003, respectively. Whereas the nonlinear regression was done by SigmaPlot 10.0,
ANOVA was performed to find out the effect of moisture and heating time on the
development of browning color in the model systems. The level of statistical
significance for the process variables was defined at p < 0.05. Results reported
were the average of the data.
46 C.W. Wong et al.
With the aim of finding the peak of absorbance vs. wavelength, the absorbance
spectra between 200 and 500 nm were recorded. Simple structure compounds,
produced during the early stages of Maillard reaction, are related to absorbance in
the UV region (Zegota and Bachman 1987). The analysis of the spectra of browning
at different moisture conditions and temperatures used allowed the detection of
changes related to subsequent stages of Maillard reaction. The first stages of the
browning reaction in the systems used are characterized by the appearance of
absorption around 280 nm that can be related to heterocyclic derivatives. This
absorbing material becomes progressively reduced due to the appearance of more
complex compounds absorbing at 350 nm as the reaction proceeds (Cuzzoni
et al. 1998). Wijewickreme et al. (1997) reported that the appearance of absorbance
at 320 nm is due to the formation of soluble pre-melanoidins, while the absorbance
at 420 nm has been used as the end-point measurement for quantifying the yield of
high molecular weight melanoidins (Fogliano et al. 1999).
Maillard Reaction in Limited Moisture and Low Water Activity Environment 47
a
0.150
Abs. (280nm)
0.100
0.050
0.000
0 2 4 6 8 10 12 14 16 18 20 22 24
- 0.050
time (h)
glycine-glycerol (0% moisture)
glycine-glycerol (5% moisture)
glycine-glycerol (10% moisture)
b
1.600
1.200
Abs. (280nm)
0.800
0.400
0.000
0 2 4 6 8 10 12 14 16 18 20 22 24
- 0.400
time (h)
fructose-glycine-glycerol lactose-glycine-glycerol
glucose-glycine-glycerol glycine-glycerol
Fig. 1 (a) Absorbance of control (glycine-glycerol) at three moisture levels monitored at 280 nm.
(b) Comparison of absorbance (280 nm) between sugar-glycine-glycerol model systems with
control (glycine-glycerol) at moisture level 0 %
In this experiment, the absorbance peak was observed along the wavelength
between 282 and 293 nm for glucose and lactose and 278–293 nm for fructose.
Absorbance peaks also occurred between 335 and 344 nm for glucose. In any case,
this range of wavelength has been associated with an early stage of the reaction and
low molecular weight of Maillard reaction product such as furans, pyrroles, and
pyrazines (Lu et al. 2005a). Pyrazine compounds were present abundantly in the
glycerol model system (Lu et al. 2005b). Figure 2a illustrated the trend of absor-
bance spectra of three sugars at 0 % moisture level when heated for 24 h. Obvi-
ously, the absorbance peak of these three sugars was not occurring at the exactly
same wavelength. Absorbance peaks were shifted to a slightly higher wavelength if
moisture was present which is shown in Fig. 2b. This indicated that the brown
pigments produced in each reducing sugar-amino acid-glycerol model system
48 C.W. Wong et al.
a
3
2.5
Absorbance
2 288nm
1.5 278nm
293nm
1
0.5
0
200 300 400 500
wavelength(nm)
glu-gly-glycerol fruc-gly-glycerol lac-gly-glycerol
b 2.5
2
Absorbance
1.5 280nm
0.5
0
200 300 400 500
wavelength (nm)
0% 5% 10%
Fig. 2 (a) Absorbance spectra of glucose-, fructose-, and lactose-glycine-glycerol model system
at 0 % moisture level and heated for 24 h. (b) Absorbance spectra of fructose-glycine-glycerol
model system at three moisture levels and heated for 24 h. All values refer to diluted samples
(5:100 v/v)
All sugars in the glycine-glycerol systems showed the highest browning at moisture
content 0 % (w/w), followed by 5 % (w/w), and 10 % (w/w) showed the least
browning (Fig. 3). The intensity of the browning increased over heating time.
This observation indicated that the role of water, which used to increase the
solubility of the reactant and mobility of the system, thus the rate of the chemical
reaction, could be replaced by using glycerol. This finding once again proved that
glycerol was able to shift the water activity downward (Warmbier et al. 1976;
Eichner and Karel 1972; Mustapha et al. 1998) and hence the water activity for the
maximum browning to occur either due to the hygroscopic nature (Mustapha
et al. 1998) or the plasticizing effect of the glycerol itself (Eichner and Karel
1972). In any case, glycerol was proved to enhance the molecular mobility of the
system and brought the maximum browning to the system at lower water activity.
Water activity, in this case, was not the major factor for browning.
In this study, solubilization of the reactants (sugars and glycine) was visually
observed whether there was sediment at the bottom of the test tubes. High browning
was observed when no water was added to the sugar-glycine-glycerol model system
(0 % moisture level), even though the reactants especially for fructose- and lactose-
glycine were not completely soluble in glycerol during the first few hours heating time.
This finding was in line with the study from Mustapha et al. (1998) who also observed
the same phenomenon in lysine and xylose in the glycerol model system than in an
aqueous matrix. However, all these three sugars were confirmed to be solubilized
completely in the glycerol at 0 % moisture level after a few hours of heating time.
Glucose was solubilized in the glycerol at the first hour heating time at 100 C.
The rate of the Maillard reaction is dependent on the temperature, moisture, and
heating time (Ames 1990; Misharina et al. 1992). The increasing of the browning
reaction was being observed along the increasing of heating time. The browning
reaction decreased as the moisture content in the glycerol model system increased.
50 C.W. Wong et al.
Water, which acts as a strong polar solvent, decreased the browning rates for at least
two reasons: first, a more structured liquid would be formed due to the presence of
water in the solution impeding the collision of reactants by hydrogen bonding and
dipole interaction between water molecules and the reactants or intermediates that
developed in the solution; second, the activation step may be involved in liquidating
electrical charge in the reactive molecules, and this can cause the browning rate to
decrease if the content of the more polar molecule water increased.
The tristimulus coordinates (L*, a*, and b* values) for each samples in the model
system were measured. ANOVA was performed to determine the significant effect
of the factors such as heating time and moisture content on the color changes in
each sugar-glycine in glycerol model system. The measurement of glucose-, lac-
tose-, and fructose-glycine in glycerol model systems showed that moisture content
and heating time significantly affected L*, a*, and b* of the color development in
the system ( p < 0.05) during heating.
A detailed analysis with one-way ANOVA was performed to determine the effect
of each moisture content and heating time on the color changes in the model system.
For lactose-glycine in glycerol model system, there was no significant difference over
the range of moisture content studied in this experiment in L* and a* values
( p > 0.05), only at the first heating hour, while b* value was significantly affected.
Other than that, the effect of heating time on L*, a*, and b* values for all reducing
sugar-glycine in glycerol model system was significant ( p < 0.05).
Here, L* represents the “lightness,” while a* and b* are the color intensity
coordinates where +a* is the red direction, a* is the green direction, +b* is the
yellow direction, and b* is the blue direction (Gu et al. 2006; Ozdemir and Devres
2000). Redness (a*) and yellowness (b*) color parameters for glucose-, fructose-,
and lactose-glycine in glycerol model system were measured and plotted in Fig. 4 at
three moisture levels (a) 0 %, (b) 5 %, and (c) 10 %.
As can be seen in Fig. 4a, at moisture level 0 %, the redness (a*) and yellowness
(b*) values for the three sugar-glycines in glycerol model system were unexpectedly
decreased over time, even though the intensity of the color getting darker (observed
visually) or the L* values (not shown in data) were decreased over time. The trend of
the L* values was in the same direction as a* and b* values’ trend; as a* and b* values
decreased, L* values decreased at the same time. The a* and b* values in lactose-
glycine in glycerol model system increased at the beginning, but the yellowness (b*)
decreased when it reached the maximum point at 2 h heating time and shifted to +a*
value at 3 h heating. Both a* and b* values decreased over time after that. For
fructose-glycine in glycerol model system, b* value reached the maximum point at
1 h heating time, but started to decrease at 2 h heating time. The band was shifted to
+a* or the redness was increased at 2 h heating time, and both a* and b* values
decreased over time after 2 h heating. Glucose-glycine in glycerol model system
decreased over time after both a* and b* values reached the maximum point at the first
Maillard Reaction in Limited Moisture and Low Water Activity Environment 51
a
18.00 b*
16.00
14.00
12.00
10.00
8.00
6.00
4.00 a*
2.00
0.00
-20.00 -15.00 -10.00 -2.000.00
-5.00 5.00 10.00 15.00 20.00
fructose 0% lactose 0%
glucose 0% starting point
b
18.00 b*
16.00
14.00
12.00
10.00
8.00
6.00
4.00
2.00
0.00
-20.00 -15.00 -10.00 -2.000.00
-5.00 5.00 10.00 15.00 20.00
a*
fructose 5% lactose 5% glucose 5%
c
b*
16.00
14.00
12.00
10.00
8.00
6.00
4.00
2.00
0.00
-20.00 -15.00 -10.00 -2.00 0.00
-5.00 5.00 10.00 15.00 20.00
a*
Fig. 4 a* vs. b* plot with the heating time (from 0 to 24 h) at 100 C for fructose-, lactose-, and
glucose-glycine-glycerol system at moisture content (a) 0 %, (b) 5 %, and (c) 10 %
52 C.W. Wong et al.
hour heating time. Generally, the behavior for all these three sugar-glycines in
glycerol model system in a* and b* plot was similar at 0 % moisture content as a*
and b* values decreased over time after they reached the maximum point, respec-
tively, even though marked browning was visually observed.
In contrast, for the system at 5 and 10 % moisture levels, as illustrated in Fig. 4b, c,
yellowness (b*) increased over time, but there were slight inconsistent changes in a*
values, especially for 10 % moisture level model system. It should be noted that the trend
for glucose and fructose model system in b* values was similar to the model system at
0 % moisture level during the end of the process, when the color of the samples began
getting darker (visually observed) or L* values decreased as heating time increased. The
trends for both sugar model systems shifted to +a* at the end of few hours heating time,
and yellowness (b*) was decreased significantly for glucose model system and slightly
decreased for fructose model system. The similar trend was expected for lactose model
system if longer heating time was tested, since the drawn band was shifted slightly to
+a*, and it was postulated that the trend of a* and b* values will decrease after they reach
the maximum point, as the other two sugars behaved.
The decrease of yellowness and redness over time was observed in 0 % moisture
and also in some samples in 5 % moisture model system, especially in glucose
model system, even though strong browning was visually observed. However, it
could be explained as below:
(a) The decrease of lightness (L*) over time was attributed to the less amount of
light scattering. Reactants had reacted in the glycerol and browning formed
over time; hence, it is reasonable to find that browning color of the solution
became stronger as the heating time increased.
(b) The decrease of a* and b* values might be due to the brown color being too
dark to be detected. Referring to Fig. 5 and the picture illustrated in Fig. 6, the
b*
20.00
increasing 15.00
time
10.00 increasing
5.00
0.00
-20.00 -15.00 -10.00 -5.00 0.00 5.00 10.00 15.00 20.00
-5.00 a*
fructose 0% fructose 5%
fructose 10% starting point
Fig. 5 a* vs. b* plot with the heating time (from 0 to 24 h) at 100 C for fructose-glycine-glycerol
model system at moisture content 0, 5, and 10 %. Note: 2 points were overlapping at 17 and 24 h at
moisture level 0 %
Maillard Reaction in Limited Moisture and Low Water Activity Environment 53
b* value reached the maximum point when the sample turned from the
optically transparent color into light yellow brown at heating time 1 h
(Fig. 6 1 h moisture 0 %,). The trend was then shifted to +a* (redness) while
going down to b* direction at the 2 h heating time. Both a* and b* values
tended to decrease from heating time at 3 h. This indicated that the colorimeter
could only be able to detect the formation of yellowish color up to 1 h in
fructose-glycine in glycerol model system at 0 % moisture content (Fig. 6 1 h
moisture 0 %), while formation of redness until 2 h (Fig. 6 2 h moisture 0 %).
(c) For 5 % moisture level, the b* values (yellowness) increased over time up to
17 h. The trend tended to shift to the +a* (redness) after 10 h heating time. The
b* value (yellowness) decreased after 17 h heating time (Fig. 5). The color-
imeter could only detect the formation of yellowish color up to 17 h as shown
in Fig. 6 (17 h, moisture 5 %).
(d) Therefore, it can be concluded that the tolerance of the colorimeter in
detecting yellowish pigment was only up to the color intensity as illustrated
in Fig. 6 (17 h, moisture 5 %; 1 h, moisture 0 %) or lighter. Therefore, it seems
that it is not a problem for the model system consisting of 10 % moisture
content, due to the lightness of the brown color formed in this system.
Sherwin and Labuza (2003) reported that the change of b value showed a
consistent increase over time and correlates well to brown pigment production;
therefore, it was thought to be a most reliable measurement than absorbance
method; however, it is not the case for this experiment. Authors also deemed that
54 C.W. Wong et al.
L and a* values were not useful indicators for brown pigment production, due to the
variability of these parameters which was greater than any consistent increase or
decrease (Sherwin and Labuza 2003). This is true for a* values in this experiment
but not L* values. L* values were showing consistent decrease (data not shown) as
the darkness of the samples increased.
The color difference (ΔE*) of three sugar model systems was plotted in Fig. 7a–c
at three moisture levels. Clearly, the formation of color in these three sugar model
systems was very high at 0 % moisture level. For glucose-glycine model system
(Fig. 7a), the color changed significantly during the first induction time, but after all,
it remained unchanged over the heating time. A similar trend was also observed in
fructose and lactose model systems that increased significantly at the first few hours
of heating time and remained constant after reaching the maximum point at a certain
heating time. At 5 and 10 % moisture levels, the color changes (ΔE*), for glucose and
fructose model system increased steadily over time, but less significant color change
was observed for lactose model system.
Absorbance was measured at 280 and 420 nm, and the results showed that color
development was significantly affected by moisture content and heating time
( p < 0.05). The rate of nonenzymatic browning reaction at three different moisture
levels of the model systems increased with the heating time. As expected, the
browning rate was the highest in 0 % moisture content model system, since the
strong browning was visually observed as illustrated in Fig. 8, followed by 5 % and
then 10 % moisture content model system which gave the least browning rate.
Comparing both absorbances at 280 and 420 nm, the rate obtained at 280 nm was
higher than the one at 420 nm, and this has been reported by many other researchers
(Warmbier et al. 1976; Carabasa-Giribet and Ibarz-Ribas 2000; Miao and Roos
2004) (Fig. 8). This trend was typical for food models with low and intermediate
water contents (Warmbier et al. 1976).
Generally, the nonenzymatic browning reaction in foods follows either first
order (see Eq. (2)) or zero order (see Eq. (1)) (Warmbier et al. 1976; Kamman
and Labuza 1985; Laroque et al. 2008; Miao and Roos 2004). When the change in
intermediate Maillard reaction products (MRPs) yields a linear plot versus reaction
time, it is regarded as zero-order reaction; when the production of MRPs increases
in the course of the Maillard reaction and finally reaches a plateau phase, the
reaction is called first order with respect to intermediate MRPs (Laroque
et al. 2008). However, it is not always possible to apply kinetics as simple as zero
or first order to describe the color changes in Maillard reaction (Ibarz et al. 1999).
Two stage mechanisms in the nonenzymatic browning reaction were proposed by
Ibarz et al. (1999) on the basis of the color deterioration produced in thermal
treatments. The first stage consists of browned polymeric compound formation
which is regarded as a zero-order kinetics, while the second stage is the decompo-
sition of these brown pigments to give colorless compounds (or color destruction)
Maillard Reaction in Limited Moisture and Low Water Activity Environment 55
a
35
30
25
20
ΔE*
15
10
5
0
0 2 4 6 8 10 12 14 16 18 20 22 24
time (h)
glucose 0% glucose 5% glucose 10%
b 35
30
25
20
ΔE*
15
10
5
0
0 2 4 6 8 10 12 14 16 18 20 22 24
time (h)
fructose 0% fructose 5% fructose 10%
c 35
30
25
ΔE* 20
15
10
5
0
0 2 4 6 8 10 12 14 16 18 20 22 24
time (h)
lactose 0% lactose 5% lactose 10%
Fig. 7 The color difference (ΔE*) of three sugar-glycines in glycerol model system, namely, (a)
glucose, (b) fructose, and (c) lactose, at 0, 5, and 10 % moisture content
56 C.W. Wong et al.
1.600 a
0.400
1.200
0.300
Abs. 280nm
Abs. (420nm)
0.800
0.200
0.400
0.100
0.000 0.000
0 2 4 6 8 10 12 14 16 18 20 22 24 0 2 4 6 8 10 12 14 16 18 20 22 24
-0.400 -0.100
time (h) time (h)
glucose 0% glucose 5% glucose 10%
1.600 b
0.300
1.200
Abs. (280nm)
0.200
Abs. (420nm)
0.800
0.400 0.100
0.000 0.000
0 2 4 6 8 10 12 14 16 18 20 22 24 0 2 4 6 8 10 12 14 16 18 20 22 24
-0.400 time (h) -0.100 time (h)
1.600 c 0.300
1.200
0.200
Abs. (420nm)
Abs. (280nm)
0.800
0.100
0.400
0.000 0.000
0 2 4 6 8 10 12 14 16 18 20 22 24 0 2 4 6 8 10 12 14 16 18 20 22 24
-0.400 time (h) time (h)
-0.100
lactose 0% lactose 5% lactose 10%
Fig. 8 Plot of absorbance at 280 nm (a) and 420 nm (b) vs. heating time for glucose-glycine in
glycerol model system at 0, 5, and 10 % moisture level. Data were fitted to the kinetic model of
Eq. (1) (a and b) and Eq. (4) (c). The kinetic prediction data is also shown on each graph
formation color and colorless compounds in the reaction can usually be described as
zero-order kinetics at short heating times, but become complex at longer heating
time and higher temperatures.
In contrast, lactose in 0 % moisture glycerol model system showed a linear
increase over heating time and fitted well to zero-order kinetic model. Plots of
absorbance for the reducing sugar-glycine in glycerol at 5 and 10 % moisture level
model system were also fitted to linear regression and showed that both systems
followed zero-order kinetic reaction. All of these fitted well to zero-order kinetics
probably due to the low initial color formation in the system and also slow
browning rate.
a 1.600
1.200
Abs. (280nm)
0.800
0.400
0.000
0 2 4 6 8 10 12 14 16 18 20 22 24
-0.400
time (h)
fructose 0% lactose 0% glucose 0%
b 0.400
0.300
Abs. (280nm)
0.200
0.100
0.000
0 2 4 6 8 10 12 14 16 18 20 22 24
-0.100
time (h)
c 0.200
0.150
Abs. (280nm)
0.100
0.050
0.000
0 2 4 6 8 10 12 14 16 18 20 22 24
-0.050
time (h)
fructose 10% lactose 10% glucose 10%
Fig. 9 Time course of the absorbance (280 nm) of fructose-, glucose-, lactose-glycine in glycerol
model system at (a) 0 %, (b) 5 %, and (c) 10 % moisture level after heated at 100 C with assumed
pH 7
Maillard Reaction in Limited Moisture and Low Water Activity Environment 59
0.070
Absorbance (420nm) 0.060
0.050
0.040
0.030
0.020
0.010
0.000
-0.010 0 2 4 6 8 10 12 14 16 18 20 22 24
time (h)
fructose 5% lactose 5% glucose 5%
0.025
0.020
Absorbance (420nm)
0.015
0.010
0.005
0.000
0 2 4 6 8 10 12 14 16 18 20 22 24
-0.005
time (h)
fructose 10% lactose 10% glucose 10%
Fig. 10 Comparison of absorbance plot of sugar-glycine in glycerol model system with different
moisture levels at 420 nm
is similar to the plot at 280 nm (Fig. 9c). Lactose still remained the least reactive
sugar, which showed a lower browning rate than the others.
The descending order of the reactivity of reducing sugars,
glucose > fructose > lactose, in the browning development is widely recognized
and supported by many studies. However, there are still major conflicts on the
reactivity of sugars on the degree of browning, as some authors noted fructose is
more reactive than glucose (Benjakul et al. 2005; Brands and van-Boekel 2002)
which was also noted in our study in 5 and 10 % moisture glycerol model systems.
These discrepancies were sometimes explained as higher participation of ketoses to
browning via caramelization compared to aldoses (Jing and Kitts 2004; Ajandouz
et al. 2001). However, this was not the case in the present study. When sugars were
heated alone in glycerol with moisture or without moisture, the solutions were
visually clear, and no color changes were noted, either by assessing the values of
L*, a*, and b* or absorbance, even though the samples were heated at temperature
100 C for 24 h (assumed pH 7). This finding is on the contrary to the study from
Ajandouz et al. (2001) who noted caramelization occurred in aqueous solutions of
60 C.W. Wong et al.
Maillard reaction has proved to occur at low moisture environment, even as low as
0 % moisture content, in all reducing sugar-glycine-glycerol model systems. The
browning rate was found maximum at 0 % moisture model system, followed by
5 %, and then 10 %. Once again, it has been proved that water activity was not one
of the important parameters for Maillard reaction to take place.
In the present study, the browning rate for glucose- and fructose-glycine in
glycerol at 0 % moisture model system obeyed the pseudo-first-order kinetics, a
combined equation of zero-order and first-order reaction which has been reported
previously, while others besides fructose and glucose in 0 % moisture model system
followed to zero-order reaction.
Maillard Reaction in Limited Moisture and Low Water Activity Environment 61
In the case of the reactivity reducing sugars in the early stage of the reaction
(absorbance 280 nm) in the present study, glucose was noted more reactive than
fructose in 0 % moisture model system, but fructose became more reactive than
glucose when the moisture content in the model system increased to 5 and 10 %.
Lactose remained the least reactive reducing sugars. The phenomena have been
discussed widely in the past; however, the reaction pathway that these aldo- and
ketohexoses took place in either aqueous or glycerol model system is unclear. This
needs further investigation.
Maillard reaction is very complex since the rate of nonenzymatic browning was
not only influenced by water activity, but also temperature, heating time, solvent
used in the system, type of reducing sugars, and amino acids. Other parameters
include pH and the presence of transition metal ions which were not included in the
present study. All these parameters will influence the browning rate that needs to be
studied further.
Simple reaction pathway of the Maillard reaction has been established, but the
complex reaction pathway remains unclear. The study of the reaction pathway due
to the discovery of the Maillard reaction occurring in glycerol model system in a
limited moisture environment will pose another big challenge ahead in Maillard
chemistry field.
Since Maillard reaction has proved to occur at low moisture content, with the
high browning rate, it is recommended that the next research studies the profile of
the browning compounds that generated from such a low moisture system. Few
studies had been carried out in the past to investigate the aroma and flavor profile or
mutagenic compounds in sugar-amino acid-glycerol model system at different
water activities and relatively high temperatures. However, more effort should be
put to indentify more volatile and nonvolatile compound, potential antioxidant, as
well as carcinogenic compounds (if any) of the resulting MRPs, by testing different
types of sugars and amino acids at different moisture levels and heating tempera-
tures. By obtaining all these knowledge, Maillard reaction is therefore able to be
controlled, in order to obtain desired color, flavors and aromas, potential antioxi-
dant, and low carcinogenic compounds.
Besides, it is also suggested that the experiments can be conducted in a real food
system by applying the knowledge from the model system. Since Maillard reaction
proved to occur at low moisture content, the success of this research will help to
enhance the shelf life of the intermediate moisture food (IMF) products, and hence,
great contribution could be made in the food industry.
References
Ajandouz EH, Tchiakpe LS, Ore FD, Benajiba A, Puigserver A (2001) Effects of pH on
caramelization and Maillard reaction kinetics in fructose-lysine model systems. J Food Sci
66(7):926–931
Ames JM (1990) Control of the reaction in food systems. Trends Food Sci Technol 1:150–154
62 C.W. Wong et al.
Belitz H D, Grosch W, Schieberle P (2004) Food chemistry, 3rd revised edn. Springer, Berlin
Benjakul S, Lertittikul W, Baurer F (2005) Antioxidant activity of Maillard reaction products from
a porcine plasma protein-sugar model system. Food Chem 93:189–196
Bhandari BR (2010) Beyond water: waterlike functions of other biological compounds in a
waterless system. In: Reid DS, Sajjaanantakul T, Lillford PJ, Charoenrein S (eds) Water
properties in food, health, pharmaceutical and biological systems: ISOPOW 10. Wiley-
Blackwell, Ames, IA, pp 157–164
Brands CMJ, vanBoekel MAJS (2002) Kinetic modeling of reactions in heated monosaccharide-
casein systems. J Agric Food Chem 50:6725–6739
Buera MDP, Chirife J, Resnik SL, Wetzler G (1987) Nonenzymatic browning in liquid model
systems of high water activity: kinetics of color changes due to Maillard’s reaction between
different single sugars and glycine and comparison with caramelization browning. J Food Sci
52(4):1063–1067
Carabasa-Giribet M, Ibarz-Ribas A (2000) Kinetics of colour development in aqueous glucose
systems at high temperatures. J Food Eng 44:181–189
Cerny C, Guntz-Dubini R (2006) Glycerol, another pyrazine precursor in the Maillard reaction.
Recent Advances and Trends, Flavor Science, pp 351–354
Cuzzoni MT, Stoppini G, Gazzani G (1998) Influence of water activity and reaction temperature of
ribose-lysine and glucose-lysine Maillard systems on mutagenicity, absorbance and content of
furfurals. Food Chem Toxicol 26(10):815–822
Eichner K, Karel M (1972) The influence of water content and water activity on the sugar-amino
browning reaction in model systems under various conditions. J Agric Food Chem 20
(2):218–223
Fogliano V, Monti SM, Musella T, Randazzo G, Ritieni A (1999) Formation of coloured Maillard
reaction products in a gluten-glucose model system. Food Chem 66:293–299
Gu YS, Corradini MG, McClemnts DJ, DesRochers J (2006) Properties of low moisture composite
materials consisting of oil droplets dispersed in a protein-carbohydrate-glycerol matrix: effect
of continuous phase composition. J Agric Food Chem 54:417–424
He X, Fowler A, Toner M (2006) Water activity and mobility in solutions of glycerol and small
molecular weight sugars: Implication for cryo- and lyopreservation. J Appl Phys 100
(074702):1–11
Hofmann T (1998) Characterization of the chemical structure of novel colored Maillard reaction
products from furan-2-carboxaldehyde and amino acids. J Agric Food Chem 46:932–940
Ibarz A, Pagan J, Garza S (1999) Kinetic models for color changes in pear puree during heating at
relatively high temperatures. J Food Eng 39:415–422
Jing H, Kitts DD (2004) Antioxidant activity of sugar-lysine Maillard reaction products in cell free
and cell culture systems. Arch Biochem Biophys 429:154–163
Kamman JF, Labuza TP (1985) A comparison of the effect of oil versus plasticized vegetable
shortening on rates of glucose utilization in nonenzymatic browning. J Food Process Preserv
9:217–222
Katkov II, Levine F (2004) Prediction of the glass transition temperature of water solutions:
comparison of different models. Cryobiology 49:62–82
Labuza TP, Saltmarch M (1981) The nonenzymatic browning reaction as affected by water in
foods. In: Rockland LB, Stewart GF (eds) Water activity: influences on food quality. Academic
Press, Inc., New York, pp 605–650
Laroque D, Inisan C, Berger C, Vouland E, Dufosse L (2008) Kinetic study on the Maillard
reaction. Consideration of sugar reactivity. Food Chem 111:1032–1042
Lu CY, Hao Z, Payne R, Ho CT (2005a) Effects of water content on volatile generation and
peptide degradation in the Maillard reaction of glycine, diglycine, and triglycine. J Agric Food
Chem 53:6443–6447
Lu CY, Payne R, Ho CT (2005b) The effect of water content on volatile formation from the
Maillard reaction of peptides and glucose. Ann N Y Acad Sci 1043:894
Maillard Reaction in Limited Moisture and Low Water Activity Environment 63
Martins SIFS, Van-Boekel MAJS (2005) A kinetic model for glucose/glycine Maillard reaction
pathways. Food Chem 90:257–269
Mauron J (1990) Influence of processing on protein quality. J Nutr Sci Vitaminol 36:557–569
Miao S, Roos H (2004) Nonenzymatic browning kinetics of a carbohydrate-based low-moisture
food system at temperatures applicable to spray drying. J Agric Food Chem 52(16):5250–5257
Misharina TA, Golovnya RV, Yakovleva VN (1992) Effect of time and temperature on the
preparation of pyrazines in model reactions of the synthesis of aroma-forming substances.
Russ Chem Bull 41(7):1258–1263
Mustapha WAW, Hill SE, Blanshard JMV, Derbyshire W (1998) Maillard reactions: do the
properties of liquid matrices matter? Food Chem 62(4):441–449
Naranjo GB, Malec LS, Vigo MS (1998) Reducing sugars effect on available lysine loss of casein
by moderate heat treatment. Food Chem 62(3):309–313
Obanu ZA, Ledward DA, Lawrie RA (1977) Reactivity of glycerol in intermediate moisture
meats. Meat Sci 1:177–183
Oliver CM, Melton LD, Stanley RA (2006) Glycation of caseinate by Fructose and fructo-
oligosaccharides during controlled heat treatment in the ‘dry’ state. J Sci Food Agric
86:722–731
Ozdemir M, Devres O (2000) Kinetics of color changes of hazelnuts during roasting. J Food Eng
44:31–38
Schwartzberg HG, Hartel RW (1992) Physical chemistry of foods. Marcel Dekker, New York
Seow CC, Cheah PB (1985) Reactivity of sorbic acid and glycerol in non-enzymatic browning in
liquid intermediate moisture model systems. Food Chem 18:71–80
Sherwin CP, Labuza TP (2003) Role of moisture in Maillard browning reaction rate in interme-
diate moisture foods: comparing solvent phase and matrix properties. J Food Sci 68:588–594
Suarez G, Rajaram R, Bhuyan KC, Oronsky AL, Goidl JA (1988) Administration of an aldose
reductase inhibitor induces a decrease of collagen fluorescence in diabetic rats. J Clin Investig
82:624–627
Warmbier HC, Schnickels RA, Labuza TP (1976) Effect of glycerol on non-enzymatic browning
in a solid intermediate moisture model food system. J Food Sci 41(3):528–531
Wijewickreme AN, Kitts DD (1997) Influence of reaction conditions on the oxidative behavior of
model Maillard reaction products. J Agric Food Chem 45:4571–4576
Wijewickreme AN, Kitts DD, Durance TD (1997) Reaction conditions influence the elementary
composition and metal chelating affinity of nondialyzable model Maillard reaction products.
J Agric Food Chem 45:4577–4583
Zegota A, Bachman S (1987) Nonenzymatic browning induced by gamma-irradiation in model
system. Z Lebensmittelunters Forsch 184:3–7
Carbohydrates and Proteins
as Nonequilibrium Components
of Biological Materials
Y.H. Roos
Abbreviations
aw Water activity
C1 Universal coefficient
C2 Universal coefficient
T Temperature
tcr Crystallization time
Tg Glass transition temperature
WLF Williams-Landel-Ferry relationship
1 Introduction
Carbohydrates and proteins form the main hydrophilic phases in biological mate-
rials. Water solubility and formation of glass-like structures at high solids contents
are well-known properties of carbohydrates, particularly of low molecular weight
sugars (White and Cakebread 1966). Natural biological materials, including foods,
often contain carbohydrates mixed with proteins and other hydrophilic substances
in their aqueous phase.
Food materials can be preserved by using dehydration or freezing processes that
reduce water content or the effective concentration of water (water activity, aw) for
the support of biological functions, as microbial growth. Such processes remove
water from hydrophilic solids and produce concentrated solids systems that
Dielectric relaxation times derived from the α-relaxation have been reported for
several carbohydrates (Noel et al. 1996; Shinyashiki et al. 2008). The comprehen-
sive data of Shinyashiki et al. (2008) for noncrystalline fructose at various water
contents were particularly interesting. Our analysis of their data showed that the
structural relaxation times decreased in line with the Williams-Landel-Ferry (WLF)
relationship (Williams et al. 1955) with increasing temperature above Tg. Structural
relaxation times at constant water content decreased, showing typical water plasti-
cization of hydrophilic food solids. This finding was also apparent in an “Angell”
plot showing relaxation times against Tg/T (Angell et al. 1994; Angell 2002). Water
Carbohydrates and Proteins as Nonequilibrium Components of Biological Materials 67
a 10
40°C
45°C 50°C
55°C
8
Dielectric Loss, ε’’
60°C
6
0
0.01 0.1 1 10 100
30°C 35°C
Frequency (kHz)
b7
55°C
6 60°C
65°C 70°C 75°C
Dielectric Loss, ε’’
2
1
0
0.01 0.1 1 10 50°C 100
Frequency (kHz) 80°C
Fig. 1 Dielectric loss, ε00 , for amorphous fructose (a) and glucose (b) at various temperatures and
frequencies
68 Y.H. Roos
0.015 2 τmax=100 s
Fructose Fructose
Glucose Glucose
0
0.010
High Practical -2
Importance
High Practical
0.005 -4 Importance
-6
0
0 10 20 30 40 50 60 0.84 0.86 0.88 0.90 0.92 0.94 0.96 0.98 1.00
T - Tg (K) Tg / T
Fig. 2 Temperature dependence of dielectric relaxation times of amorphous fructose and glucose
above glass transition as a function of T Tg and Tg/T (Angell plot). The difference in Tg/T plotting
is due to the differences in glass transition temperatures (fructose Tg ¼ 282 K; glucose Tg ¼ 304 K)
systems for their analogous dielectric and mechanical relaxation properties over the
range from Tg to Tg + 30 C at minimum. It should be noted that the WLF
relationship shows differences in relaxation times as a function of temperature
above the calorimetric onset Tg (i.e., T Tg), while the Tg/T scaling of Angell
plots cannot be used to compare materials with different Tg values. Consequently, a
comparison of relaxation times of the same material with different water plastici-
zation levels using an Angell plot is not appropriate and may give highly misleading
information on the effects of plasticization on fragility and relaxation behavior.
Accordingly, fragility values, as defined by Angell (e.g., Angell 2002), cannot be
used to assess relaxation properties of glass formers above their Tg.
Numerous studies have shown that the mechanical and dielectric behavior of
simple low molecular weight sugars is typical of nonequilibrium liquids (Noel
et al. 1996; Silalai and Roos 2011; Potes et al. 2012). Such sugars show dramatic
changes in modulus and dielectric behavior around the glass transition. When
carbohydrates and sugars are dissolved in water with suspended proteins, such as
dairy solids or soy proteins, the glass transition behavior of the systems seems to
follow that of the carbohydrate components (Zhou and Roos 2011). For a phase-
separated system, the Tg when plotted against aw is composition independent, and
the data overlap that of the component suffering the glass transition. When the
carbohydrate composition is modified by alteration of the low and high molecular
weight carbohydrate fractions, it may be noted that glass transition occurs at a
higher temperature with increasing average molecular weight (Silalai and Roos
2011; Potes et al. 2012). Water plasticization studies of such systems need to
consider whether multiple glass formers are plasticized as a mixture of miscible
components (e.g., high DE maltodextrins with sugars) or do components, such as
proteins and sugars, exhibit phase separation. Phase separation of food components
may also result in significant differences in dielectric and mechanical relaxation
properties. Dielectric properties may be dominated by relaxation properties of small
molecular weight carbohydrates and their water plasticization. Conversely,
Carbohydrates and Proteins as Nonequilibrium Components of Biological Materials 69
3 Fluidness Characteristics
2
SiO2 C2 ~ 1500 K
0
s
es
-2
idn
log τ (S)
Flu
-4 Glucose
C2 ~ 50 K
-6 Glycerol
Fig. 3 Fluidness of common glass formers in foods shown by a plot of the structural relaxation
time (log τ) against temperature above the glass transition onset temperature (T Tg). The C2
value of the WLF relationship shown is a measure of fluidness, i.e., fluidness decreases with
increasing C2 value
70 Y.H. Roos
relaxation times above the Tg. It should be noted, however, that significant differences
in measured relaxation times of complex food systems may result from the phase
separation of components and differences in their molecular weight (Silalai and Roos
2011; Zhou and Roos 2011; Potes et al. 2012). The WLF constant C2 describes the
steepness of the logarithmic decrease in structural relaxation times with increasing
T Tg (Fig. 3). The C2 value provides a measure of “fluidness” of food components.
Differences in fluidness indicate variations in molecular mobility of different glass
formers above the glass transition and give a clear measure for comparison of effects
of glass transition on mobility-related properties.
Structural relaxation times and their variations during food processing are signifi-
cantly dependent on dynamic changes in water content. Typical food processes
where structural relaxation times control solids mobility and concomitant formation
of microstructure include food freezing, dehydration processes, baking, extrusion,
and manufacturing of confectionary products. The resultant materials are typically
noncrystalline solids showing glass transitions for at least some of the components
or mixtures of miscible components.
Formation of food microstructure relates to fluidness of solids through the rate at
which food solids can relax during dehydration to respond to the loss of volume
resulting from evaporation of water (Karel et al. 1994). In natural foods, several
factors, including the characteristics of cell wall polymers, further contribute to the
overall shrinkage of complex food solids. Freeze-drying is a typical process where
microstructural changes during sublimation are minimal. Volume retention in
freeze-drying benefits from glass formation of solids as a result of freeze concen-
tration during the freezing process and concomitant formation of a separate ice
phase (Harnkarnsujarit et al. 2012). A successful freeze-drying process retains the
highly freeze-concentrated solids network that can support its own weight during
sublimation of ice. A glassy state or high enough viscosity of solids is required to
resist the gravitational stress appearing as sublimation increases void volume
(porosity). The freeze-concentrated solids remain as a membrane network, hosting
pores left by the ice crystals. In other dehydration processes, some water plastici-
zation of solids enhances viscous flow, i.e., the microstructure of the dehydrated
solids is formed during evaporation of water. It may be assumed that the fluidness
characteristics of food solids and their level of water plasticization during formation
of microstructures in food processing are significant factors affecting the final
properties of the materials, particularly shrinkage in dehydration and formation of
expanded solids in extrusion or puffing.
Carbohydrates and Proteins as Nonequilibrium Components of Biological Materials 71
6 16
4 14
2 ~10 s 12
log Viscosit (Pa s)
0 Contact Time ~1 s 10
log t (s)
-2 8
Relaxation Time Viscosity
-4 6
-6 4
-8 2
0 10 20 30 40 50
T - Tg (°C)
Fig. 4 Relationships of structural relaxation time, particle contact time, and viscosity according to
data of Downton et al. (1982) and the WLF relationship using the universal constants C1 ¼ 17.44
and C2 ¼ 51.6 K
72 Y.H. Roos
from Downton et al. (1982). The relationships established in Fig. 4 showed that the
contact times of 1–10 s corresponded to structural relaxation times of 104 s to
102 s at 16–27 C above the Tg. It is important to note that the Tg values used are
onset temperatures of the calorimetric glass transitions at which the structural
relaxation times are approximately 100 s (Angell 2002).
8 8
2 months
2 weeks Time to complete
6 2 days crystallization 6
2 hours
4 20 min 4
log τcr (S)
10 min
log τ (S)
2 min
2 2
3s
1s Structural
0 0.5 s relaxation time 0
10 ms
-2 3 ms -2
1 ms
0.3 ms
-4 -4
0 10 20 30 40 50
T - Tg
5 Conclusions
References
Angell CA (2002) Liquid fragility and the glass transition in water and aqueous solutions. Chem
Rev 102:2627–2650
Angell CA, Bressel RD, Green JL, Kanno H, Oguni M, Sare EJ (1994) Liquid fragility and the
glass transition in water and aqueous solutions. J Food Eng 22:115–142
Downton GE, Flores-Luna JL, King CJ (1982) Mechanism of stickiness in hygroscopic, amor-
phous powders. Ind Eng Chem Fundam 21:447–451
Haque MK, Ross YH (2004) Water plasticization and crystallization of lactose in spray-dried
lactose/protein mixtures. J Food Sci 69(1):FEP23–FEP29
Harnkarnsujarit N, Charoenrein S, Roos YH (2012) Microstructure formation of maltodextrin and
sugar matrices in freeze-dried systems. Carbohydr Polym 88:734–742
Karel M, Anglea S, Buera P, Karmas R, Levi G, Roos Y (1994) Stability-related transitions of
amorphous foods. Thermochim Acta 246:249–269
Le Meste M, Champion D, Roudaut G, Blond G, Simatos D (2002) Glass transition and food
technology: a critical appraisal. J Food Sci 67:2444–2458
Noel TR, Parker R, Ring SG (1996) A comparative study of the dielectric relaxation behaviour of
glucose, maltose, and their mixtures with water in the liquid and glassy states. Carbohydr Res
282:193–206
Potes N, Kerry JP, Roos YH (2012) Additivity of water sorption, alpha-relaxations and crystalli-
zation inhibition in lactose-maltodextrin systems. Carbohydr Polym 89(4):1050–1059
Roos Y (1993) Melting and glass transitions of low molecular weight carbohydrates. Carbohydr
Res 238:39–48
Roos Y, Karel M (1990) Differential scanning calorimetry study of phase transitions affecting the
quality of dehydrated materials. Biotechnol Prog 6:159–163
Roos Y, Karel M (1991) Plasticizing effect of water on thermal behavior and crystallization of
amorphous food models. J Food Sci 56:38–43
Roos Y, Karel M (1992) Crystallization of amorphous lactose. J Food Sci 57:775–777
Roudaut G, Farhat I, Poirier-Brulez F, Champion D (2009) Influence of water, temperature and
sucrose on dynamics in glassy starch-based products studied by low field 1H NMR. Carbohydr
Polym 77:489–495
Sablani SS, Kasapis S, Rahman MS (2007) Evaluating water activity and glass transition concepts
for food stability. J Food Eng 78:266–271
Shinyashiki N, Shinohara M, Iwata Y, Goto T, Oyama M, Suzuki S, Yamamoto W, Yagihara S,
Inoue T, Oyaizu S, Yamamoto S, Ngai KL, Capaccioli S (2008) The glass transition and
dielectric secondary relaxation of fructose-water mixtures. J Phys Chem B 112:15470–15477
Silalai N, Roos YH (2011) Mechanical relaxation times as indicators of stickiness in skim milk-
maltodextrin solids systems. J Food Eng 106:306–317
Slade L, Levine H (1991) Beyond water activity: recent advances based on an alternative approach
to the assessment of food quality and safety. Crit Rev Food Sci Nutr 30:115–360
Soesanto T, Williams MC (1981) Volumetric interpretation of viscosity for concentrated and
dilute sugar solutions. J Phys Chem 85:3338–3341
Toufeili I, Lambert IA, Kokini JL (2002) Effect of glass transition and cross-linking on rheological
properties of gluten: development of a preliminary state diagram. Cereal Chem 79:138–142
White GW, Cakebread SH (1966) The glassy state in certain sugar-containing food products. J
Food Technol 1:73–82
Williams ML, Landel RF, Ferry JD (1955) The temperature dependence of relaxation mechanisms
in amorphous polymers and other glass-forming liquids. J Am Chem Soc 77:3701–3707
Zhou Y, Roos YH (2011) Characterization of carbohydrate-protein matrices for nutrient delivery.
J Food Sci 76:E368–E376
Low-Temperature Mobility of Water
in Sugar Glasses: Insights from Thermally
Stimulated Current Study
Abbreviations
D 2O Water
DSC Differential scanning calorimetry
HB Hydrogen bonding
LDA Low-density amorphous solid water
NMR Nuclear magnetic resonance
P1 Calorimetric peak
Px Calorimetric peak
RH Relative humidity
T1 Spin-lattice relaxation times
Tg Glass transition temperature
Tgw Glass transition of water in sugar matrix
S. Ewing
Pfizer Inc, Groton, CT, USA
A. Hussain
Pfizer Inc, Groton, CT, USA
New Jersey Institute of Technology, Newark, NJ, USA
G. Collins
New Jersey Institute of Technology, Newark, NJ, USA
C. Roberts
Department of Chemical Engineering, University of Delaware, Newark, DE 19716, USA
E. Shalaev (*)
Pfizer Inc, Groton, CT, USA
Allergan Inc, Irvine, CA, USA
e-mail: [email protected]
Tp Polarizing temperature
TSC Thermally stimulated current
TSdC Thermally stimulated depolarization current
TSpC Thermally stimulated polarization current
1 Introduction
In this study, we used the thermally stimulated current (TSC) method to explore
if the decoupling in mobility between sucrose and water molecules, which was
reported by Girlich and Lüdemann (1994), persists in the wide temperature range,
in particular below the calorimetric glass transition temperature. In addition, we
probed whether similar behavior can be expected in other polyhydroxy compounds
by using sorbitol, a sugar alcohol. TSC is a thermal analytical technique that has
been used to study dipole relaxation in different types of materials, including
synthetic polymers, small molecular-weight glass-formers, and proteins (Teyssedre
et al. 1995; Fagegaltier et al. 1997; Correia et al. 2000; Galop and Collins 2001;
Shmeis et al. 2004; Reddy et al. 2009). In addition, TSC has been used to study
low-temperature molecular mobility in aqueous sugar solutions (Pissis et al. 1983)
and hexagonal ice (Johari and Jones 1975; Apekis et al. 1983). A brief description
of the TSC method is provided in Sect. 2. In the present study, we investigated
highly concentrated sucrose-water and sorbitol-water solutions and freeze-dried
amorphous sucrose, i.e., systems in which no ice is formed upon cooling. Ice
formation would result in two-phase systems, with water molecules distributed
between crystalline ice (usually, hexagonal) phase and an amorphous phase
consisting of the solute and remaining water. With such a two-phase system, the
experimental TSC signal would have contributions from both water molecules in
ice and water in the amorphous phase, which could complicate interpretation of the
results.
2.1 Materials
acetate solution (23 % RH). Vacuum was pulled to evacuate the desiccators, and the
freeze-dried samples were allowed to equilibrate for at least 4 days. To estimate
water content in these samples, the Tg was measured by DSC, and the Tg vs. water
content curve from Girlich and Lüdemann (1994), reproduced in Fig. 4, was used to
determine corresponding water content. Water contents were determined to be
1.6 % w/w (range of 0.5–2.8 % w/w), 3.9 % w/w (range of 3.3–4.4 % w/w), and
6.8 % w/w for the freeze-dried sample and after equilibration at 11 % and 23 % RH,
respectively, based on the Tg values (midpoint) of 61.5 9 C, 44.5 4 C, and
22.5 C, respectively.
Thermally stimulated current detects the mobility of dipoles and other permanent or
induced electric charges. There are two basic TSC techniques, namely thermally
stimulated polarization current (TSpC) and thermally stimulated depolarization
current (TSdC) (Vanderschueren and Gasiot 1979). In the depolarization experi-
ments (TSdC), a sample is polarized by a permanent electric field at a relatively
high temperature where there is significant molecular mobility to allow the attain-
ment of equilibrium. The polarization is frozen in by quenching the sample to a low
temperature, and then the electric field is removed. Depolarization is measured
during heating of such a sample without an electric field. When the dipoles become
mobile, an electric current is induced by the spontaneous depolarization, and this
current is measured by an electrometer. In the polarization method (TSpC), an
electric field is applied at low temperature, where dipoles are immobile and
polarization of dipoles by the electric field is detected during heating of the sample.
A potential advantage of the TSpC method is that the sample is not exposed to
higher temperature prior to the measurements (as with TSdC experiments), thus
allowing characterization of the sample with “as is” thermal history and minimizing
the potential for both chemical degradation and physical changes (e.g., crystalliza-
tion) of temperature-sensitive materials. In addition, TSpC experiments are often
performed prior to TSdC measurements on an unknown sample, in order to choose
polarization temperature.
Thermally stimulated current experiments were performed with a TSC/RMA
instrument (Thermold, CT) equipped with stainless steel electrodes. The TSC
instrument was calibrated using melting of indium at the heating rate 7 C/min.
Two different sample preparation methods were used. In the first method, a sample
holder consisting of a differential scanning calorimeter (DSC) aluminum pan, a
Teflon sheet, a Teflon ring that serves as a spacer, and a Teflon lid was used. Powder
samples and solutions were loaded into the small cavity formed by a Teflon ring on
top of a Teflon sheet that was fitted into an aluminum Perkin-Elmer DSC pan. The
samples were compacted into the cavity and then covered with a Teflon lid that was
crimped onto the pan containing the sample. The Teflon lid was used to insulate the
electrode from charge injection at the sample surface and to minimize contributions
Low-Temperature Mobility of Water in Sugar Glasses: Insights from Thermally. . . 79
from the ionic conductivity of the sample to the TSC signal. This approach is
known as a blocking electrode. In the second method, a solution was sandwiched
between two Teflon sheets in the cavity of the Teflon ring placed. Both methods
gave comparable results.
In TSpC experiments, the sample was cooled to approximately 170 C, and
then a polarizing static field of 100 V was applied during heating at 7 C/min. In
several experiments, higher voltages of 200 and 300 V were used. Polarization
current was measured while heating in the presence of the applied field. TSdC
experiments were performed by first heating the sample to a polarizing temperature
(Tp), and then holding at Tp isothermally in the presence of a polarizing field for
2 min. Polarization temperature of 50 C was used for 70 and 80 % sorbitol, and
20 C for 90 % sorbitol. Typically, a polarization voltage of 100 V was used;
several experiments were performed at a higher voltage of 200 and 300 V. The
sample was then cooled using liquid nitrogen in the presence of the applied field to
sub-zero temperature and held for 1 min. The polarizing field was then removed,
and the depolarization current was measured while heating the sample at 7 C/min.
For the sorbitol system, both polarization (TSpC) and depolarization (TSdC)
experiments were performed, while the sucrose system was investigated using
TSpC mode. Temperature of the TSC events was measured manually from a
position of a peak on a TSC curve.
3 Results
Representative TSpC heating curves of 73.1 % w/w sucrose solution are shown in
Fig. 1. The experiments were performed at 100 V, 200 V, and 300 V, respectively,
using both positive and negative voltage. Three broad signals, marked as P1, P2,
and P3, were observed. At lower temperatures below approximately 60 C, the
negative and positive curves represent nominally mirror images, whereas above
60 C significant difference between negative and positive curves was observed.
In addition, the lower-temperature portions of the curves scaled with the field. This
observation, i.e., proportionality with the field and mirror image of P1 and P2
events, is indicative of dipole relaxation process. On the other hand, the nature of
higher temperature events is less certain and probably involves additional
80 S. Ewing et al.
7e-13
-300V
-200V
Current, Amps
-100V
0 0V
100V
200V
300V
-7e-13
P1 P2 P3
Fig. 1 TSpC heating curves for 73.12 % sucrose solution in water obtained at variable polariza-
tion voltages (100, 200, 300, and 0 V, as shown in the graph) and with two different directions of
the polarizing field
0.0
-5.0x10-13 0
Current, Amps
Px
Current, Amps
100 V -2*10-14
100 V
-1.0x10-12 200 V
P1 300 V
300 V
Temperature, °C
Fig. 2 Representative TSpC heating curves of freeze-dried sucrose. Insert: magnified part of
TSpC curves showing the low-temperature P1 event at different voltages of the polarizing field.
Area of magnification is shown on the main curve as a rectangle
Representative DSC heating curves for sorbitol samples with variable concen-
tration of water are presented in Fig. 3a. In all samples, glass transition was
detected, whereas no evidence of water crystallization was observed. In solutions
with the lowest concentration of sorbitol (65 % w/w), during heating scan to higher
temperatures, i.e., above the Tg, it was observed the exotherm of water crystalli-
zation and ice melting endotherm. This finding is consistent with low-temperature
X-ray diffraction studies of concentrated solutions of sorbitol where neither water
nor sorbitol crystallization was observed on the timescale of the experiment (data
not shown). Therefore, these samples are expected to remain completely amor-
phous during TSC experiments. Typical TSpC heating curves for sorbitol glasses
with water content from 35 to 0 % w/w are shown in Fig. 3b. The low-temperature
relaxation event, P1, was observed in all sorbitol-water samples, whereas it was not
detected in amorphous sorbitol without water. Note that a relatively weak and broad
low-temperature event attributed to beta-relaxation (Johari-Goldstein relaxation)
was observed in pure sorbitol using the TSDC method (Correia et al. 2001; Ramos
et al. 2007). The apparent discrepancy between the literature, where a
low-temperature event was detected in pure sorbitol, and the present study, with a
lack of such events in pure sorbitol, might be attributed to different sample
configurations used. Indeed, the sample was in direct contact with an electrode in
the study of Ramos et al. (2007), whereas in the present study, the sample was
isolated from the electrode by Teflon. Isolating the sample from the electrode could
have an impact on the signal strength, and therefore on the detection of such low
strength signals as beta-relaxation. The P1 event was observed in both TSpC and
TSdC experiments for sorbitol-water samples, and its strength was directly
82 S. Ewing et al.
Fig. 3 Representative DSC (a) and TSpC (b) curves of sorbitol-water samples with different
concentration. Concentrations are shown in % w/w
proportional to the voltage of the polarizing field (curves not shown). These
observations, i.e., the direct proportionality between the voltage and the signal,
and the mirror images produced with TSpC and TSdC are usually considered to be
the “signature” of a dipole relaxation. Several higher temperature TSC events were
also observed in the sorbitol-water system; investigation of these transitions is
outside the scope of the present study and is not considered here. Temperatures
of the P1 event increased with the increase in sorbitol concentration, as can be seen
in Fig. 4. P1 values are plotted in Fig. 4, along with the calorimetric Tg; also, Tg
data from the literature are shown for comparison.
Low-Temperature Mobility of Water in Sugar Glasses: Insights from Thermally. . . 83
50
se
Temperature (°C)
cro
su
-50
Tg
l
bito
sor
Tg
-100
e
cros a, d
su b
-150 T0 c
T0 water
lh
0 20 40 60 80 100
Solute (Sucrose or sorbitol) (wt%)
Fig. 4 Calorimetric Tg and TSC events in sucrose-water: Solid black line represents the Tg curve
of sucrose-water (Girlich and Lüdemann 1994). Blue line and blue symbols represent the sorbitol-
water systems. Dotted line is the T0 values for water-sucrose system according to Girlich and
Lüdemann (1994). The Tg line for sorbitol was constructed by fitting the DSC Tg (onset) data from
Talja and Roos (2001) (symbols not shown) and the present study (blue filled squares) to the
Gordon-Taylor equation (Gordon and Taylor 1952) with K as a fitting parameter Tg ¼
(w1Tg1 + Kw2Tg2)/(w1 + Kw2), where Tg1 and Tg2 ¼ 136 K are the glass transition temperatures
and w1 and w2 are the weight fractions of sorbitol and water, respectively, and K is a constant.
K was determined to be 3.1 0.2. The arrows indicate temperature of the Tg for pure water
(137 C for hyperquenched glassy water (Johari et al. 1987), 144 C for the produced LDA
(Mayer 1991), 149 C for LDA (Handa and Klug 1988), 138 C for vapor-condensate
(Sugisaki et al. 1968)) and for the rotational mobility of water in hexagonal ice (Johari and
Jones 1975). Schematic representation of the microscopic structure of a sorbitol-water system is
presented at the top, with three regions of different water concentration. Water and sorbitol
molecules are shown as blue and yellow objects, respectively. The Figure was published previ-
ously in Authelin et al. (2014)
4 Discussion
A low-temperature TSC event, P1, was detected in both sucrose-water and sorbitol-
water glasses at approximately 153 to 128 C (sucrose-water) and 150 to
125 C (sorbitol-water). While there are several potential types of physical
processes which could cause a TSC signal (such as space charge polarization),
84 S. Ewing et al.
the P1 event indicates all of the main signatures of dipole relaxation, i.e., linear
response of the strength with voltage, and mirror image of signals obtained with +/
directions of the polarizing field (sucrose) and TSpC/TSdC (sorbitol). In addition,
the fact that the P1 event was not detected in pure sorbitol (i.e., at 0 % water), and
that the intensity of the signal was much lower in low water samples as compared
with higher water samples in both sucrose-water and sorbitol-water systems, allows
us to conclude that the origin of a P1 event is due to dielectric relaxation of water
dipoles, probably because of the rotational mobility of water molecules, and that
the temperature of the P1 event corresponds to the onset of dipole relaxation of water
in sorbitol and sucrose glasses.
The concentration-related changes in P1 temperature were similar in both
sucrose-water and sorbitol-water systems; this observation can be attributed to
either similarity in water-sucrose and water-sorbitol interaction, or negligible
water-solute interaction as related to that of water-water contacts. The first expla-
nation (similar water-sucrose and water-sorbitol interactions) seems more likely, as
the P1 temperature changed with the water content, whereas a lack of concentration
dependence could be expected under the second scenario (i.e., predominantly
water-water interactions).
It is interesting to compare the P1 temperature in sucrose and sorbitol glasses
with the glass transition temperature reported for pure amorphous water. Indeed,
from a general perspective, the glass transition event corresponds to the onset of
rotational and translational cooperative movements (during heating), and therefore
the P1 event may resemble the glass transition behavior. To highlight this resem-
blance, we attribute the P1 event to the glass transition of water in sugar matrix
(Tgw). In low-density amorphous solid water (LDA), the glass transition temper-
ature is reported to lie in the range of 150 to 135 C, depending on the sample
preparation and measurement technique (Sugisaki et al. 1968; Johari et al. 1987;
Handa and Klug 1988; Mayer 1991). Interestingly, using 150 C as a lower
boundary indicates that Tgw decreases with increasing water content and has
reached the 150 C boundary by the point of 40 % w/w water (see Fig. 4). A
plausible interpretation of these observations is that Tgw measured by TSC for
lower water contents reflects the average orientational mobility of water molecules
within an effectively rigid matrix of sugar molecules that form a percolating sugar-
hydrogen bonding (HB) network. Above this threshold (i.e., at lower sugar con-
centration), the glass constitutes a percolating HB network of water molecules, with
a competing or complementary network of sugar-sugar and sugar-water HBs. If this
network is similar to that of pure water in terms of activation barriers for rotational
motion that liberate water-water and water-sugar HBs, it can be expected that the
Tgw would be similar to Tg for pure water at these lower sugar concentrations.
On the lower water concentration side of the threshold level near 40 % w/w
water, it seems reasonable that Tgw first increases with decreasing water content, if
more and more water molecules are forced to balance between HBs with other
water molecules that are more rotationally mobile, and sugars that are locked into a
percolating sugar HB matrix that is well below its Tg. Eventually, at low enough
water levels, most water molecules become isolated from other water molecules
Low-Temperature Mobility of Water in Sugar Glasses: Insights from Thermally. . . 85
5 Conclusions
Acknowledgements The study was supported in part by a Pfizer Global Research and Develop-
ment summer internship provided to Ali Hussain.
References
Aldous BJ, Franks F, Greer AL (1997) Diffusion of water within an amorphous carbohydrate.
J Mater Sci 32:301–308
Apekis L, Pissis P, Boudoris G (1983) Depolarization thermocurrents in ice Ih at low temperature
depending on the electrode material. Polarization mechanism. Nuovo Cimento 2D:932–946
Authelin J-R, MacKenzie AP, Rasmussen DH, Shalaev EY (2014) Water clusters in amorphous
pharmaceuticals. J Pharm Sci 103:2663–2672
Chou SG, Soper AK, Khodadadi S, Curtis JE, Krueger S, Cicerone MT, Fitch A, Shalaev EY
(2012) Pronounced micro-heterogeneity in a sorbitol-water mixture observed through variable
temperature neutron scattering. J Phys Chem B 116:4439–4447
Correia NT, Alvarez C, Moura Ramos JJ, Descamps M (2000) Molecular motions in molecular
glasses as studied by thermally stimulated depolarisation currents. Chem Phys 252:151–163
Correia NT, Alvarez C, Rames JJM, Descamps M (2001) The β α branching in D-sorbitol as
studied by thermally stimulated depolarization currents (TSDC). J Phys Chem B
105:5663–5669
Fagegaltier N, Lamure A, Lacabanne C, Caron A, Misfud H, Bauer A (1997) Thermal analysis of
amorphous phase in a pharmaceutical drug. J Therm Anal 48:459–464
Franks F (1997) Phase changes and chemical reactions in solid aqueous solutions: science and
technology. Pure Appl Chem 69:915–920
Galop M, Collins GL (2001) Thermally stimulated current observed in pharmaceutical products.
Thermochim Acta 37:367–368
Girlich D, Lüdemann H-D (1994) Molecular mobility of the water molecules in aqueous sucrose
solutions, studied by 2H-NMR relaxation. Z Naturforsch C J Biosci 49c:250–257
Gordon M, Taylor JS (1952) Ideal copolymers and the second-order transitions of synthetic
rubbers. I. Non-crystalline copolymers. J Appl Chem 2(9):493–500
Low-Temperature Mobility of Water in Sugar Glasses: Insights from Thermally. . . 87
Handa YP, Klug DD (1988) Heat capacity and glass transition behavior of amorphous ice. J Phys
Chem 92:3323–3325
Johari GP, Jones SJ (1975) Study of the low-temperature “transition” in ice Ih by thermally
stimulated depolarization measurements. J Chem Phys 62:4213–4223
Johari GP, Hallbrucker A, Mayer E (1987) The glass–liquid transition of hyperquenched water.
Nature 330:552–553
Koster KL (1991) Glass formation and desiccation tolerance in seeds. Plant Physiol 96:302–304
Mayer E (1991) Calorimetric glass transition in the amorphous forms of water: a comparison.
J Mol Struct 250:403–411
Pissis P, Diamanti D, Boudouris G (1983) Depolarisation thermocurrents in frozen aqueous
solutions of glucose. J Phys D Appl Phys 16:1311–1322
Ramos JJM, Diogo HP, Pinto SS (2007) Effect of physical aging on the Johari-Goldstein and α
relaxations of D-sorbitol: a study by thermally stimulated depolarization currents. J Chem Phys
126:144506
Reddy R, Chang L, Luthra S, Collins G, Lopez C, Shamblin SL, Pikal MJ, Gatlin LA, Shalaev EY
(2009) The glass transition and sub-t-g-relaxation in pharmaceutical powders and dried pro-
teins by thermally stimulated current. J Pharm Sci 98:81–93
Shmeis RA, Wang Z, Krill SL (2004) A mechanistic investigation of an amorphous pharmaceu-
tical and its solid dispersions. Part I. A comparative analysis by thermally stimulated depolar-
ization current and differential scanning calorimetry. Pharm Res 21:2025–2030
Sugisaki M, Suga H, Seki S (1968) Calorimetric study of glassy state. V. Heat capacities of glassy
water and cubic ice. Bull Chem Soc Jpn 41:2591–2599
Talja RA, Roos YH (2001) Phase and state transition effects on dielectric, mechanical, and thermal
properties of polyols. Thermochim Acta 380:109–121
Teyssedre G, Bernes A, Lacabanne C (1995) Investigation of dielectric relaxations associated with
the glass transition of vinylidene fluoride and trifluoroethylene copolymers by thermally
stimulated current. J Polym Sci B Polym Phys 33:2419–2428
Vanderschueren J, Gasiot J (1979) Field-induced thermally stimulated currents. In: Bräunlich P
(ed) Thermally stimulated relaxation in solids. Springer, Berlin, pp 135–223
Functional Behavior of Different Food
Components as Affected by Water
and Physical State
Abbreviations
aw Water activity
DE Dextrose equivalent
FD Freeze-dried
MW Molecular weight
PEGs Polyethylene glycols
SMP Skim milk powder
1 Introduction
The presence of water in foods, and the vast number of roles that water plays, is an
old and well-researched topic. There are countless interactions in which water
impacts the food chain, starting from irrigation of crops, harvested raw materials,
storing and processing, and finished goods properties such as shelf life, microbio-
logical quality and safety, functional properties, palatability and profitability, to
name a few.
In order to better understand and control the processing steps used to manufac-
ture different food products, it is essential to understand, control, and exploit the
unique properties of water and how it affects and transforms the ingredients
utilized.
In this chapter, we will focus on how different food components, namely
carbohydrates, proteins, and fat, are affected by the presence of water. We put
Fig. 2 Integrated approach to elucidate the complex interactions taking place from the moment
the consumer fills a spoon with a powder until it is possible to enjoy a chocolate drink
properties, as well as the physical processes taking place, including both heat and
mass transfer, should be considered in order to gain a deeper understanding of the
complex interactions taking place during the reconstitution process.
In order to exemplify the important role of material state on the dissolution process,
a simple example is presented in Fig. 3. The data was measured in a Setaram C80
Calvet-type calorimeter. The curves show the endo- or exothermic effect observed
as a result of the state of sucrose during the dissolution process. When completely
amorphous sucrose is dissolved, a fast and exothermic response (10 J/g) is
observed. In contrast, when crystalline sucrose is dissolved, a highly endothermic
curve results (18 J/g). When a partially recrystallized sample is dissolved, a mixed
reaction is obtained. It is worth noting that, in this case, and for many other samples
in which amorphous and crystalline materials are present simultaneously, the
exothermic peak is always observed first.
The fact that different physical states show opposite thermodynamic behavior
highlights that the kinetics of the process might be related to the thermal response
observed. This is exemplified by a simple example in which crystalline sucrose
spheres were dissolved in different liquid mixtures (water and polyethylene
glycols, PEGs) at 30 C. As shown in Fig. 4, a linear correlation was observed
Fig. 3 Thermal response upon dissolution of sucrose in amorphous and crystalline states;
a partially recrystallized sample is also shown
Functional Behavior of Different Food Components as Affected by Water. . . 93
Fig. 4 Correlation between enthalpy of dissolution and dissolution rate constant (k) for sucrose
spheres dissolved in different solutions (adapted from Marabi et al. 2008a)
between the dissolution rate constant and the enthalpy of dissolution. In this case,
more endothermic thermal responses were related to slower dissolution rates. This
example highlights the importance of the interaction between the solid and the
liquid in which it is dissolved, and also the fact that the thermodynamics of the
process can be effectively used to predict or even “tune” the dissolution rate. This
has further practical applications, as it is well known that the fastest dissolution
rates are not always desired, since particles that dissolve too quickly might lead to
the generation of lumps, in which the outer layers are dissolved and generate a
highly viscous boundary enclosing and avoiding the dissolution of the dry powder
inside.
The impact of the water content of maltodextrin (dextrose equivalent, DE, 21)
and the physical state of skim milk powder (SMP) samples was also studied in
terms of its thermodynamic response upon dissolution (Figs. 5 and 6). The samples
were characterized by their enthalpy of dissolution at three different conditions:
freeze-dried (FD); after equilibration at 54.4 % RVP; and after FD of the equili-
brated samples. The dissolution was found to be exothermic for all tested samples,
and the curves showed similar shapes. A clear decrease in the exothermic response
is observed when FD and equilibrated samples are compared. Lower exothermic
response is linked to higher moisture content of the samples. However, there is a
significant difference between the two samples. In the case of maltodextrin, exo-
thermic heat of dissolution is fully recovered after the equilibrated sample is dried.
In contrast, when the equilibrated SMP sample is dried, only ca. 50 % of the heat of
94 A. Marabi et al.
Fig. 5 Typical dissolution calorimetry curves of tested maltodextrin powders. All samples
showed an exothermic response; however, increased moisture content led to a less exothermic
dissolution process (adapted from Marabi et al. 2007)
Fig. 6 Typical dissolution calorimetry curves of tested skim milk powders. All samples showed
an exothermic response; however, increased moisture content and recrystallized lactose led to a
less exothermic dissolution process (adapted from Marabi et al. 2007)
Functional Behavior of Different Food Components as Affected by Water. . . 95
Apart from carbohydrates, proteins are another important category of key constit-
uents of food powders. Dissolution behavior of proteins can thus affect the overall
rehydration of several food products, such as infant formulae, and be a limiting step
for effective reconstitution.
For instance, Fig. 8 shows the overall rehydration behavior of micellar casein
and Na-caseinate in water. The evolution of concentration across time is measured
by an electric conductivity probe.
It can be observed that, compared to carbohydrates, the timescale for complete
dissolution of proteins is much longer (minutes vs. seconds), and it is strongly
dependent on the protein structure. Na-caseinate dissolves much more slowly than
micellar casein, due to its open structure, in which hydrophobic sites are exposed.
A similar effect can be noticed when proteins are denatured; for instance, denatured
β-lactoglobulin is almost insoluble, while its native form can easily be rehydrated.
It can also be noticed that the initial water activity (and thus water content) of the
powder plays an important role. For Na-caseinate, higher water activity leads to
statistically significant faster initial dissolution. However, for other proteins oppo-
site behavior has been observed.
Fig. 8 Overall reconstitution behavior of micellar casein and Na-caseinate powders equilibrated
at aw ¼ 0.22 and 0.71 (dissolution in water, 2 % w/v, 750 rpm stirring)
Functional Behavior of Different Food Components as Affected by Water. . . 97
Fig. 9 Evolution of effective contact angle between a water drop and a layer of powder along time
(left: micellar casein layer; right: Na-caseinate layer). In both cases, powders were equilibrated at
aw of 0.22 and 0.71
Fig. 10 Dissolution calorimetry curves with calculated enthalpy of dissolution of micellar casein
and Na-caseinate powders, equilibrated at different water activities
4 Conclusions
• Food powders are an important food product category, and they account for a
significant portion of new products launched every year.
• Key properties of ingredients and finished products should be studied and
understood in order to optimize their functionality.
• The physical state and water content of food powders will impact their ultimate
behavior, and should therefore be controlled in order to optimize performance
and ensure consumer satisfaction.
• Measuring and understanding the thermodynamics of the dissolution process
provided additional insight, which, coupled with dissolution kinetics, will result
in products adapted to consumer needs.
• Incorporation of new ingredients in existing products presents a challenge, and
therefore there is a need to carefully design and present all aspects of newly
developed products.
References
Dilworth SE, Buckton G, Gaisford S, Ramos R (2004) Approaches to determine the enthalpy of
crystallisation, and amorphous content, of lactose from isothermal calorimetric data. Int J
Pharm 284:83–94
Gaiani C, Scher J, Schuck P, Hardy J, Desobry S, Banon S (2006) The dissolution behaviour of
native phosphocaseinate as a function of concentration and temperature using a rheological
approach. Int Dairy J 16:1427–1434
Gaiani C, Schuck P, Scher J, Desobry S, Banon S (2007) Dairy powder rehydration: influence of
protein state, incorporation mode, and agglomeration. J Dairy Sci 90:570–581
Functional Behavior of Different Food Components as Affected by Water. . . 99
Gao D, Rytting JH (2006) Use of solution calorimetry to determine the extent of crystallinity of
drugs and excipients. Int J Pharm 151:183–192
Haque MK, Roos YH (2006) Differences in the physical state and thermal behavior of
spray-dried and freeze-dried lactose and lactose/protein mixtures. Innovat Food Sci Emerg
Technol 7:62–73
Marabi A, Mayor G, Raemy A, Bauwens I, Claude J, Burbidge AS, Wallach R, Saguy IS (2007)
Solution calorimetry: a novel perspective into the dissolution process of food powders. Food
Res Int 40:1286–1298
Marabi A, Mayor G, Raemy A, Burbidge AS, Wallach R, Saguy IS (2008a) Assessing dissolution
kinetics of powders by a single particle approach. Chem Eng J 139:118–127
Marabi A, Raemy A, Bauwens I, Burbidge AS, Wallach R, Saguy IS (2008b) Effect of fat content
on the dissolution enthalpy and kinetics of a model food powder. J Food Eng 85:518–527
Schubert H (1987) Food particle technology. I. Properties of particles and particulate food
systems. J Food Eng 6:1–32
Effect of Different Components of Edible/
Biodegradable Composite Films on Water
Relationships in the Polymer Matrix
Abbreviations
1 Introduction
Fig. 1 SEM micrographs of the cross section of sodium caseinate-glycerol (1:0.30) films without
(control) and with lipids (oleic acid beeswax mixtures in the ratios 1:0, 0.3:0.7 and 0:1)
The protein-lipid ratio was 1:0.5 in all cases. The lipid dispersed phase can be
appreciated, showing clear differences depending on the lipid composition. The
heterogeneity of the structure is also reflected in the film surface, as has been
observed by the atomic force microscopy (AFM) (Fabra et al. 2009a, c). The
different surface topography is the result of the initial lipid particle size in the
film forming dispersion and the destabilization phenomenon, which occur during
film drying and which imply flocculation, coalescence, and creaming of lipid
particles. The more unstable dispersions lead to a greater surface roughness of the
films due to the faster progression of destabilization phenomena and, as a result, the
lipid migration to the film surface is more intense. Surface roughness has a high
impact on the film gloss; the greater the surface roughness, the lower the gloss
(Villalobos et al. 2005; Fabra et al. 2009a).
In polar lipids, such as fatty acids, the lipid self-association phenomenon plays
an important role in the film microstructure. This phenomenon is determined by
molecular interactions in the film forming dispersion and their changes in line with
the water evaporation during the film formation. Several authors observed a mul-
tilayered structure for saturated fatty acids (lauric, palmitic, and stearic) in sodium
caseinate (Fabra et al. 2009c) and hydroxypropyl methylcellulose (Jiménez
et al. 2010) films. This was associated with the growth of the molecular
104 A. Chiralt et al.
aggregations in the bilayer self-association of the lipid molecules during film drying
and the final crystallization of fatty acids in the polymer matrix. The length of lipid
layers in the matrix was greater for lauric acid and decreased as the fatty acid chain
length increased. This indicates that the growth of the molecule aggregates during
film drying was more effective when the fatty acid chain was shorter, probably due
to its greater molecular mobility. These successive lipid layers in the film matrix
greatly contribute to reduce water vapor permeability, thus improving the water
barrier properties; the longer the lipid layers, the lower the water vapor permeability
(WVP) values. Films containing stearic acid, of a more hydrophobic nature, showed
the poorest water barrier properties, which can be explained by these structural
effects. These considerations underline the importance of relating the properties of
the film forming dispersions with the relevant characteristics of the films. The
properties related with the stability of the film forming dispersion such as rheology,
particle size distribution, zeta-potential, or contact angle on a determined surface
affect the properties related to the film functionality to a great extent.
Optical properties of the films such as gloss and transparency are important,
since they have an impact on product appearance and are greatly affected by the
degree of the film’s structural heterogeneity (Villalobos et al. 2005; Ozdemir and
Floros 2008; Fabra et al. 2009a). Nevertheless, one of the most relevant properties
of the films/coatings is the water affinity and water sorption behavior, since
equilibrium water content of the film impacts directly on its mass transport prop-
erties (water vapor and gas permeability) and mechanical resistance, which are
decisive for the film’s practical applications.
Components of the hydrocolloid matrix can interact to a different degree with
the polymer chain/groups, giving rise to systems where water relationships can
change considerably, as compared with pure hydrocolloid. Some examples of how
the compounds incorporated into the polymer matrix can affect water sorption
capacity, phase transitions, or molecular mobility are commented on for some
protein or polysaccharide matrices.
Plasticizers, such as glycerol, or sorbitol, are usual components of films they have a
great impact on the sorption behavior and water sorption capacity (WSC) of the
hydrocolloid films due to their hydrophilic nature. Figure 2a shows the sorption
isotherms, at 5 and 20 C, of films based on soy protein isolate (SPI), both pure and
combined with glycerol in a ratio 1:0.3, where the promotion of the WSC of the
matrix due to the glycerol addition can be observed at both temperatures. This
promotion is particularly intense at high water activity levels, when the solvent
effects of water are exhibited for films containing low molecular weight polar
solutes, such as glycerol.
Effect of Different Components of Edible/Biodegradable Composite Films. . . 105
a b 0,8
0,8
0,7
0,7
0,6
0,6
0,5
0,5
we(d.b.)
we (d.b.)
0,4
0,4
0,3 0,3
0,2 0,2
0,1 0,1
0 0
0 0,2 0,4 0,6 0,8 aw 1 0 0,2 0,4 0,6 0,8 aw 1
Fig. 2 Effect of plasticizer (glycerol) on water sorption isotherms of biopolymer films based on
soy protein isolate (a), at 5 C (triangles) and 20 C (rhombus), and sodium caseinate (b), at 25 C.
Closed symbols: samples without glycerol, open symbols: samples with glycerol
Lipids are also usually added to the film formulation in order to improve water
barrier properties. The incorporation of these compounds also changes the WSC of
the film. The effect of lipids can be very different, depending on the hydrocolloid-
lipid interactions in the matrix. The incorporation of lipids usually reduces water
sorption due to the fact that lipids correspond to a fraction of solids with small water
uptake capacity, especially for the more hydrophobic compounds. This has been
observed for films of sodium caseinate (with 0.3 g of glycerol/g protein) containing
oleic acid (OA) and beeswax (BW) mixtures (Fabra et al. 2010). In this case, the
amount of water uptake was estimated from the values obtained for the lipid-free
film and the mass fraction of the nonlipid solids (caseinate plus glycerol) and
106 A. Chiralt et al.
Fig. 3 Effect of oleic acid on water sorption isotherms of soy protein isolate films containing
glycerol (1:0.3) in different lipid-protein ratios (0.5 (triangles) and 0.25 (squares)) at 5 C (a) and
20 C (b)
predicted values were compared with the experimental ones. The comparison
showed that the values agreed quite closely. Nevertheless, for films containing
oleic acid experimental values are slightly greater than predicted, which means that
a certain amount of water uptake in this case is retained by the lipid. Nevertheless,
the opposite behavior was observed for films containing pure BW, where predicted
values are lower than the experimental ones, thus indicating that BW seems to
inhibit the water sorption capacity of the protein matrix, probably due to the
promotion of hydrophobic interactions in the lipid-protein matrix which reduce
the active points for water adsorption.
Figure 3 shows the effect of the incorporation of OA to SPI matrices containing
glycerol (0.3 g/g protein) on water sorption behavior at 5 and 20 C. A different
effect of lipid can be observed at each temperature for different lipid-protein ratios
(0.25 and 0.5 g/g protein). For the highest ratio, OA promotes water adsorption at
both 5 and 20 C, which indicates that a greater amount of water sorption active
groups are exhibited in the matrix when OA interacts with the SPI chains. Never-
theless, when the OA-protein ratio decreased to 0.25 g lipid/g protein, a clear
decrease in the water sorption capacity was observed when film was equilibrated
at 20 C. This seems to indicate that changes in molecular interactions are induced
by the temperature increase, which affects the number of water sorption active
points.
Figure 4 shows the sorption behavior of SPI-glycerol films, at 5 and 20 C, when
BW was incorporated at two different ratios, 0.25 and 0.5 g lipid/g protein. Very
similar behavior can be observed in both cases where WSC of the matrix decreased
because of the addition of BW.
By only considering the contribution of the nonlipid fraction, the predicted
moisture content values of lipid-containing SPI films were compared with the
experimental values (as shown in Fig. 5). For OA at 5 C, points were observed
to deviate greatly from the diagonal, as expected from the previously mentioned
promotion of WSC, clearly showing the increase in the number of active points in
Effect of Different Components of Edible/Biodegradable Composite Films. . . 107
Fig. 4 Effect of beeswax on water sorption isotherms of soy protein isolate films containing
glycerol (1:0.3) in different lipid-protein ratios (0.5 (triangles) and 0.25 (squares)) at 5 C (a) and
20 C (b)
Fig. 5 Comparison of experimental and predicted values of equilibrium water content of SPI
films containing glycerol (1:0.3 protein-glycerol ratio) and lipids: OA (triangles) and BW
(squares) in a ratio protein-lipid of 1:0.25 ( full symbols) and 1:0.50 (empty symbols), at 5 C (a)
and 20 C (b)
diagonal were observed for films containing BW at either temperature, thus indi-
cating that this component is in a separate phase, hardly interacting with the
polymer network.
The effect of the incorporation of other organic compounds, such as ethanol
extract of propolis (EEP), on the WSC of hydroxypropyl methylcellulose (HPMC)
films has been described for films equilibrated at 5 and 25 C and 55 or 75 %
relative humidity (RH) (Pastor et al. 2010). A decrease in the equilibrium water
content of the film when the content of EEP in the film increased was observed for
all of the equilibration conditions. Nevertheless, it is remarkable that when moisture
content was referred to the HPMC basis (g water/g HPMC), no significant change in
water content was detected until there was nearly 10 % of EEP in the film.
However, a significant and similar decrease was obtained for the two greatest
EEP ratios (0.167 and 0.231 g/g dried film) in the films. This reduction was greater
for samples equilibrated at low RH conditions (about 80 and 60 % reduction at
5 and 25 C, respectively) than for those equilibrated at high RH (about 10 % at
both temperatures). These results seem to indicate that whereas at low propolis
contents the interactions with the EEP compounds seem not to affect the sorption
behavior notably, at the highest propolis contents, these compounds notably modify
the water-film relationships, thus reducing the ability of the polymer chains to bond
water molecules. This fact coincides with notable differences observed in the
behavior of the most highly EEP concentrated (1.0 and 1.5 %, wt) film forming
dispersions which points to the bonding of HPMC chains to the EEP components,
thus limiting their capacity for water interactions. The degree of the polymer
bonding was greater at low moisture contents in the films and seems to reduce in
line with the water gain in the matrix. In this sense, it is remarkable that the
temperature effect on water sorption, which is related to the sorption enthalpy,
was very similar for all samples containing different ratios of EEP at high water
contents (when samples were equilibrated at about 75 % RH). Nevertheless, it is
greatly reduced at low water contents (when samples were equilibrated at 56 % RH)
for the highest ratios of propolis in the films. This indicates a great decrease of the
water sorption enthalpy in these samples, which points to the promotion of the
hydrophobicity of the matrix, associated with the above-mentioned bonding of
HPMC to the propolis compounds, which blocks the HPMC active points.
order to know how the protein-lipid interactions affect the film properties (Fabra
et al. 2010). In this sense, glass transition of the hydrocolloid matrices and lipid
melting endotherm were characterized as a function of the sample water activity
(aw). Table 1 shows the values of Tg (midpoint) obtained for pure sodium caseinate,
sodium caseinate-glycerol film (1:0.3 ratio), and sodium caseinate-glycerol-OA
film (1:0.3:0.5 ratios) as a function of the sample aw. For samples containing
BW, Tg was not obtained because of the fact that the melting endotherm of BW
extends over a wide temperature range, overlapping the glass transition of the
amorphous matrix. The expected decrease of Tg when sample aw decreases can
be observed for all samples, as corresponds to the plasticizing effect of water, which
implies a reduction in the mean molecular weight in the system and the
corresponding increase in the molecular mobility of the amorphous solid.
It is remarkable that the presence of glycerol implies a marked decrease in the Tg
values with respect to the values obtained for caseinate at the same aw, mainly at
low aw values. This is coherent with the plasticizing role of glycerol, decreasing the
mean molecular weight in the system and affecting molecular mobility and
mechanical behavior of the material. The plasticizing effect of water in this case
is very mild, since the Tg decrease with aw is very small. This indicates that the
glycerol interactions in the matrix inhibit the effect of water and no subsequent,
significant plasticization effect was induced. In the completely dried film, 0.3 g of
glycerol/g of protein allows the Tg to reach a value in the order of that obtained in
the pure caseinate film at aw ¼ 0.675.
The observed behavior indicates that water plasticization effect is more effective
in caseinate than in the caseinate-glycerol system, which has also been observed for
caseinate and pullulan matrices containing sorbitol as plasticizer (Kristo and
Biliaderis 2006) and for gluten matrices when fructose, glucose, and sucrose are
incorporated (Kalichevsky et al. 1992). The main factor responsible for this behav-
ior could be the great water bonding capacity of glycerol and other polyol plasti-
cizers that limits the water interactions with the polymer chains and determines the
final distribution of water in the system.
In the samples containing OA, no significant effect of this component on Tg
values was observed below aw ¼ 0.43, which seems to indicate that OA is in a
110 A. Chiralt et al.
separate phase and no notable interactions with the caseinate matrix take place.
However, a different tendency was observed from this aw value onwards, where Tg
in the samples containing OA showed a more decreasing tendency with aw. This
behavior indicates that the plasticizing effect of water on the caseinate-glycerol
system was promoted in the presence of OA from intermediate aw values onwards,
where the molecular mobility has reached a critical level. Interactions of OA in the
caseinate matrix could limit the protein chain aggregations to some extent, provok-
ing this effect.
When the Gordon and Taylor equation was fitted to the Tg-water content
relationship for the different samples, the above-mentioned glycerol and OA effects
on Tg were reflected in significant changes in both Tgas and k parameters (Fabra
et al. 2010). The k value is a measure of the water plasticization effect, which is very
small in samples containing glycerol (k ¼ 0.12), as compared with pure caseinate
(k ¼ 1.56), and slightly increases when OA is incorporated in the matrix (k ¼ 0.68),
in agreement with the promotion of the water plasticization effect caused by
OA. The Tg values of the anhydrous solids (Tgs) were 109, 58, and 63 C,
respectively, for caseinate, caseinate-glycerol, and caseinate-glycerol-OA films.
The protein-lipid interactions may also affect lipid phase transitions, melting
temperature range, and enthalpy. Fabra et al. (2010) analyzed lipid phase transition
in the sodium caseinate films containing OA and BW, equilibrated at different aw
values, in order to obtain complementary information about their interactions in the
film matrix. No significant differences were observed between pure BW melting
characteristics (initial melting temperature: 31.3 (2.4) C, final melting tempera-
ture: 88.3 (2.8) C, peak temperature: 69.4 (1.2) C, and melting enthalpy
159 (9) J/g) and those obtained for BW in films equilibrated at different aw. The
initial and final temperatures of the endotherm were 29 and 84 C respectively, peak
temperature was 66.4 (0.5) C, and melting enthalpy was 155 (6) J/g BW,
without significant differences associated with the equilibrium aw (Fig. 6). This
seems to indicate that no notable interactions were developed between protein and
wax, which behave as if they were in separate phases. Nevertheless, significant
differences were observed among OA melting endotherms of the pure compound
and OA films equilibrated at different aw values. Pure OA behaves in a similar way
to that described by other authors (Cede~no et al. 2001), with a solid-solid transition
(Tr1) at about 50 C (ΔH ¼ 5.0 J/g) and a solid-liquid transition (Tr2) at about
5 C (ΔH ¼ 75.5 J/g), showing a broad melting endotherm. The obtained values for
the melting enthalpy (Tr2) appear in Fig. 6, as a function of the equilibrium aw
value, where a significant reduction was observed in line with the aw increase. So, in
film samples, OA crystallization was inhibited to quite an extent, as deduced from
the melting enthalpy reduction, and this inhibition increases in line with the
increase in sample aw, despite the fact that the associated increase in molecular
mobility usually promotes crystallization rates. These results are in agreement with
that previously commented on with respect to the OA interactions in the caseinate
matrix, which seem to be promoted when aw increases. These interactions lead to
the bonding of OA molecules which then are not able to crystallize. So, a part of OA
seems to be linked to the caseinate molecules and when the water content increases
Effect of Different Components of Edible/Biodegradable Composite Films. . . 111
DH (mJ/mg lípìdo)
activity of the film 120
NaCas-Gly-OA
100
Nacas-Gly-BW
80
60
40
20
0
0 0.2 0.4 0.6 0.8 1
aw
in the sample, these bonds weaken the polymeric cross-linking in the matrix,
making it softer (lower Tg values). This has a great impact on the film’s mechanical
and barrier properties due to the influence of molecular mobility on both rheolog-
ical properties and diffusional processes. In fact, films containing OA showed a
marked decrease in both elastic module (EM) and tensile at break (TS) and a sharp
increase in elongation at break (E), enhanced by the increase in aw, as compared
with the lipid-free caseinate films (Fabra et al. 2010). It is remarkable that the film
with OA did not break under the mechanical test conditions when aw is 0.75.
However, when BW was the lipid present, the film was more fragile and broke at
much lower deformation levels, showing the effect of the matrix discontinuities and
the fact that the solid dispersed phase was less deformable (Fabra et al. 2010).
5 Final Remarks
The results commented on reflect the impact of water and lipid-protein interactions
on film functionality. Lipids are usually incorporated to modify film barrier prop-
erties which are essential to define their application. In this sense, in the composite
films obtained through emulsification of components, the development of com-
pound interactions, such as in the case of the OA-caseinate or OA-SPI systems,
gives rise to particular functional properties of the film that are not expected in films
formed with the lipid dispersed in a separate phase. These aspects are very impor-
tant in the developing of emulsified polymer films, since two kinds of fillers can be
found: inactive or active. Whereas the former only gives rise to discontinuities in
the polymer continuous matrix, the latter greatly modifies the properties of this
matrix, by changing the cross-linking behavior of the polymer chains, which leads
to more marked differences in the film behavior. Physical properties of heteroge-
neous systems are much more affected by the characteristics of the continuous
112 A. Chiralt et al.
phase than those of the dispersed phase. So, when active fillers are incorporated in
the matrix, the product can behave very differently than when the matrix is not
filled. Phase transition analysis as a function of water activity is a good approach to
identify the degree of polymer-filler interactions, which in turn are usually affected
by water content, and so to predict the final effect of the filler on the film
equilibrated at different relative humidities.
Acknowledgment The authors acknowledge the financial support from Spanish Ministerio de
Educaci
on y Ciencia throughout the project AGL2007-01009 and AGL2010-20694.
References
Alves VD, Costa N, Colehoso IM (2010) Barrier properties of biodegradable composite films
based on kappa-carrageenan/pectin blends and mica flakes. Carbohydr Polym 79:269–276
Cede~no FO, Prieto MM, Espina A, Garcı́a JR (2001) Measurements of temperature and melting
heat of some pure fatty acids and their binary and ternary mixtures by differential scanning
calorimetry. Thermochim Acta 369:39–50
Debeaufort F, Quezada-Gallo JA, Delporte B, Voilley A (2000) Lipid hydrophobicity and physical
state effects on the properties of bilayer edible films. J Membr Sci 180:47–55
Elsabee MZ, Abdou ES, Nagy KSA, Eweis M (2008) Surface modification of polypropylene films
by chitosan and chitosan/pectin multilayer. Carbohydr Polym 71:187–195
Fabra MJ, Talens P, Chiralt A (2009a) Microstructure and optical properties of sodium caseinate
films containing oleic acid–beeswax mixtures. Food Hydrocoll 23:676–683
Fabra MJ, Hambleton A, Talens P, Debeaufort F, Chiralt A, Voilley A (2009b) Influence of
interactions on water and aroma permeabilities of ι-carrageenan–oleic acid–beeswax films
used for flavour encapsulation. Carbohydr Polym 76:325–332
Fabra MJ, Jiménez A, Atarés L, Talens P, Chiralt A (2009c) Effect of fatty acids and beeswax
addition on properties of sodium caseinate dispersions and films. Biomacromolecules
10:1500–1507
Fabra MJ, Talens P, Chiralt A (2010) Water sorption isotherms and phase transitions of sodium
caseinate–lipid films as affected by lipid interactions. Food Hydrocoll 24:384–391
Gomez-Estaca J, Lopez de Lacey A, Lopez-Caballero ME, G omez-Guillén MC, Montero P (2010)
Biodegradable gelatin–chitosan films incorporated with essential oils as antimicrobial agents
for fish preservation. Food Microbiol 27:889–896
Hambleton A, Voilley A, Debeaufort F (2011) Transport parameters for aroma compounds
through i-carrageenan and sodium alginate-based edible films. Food Hydrocoll 25:1128–1133
Jiménez A, Fabra MJ, Talens P, Chiralt A (2010) Effect of lipid self-association on the micro-
structure and physical properties of hydroxypropyl-methylcellulose edible films containing
fatty acids. Carbohydr Polym 82:585–593
Kalichevsky MT, Jaroskiewicz EM, Blanshard JMV (1992) Glass transition of gluten. 1. Gluten
and gluten–sugar mixtures. Int J Biol Macromol 14:257–266
Kristo E, Biliaderis CG (2006) Water sorption and thermo-mechanical properties of water/
sorbitol-plasticized composite biopolymer films: caseinate-pullulan bilayers and blends.
Food Hydrocoll 20:1057–1071
Krochta JM, de Mulder-Johnston C (1997) Edible and biodegradable polymer films: challenges
and opportunities. Food Technol 51:61–74
Morillon V, Debeaufort F, Blond G, Capelle M, Voilley A (2002) Factors affecting the moisture
permeability of lipid-based edible films: a review. Crit Rev Food Sci Nutr 42:67–89
Effect of Different Components of Edible/Biodegradable Composite Films. . . 113
Ozdemir M, Floros JD (2008) Optimization of edible whey protein films containing preservatives
for mechanical and optical properties. J Food Eng 84:116–123
Pastor C, Sánchez-González L, Cháfer M, Chiralt A, González-Martı́nez C (2010) Physical and
antifungal properties of hydroxypropyl-methylcellulose based films containing propolis as
affected by moisture content. Carbohydr Polym 82:1174–1183
Roos YH (1995) Phase transitions in food. Academic, San Diego, CA, 360 p
Sánchez-González L, González-Martı́nez C, Chiralt A, Cháfer M (2010) Physical and antimicro-
bial properties of chitosan-tea tree essential oil composite films. J Food Eng 98:443–452
Sánchez-González L, Cháfer M, Hernández M, Chiralt A, González-Martı́nez C (2011) Antimi-
crobial activity of polysaccharide films containing essential oils. Food Control 22:1302–1310
Srinivasa PC, Ramesh MN, Tharanathan RN (2007) Effect of plasticizers and fatty acids on
mechanical and permeability characteristics of chitosan films. Food Hydrocoll 21:1113–1122
Tharanathan RN (2003) Biodegradable films and composite coatings: past, present and future.
Trends Food Sci Technol 14:71–78
Villalobos R, Chanona J, Hernández P, Gutiérrez G, Chiralt A (2005) Gloss and transparency of
hydroxypropyl-methylcellulose films containing surfactants as affected by their microstruc-
ture. Food Hydrocoll 19:53–61
Glass Transition Observed with Cross-Linked
Dextrans Containing a Small Amount
of Water
Abbreviations
1 Introduction
Many biological systems and foods take on a gelling state in their ordinary
condition. Gels consisting of various polymers are also used in the pharmaceutical
and medical industries. Thus, it is of practical importance to characterize the
polymer network in gels in order to understand their physical properties. Moreover,
it is necessary to rationalize freezing and/or freeze-drying processes of the gels on
the basis of scientific fundamentals for the preparation of frozen and freeze-dried
products (Franks 1986). For differential scanning calorimetry (DSC) studies
conducted from this standpoint, it was found by one of the present authors that
water in polymer gels sometimes remains partially unfrozen during cooling below
50 C and freezes during subsequent heating (rewarming) (Murase et al. 1982,
1986). The occurrence of ice crystallization during rewarming, especially that
2.1 Materials
this study, but G15 was also used in some cases. The density of cross-links was
highest for G15, with the smallest G number among the three Sephadexes.
Samples were first dried in a drying oven for 3 h. The temperature of drying was
kept at 80 C to prevent the occurrence of dehydration reaction between and/or
within dextran molecules when heated above that temperature. Dried samples were
placed in a desiccator containing P4O10 for 1 week and kept at 30 C to assure
complete drying. Samples containing a large amount of water were prepared
directly in aluminum pans by adding deionized water to dried samples. In the
case of preparation of hydrated polymers containing a small amount of water,
dried samples were transferred to desiccators where relative humidity was con-
trolled with aqueous solutions saturated with inorganic salts, and kept for another
1 week at 30 C. Water content in the samples was determined by measuring the
weight increase.
For the investigation of freezing behavior of gels with a large amount of water,
about 10 mg of the sample was placed in an aluminum pan, and it was hermetically
sealed. Cooling was conducted from 30 to 50 C, followed by heating to 30 C.
Cooling and heating rates were 5 C/min. For the investigation of glass transition
behavior without freezing, 15–20 mg of the polymer sample containing 0–30 wt%
water was placed in an aluminum pan. Samples, once heated to 60 C, were cooled
to 20 C, kept at that temperature for 2 min, followed by heating to 80 C. For the
measurement of enthalpy relaxation, samples held at predetermined temperatures
below Tg for different times were cooled to 20 C, followed by heating to 80 C.
Cooling and heating scans using the same samples were successively repeated
to investigate the effect of thermal history. Rates of cooling and heating were
20 C/min. No weight loss of samples was confirmed after each experiment. DSC
measurements were carried out by the use of a Pyris 1 (Perkin Elmer Inc, MA,
USA) equipped with a cooling unit.
3 Results
DSC rewarming traces obtained with G25 (Sephadex) gels containing different
amounts of water are shown in Fig. 1. For the gel containing more than 31 wt%
water, ice melting endotherm was observed, indicating the appearance of water
118 N. Ijima and N. Murase
phase separated from the hydrated cross-linked dextran. Neither ice melting endo-
therm during rewarming nor ice crystallization exotherm during precedent cooling
was observed with gel containing water less than 30 wt%; therefore, the polymer–
water system forms a single phase. An exotherm due to ice crystallization during
rewarming appeared with the gel containing more than 39 wt% water, which became
more noticeable with the increase in water content. An endothermic trend prior to the
exotherm started at ca. 18 C (Murase et al. 1986, 2007). Change in heat capacity
due to glass transition, however, was not identified, since it might overlap with the
endothermic trend, probably due to melting of small ice crystals (Fig. 1).
DSC rewarming traces obtained with G25 and G200 (Sephadex) containing a
small amount of water below 30 wt% are shown in Fig. 2. Within the range of water
content, dextran–water systems form a single phase. Change in heat capacity due to
glass transition was clearly indicated with the systems. Tgs decreased with increas-
ing water. Moreover, enthalpy relaxation was observed in the first rewarming trace
with gel samples containing water less than 14.2 wt% for G25 and less than
15.7 wt% for G200 (Fig. 2a, c). Enthalpy relaxation disappeared, however, in the
second rewarming traces after the cooling following the first rewarming heated to
80 C (Fig. 2b, d). For samples containing 30 wt% water for G25 and 29.7 wt% for
G200, Tgs were observed at ca. 16 C and 17 C, respectively. For samples
containing water less than 9.0 wt% for G25 and 10.9 wt% for G200, DSC
rewarming traces behaved complicatedly around the glass transition.
Glass Transition Observed with Cross-Linked Dextrans. . . 119
Fig. 2 DSC first rewarming and second warming traces obtained with Sephadex G25 and G200
containing a small amount of water. Second rewarming was conducted after cooling to 20 C
following first rewarming to 80 C. (a) First rewarming for G25; (b) second rewarming for G25;
(c) first rewarming for G200; (d) second rewarming for G200
Tgs were plotted against water content, and Tg-composition curves obtained by
the use of regression analysis are shown in Fig. 3. In the present study, onset
temperatures using second rewarming traces were adopted as Tg values. It was
found that a Tg-composition curve obtained for G200 was not so different from the
curve obtained for G25, but was slightly higher in temperature than that of G25.
For the comparison of glass transition behavior observed with hydrated dextrans
of different cross-link density, first rewarming traces obtained with G15, G25 and
G200 were overlaid in Fig. 4a. It can be noticed that not only is enthalpy relaxation
remarkable but also the temperature range of glass transition is narrowest with
G200 among the three Sephadexes. With the G15 sample, glass transition was
scarcely observed. Moreover, the increment of heat capacity at the glass transition
is larger with the G200 sample than at G25. In Fig. 4b, second rewarming traces are
120 N. Ijima and N. Murase
Fig. 3 Tg-composition
curves obtained with
Sephadex containing a
small amount of water
lower than 30 wt% (Murase
et al. 2007)
Fig. 4 Glass transition and enthalpy relaxation behavior dependent on density of cross-links.
Samples for DSC first rewarming (a) were held at 35 C for 10 days (G25 and G200) or 7 days
(G15) before measurement. Second rewarming (b) was carried out successively after first
rewarming, followed by cooling
overlaid. These results were obtained after heat treatment to eliminate the thermal
history. Except for the disappearance of enthalpy relaxation, glass transition behav-
ior dependent on the cross-link density was similar to that observed in first
rewarming traces. The increment of heat capacity at the glass transition obtained
with second rewarming traces was 0.2–0.3 J/g C for G25 and 0.35–0.45 J/g C for
G200, respectively.
Glass Transition Observed with Cross-Linked Dextrans. . . 121
4 Discussion
Fig. 5 Schematic
representation of cross-
linked polymer chains.
L (left): cross-link density is
high, and polymer chains
are hard to interact each
other; R (right): cross-link
density is low, and polymer
chains are ready to interact
and associate with each
other. Therefore, the
number of CRR increases
with decreasing temperature
Adam–Gibbs theory (Adam and Gibbs 1965). With decrease in temperature, the
number of CRR (cooperatively rearranging regions) increases in the case of dex-
trans of lower cross-link density, as the dextran chains are flexible and easily
associate with each other. On the other hand, the number hardly increases with
decrease in temperature in the case of dextrans of higher cross-link density, as they
are restricted in motion and interaction between dextran chains prevented by the
cross-links. A schematic representation is shown in Fig. 5.
The lowest Tg obtained with hydrated G25 of maximum water content to form a
single phase was ca. 16 C. With G25 gel containing 55–59 wt% water, the
initiation temperature of the endothermic trend prior to the exotherm during
rewarming was ca. 18 C. The coincidence of two temperatures indicates that
some water in the G25 gel forming a separated phase is vitrified together with
hydration water around dextrans at the time of freezing, as the temperature is below
Tg of hydration water.
5 Conclusions
References
Abbreviations
1 Introduction
Textural properties are key drivers for food acceptability (Chauvin et al. 2008). In
low-moisture cereal foods, quality depends mainly on textural attributes like crisp-
ness (Fontanet et al. 1997). Water is one of the most important factors affecting
texture of low-moisture foods, and its effect has been extensively studied (Labuza
et al. 1970; Roos et al. 1996; Peleg 1998; Jacoby and King 2001; Lewicki 2004;
Castro-Prada et al. 2009). It is also known that water affects glass transition temper-
ature (Chen et al. 1997), which is widely used in assessing stability (Roos 2010). In
order to measure water mobility in complex and heterogeneous systems, nuclear
magnetic resonance spectroscopy has been widely used (Chinachoti et al. 2006). On
the other hand, in low-moisture starchy food, other processes that take place at glassy
state, such as physical aging and toughening, could affect textural properties
(Suwonsichon and Peleg 1998; Chang et al. 2000; Chung and Lim 2004).
The shelf life stability and textural properties of cornflakes are closely related to
their mechanical and thermal properties. Results of a previous work indicated that
while compression force showed a maximum as a function of water content (wc),
glass transition temperature (Tg) decreased progressively as wc and water mobility
increased (Farroni et al. 2010). The aim of this study was to analyze the effect of
water and molecular mobility on textural properties of cornflakes.
2.1 Samples
Fð t Þ
σE ¼ ð1Þ
A0
128 A.E. Farroni et al.
H ðtÞ
εE ¼ ð2Þ
H0
where F(t) is force as function of time, H0 is sample initial height, H(t) is sample
height as a function of time, and A0 is the area under compression curve.
Deformability modulus was calculated as the slope of σ E vs. εE curves in the
lineal region (Eq. 3) (Mohsenin and Mittal 1977; Hicsasmaz and Rizvi 2005):
σ EðtÞ
Ed ¼ ð3Þ
εEðtÞ
The Lorentzian equation (Peng and Lu 2005; Yang et al. 2010) was fitted to
deformability modulus vs. water content data (Eq. 4):
A
y¼ 2 þ b ð4Þ
ðxxo Þ
1þ γ
where y is the value of the dependent variable (in this case Ed); x is the independent
variable (water content); A is the amplitude when y reaches maximum value; xo is
the x value when y ¼ A; γ is the width of the curve when y ¼ A/2; and b is the
asymptotic value of y.
A sensory panel was trained with the texture profile method following the pro-
cedures described by Civille and Szczesniak (1973) during 35–40 h (2 h per week)
to recognize texture attributes of hardness and crispness. The panel was composed
of nine panelists (three males, six females), all between the ages of 27–42. Standard
rating scales described by Hough et al. (1994) were used. The scale of crispness was
developed for this work using local food standards. The panel worked with the
definition of crispness and the measurement technique based on the work of
Chauvin et al. (2008). Standard samples were selected from available local food
products using the crispness scale described by Meilgaard et al. (1999) as a guide.
The samples were presented to the panelists in white plastic cups identified by
three digits chosen at random. Evaluations were performed in individual booths
under white light. The panelists were also provided with mineral water and unsalted
crackers to clean their palate between samples.
The variation of crispness with water content was modeled using the Fermi
equation, as described by Harris and Peleg (1996), and for hardness; after testing
various models, the best fit was found using the Lorentzian equation:
Sensorially and Instrumentally Detected Antiplastizicing Effect of Water. . . 129
y0 þ kx
y¼ xxc ð5Þ
1þe b
where x is the water content on a dry basis; y is the crispness value; y0 is the
crispness value at zero water content; k is the slope of the initial part of the curve;
and xc is the water content, where y ¼ y0 þkx
2
c
and b is related to the step shape at
water content near to xc.
Figure 1 shows the glass transition temperature (Tg), as function of solids content.
Tg decreased as water content increased, as expected for such systems. The water
content that corresponded to glass transition at 25 C was about 16 %db.
Figure 2 shows the T2 relaxation measured by 1H-TR-NMR. At water contents
below 6.5 (%db), only one proton population was found (indicated as T2SHahn),
which was related to solids and water closely interacting with them. At higher water
contents, a second proton population (T2LHahn) related to water molecules with
higher mobility, and therefore showing less interaction with solids, was found
(Chen et al. 1997; Choi and Kerr 2003). T2LHahn value increased as water content
increased, reflecting the increase in molecular mobility, and reached a plateau at
water content of approximately 14 %db. At water content of 25 %db, a third
population of protons was found. The Hahn spin echo sequence is limited in
measuring relaxation times due to the effect of diffusion. The CPMG sequence is
able to detect proton populations with higher mobility compared to the Hahn
sequence, but can only measure populations with relaxation times higher than the
first interpulse time value (τ) (Ruan and Chen 1998). Accordingly, the Hahn
200
150
temperature (°C)
100
50
-50
Fig. 1 Glass transition
temperature of cornflakes as
0.75 0.80 0.85 0.90 0.95 1.00
a function of solids fraction solids fraction
130 A.E. Farroni et al.
Fig. 2 NMR relaxation times for different proton populations as a function of water content.
( ) T2SHahn, ( ) T2LHahn, ( ) T2SCPMG, ( ) T2LCPMG
sequence was used to measure the T2 of protons related to solids and water
interacting with them, and CPMG was used to measure the higher mobility
populations in the range of 25–60 %db water content. At water contents higher
than 25 %db, three proton populations were present. T2SHahn, as already described;
T2SCPMG, which corresponds to the same population indicated as T2LHahn; and the
third population (T2LCPMG), which corresponds to more mobile protons. It should be
noted that at water contents lower than 25 %db, only one population was found
using the CPMG sequence (data not shown). T2SCPMG values remained almost
constant in the water content range 25–60 %db, while T2LCPMG value showed a
sharp increase, indicating the increment in mobility of the free water population. It
is important to note that the occurrence of freezable water found by DSC was
located at water content of 29 %db.
Figure 3 shows a graph of deformability modulus (Ed) vs. water content. Ed showed
a maximum at 13 %db, indicating the antiplasticizing effect of water. At low water
content, low force was required to deform cornflake samples. As water content
increased beyond 13 %db, Ed showed an important decrease, evidencing the
plasticizing effect of water.
Sensorially and Instrumentally Detected Antiplastizicing Effect of Water. . . 131
Ed (MPa)
to Lorentzian function; 0.4
(vertical line) standard R2: 0.6645
deviation. Best fit 0.3
parameters are indicated in
the graph 0.2
0.1
0.0
0 5 10 15 20 25
water content (%db)
Fig. 4 Sensory properties of cornflakes as a function of water content. (a) crispness; (b) hardness;
( filled circle) experimental measurements; (solid line) theoretical curves; (vertical line) standard
deviation. Best fit parameters are indicated in the graph
Figure 4a, b shows the results of oral textural measurements. Crispness value
showed a sharp decrease when water content reached a critical value
(14.0 0.3 %db) and at higher water contents decreased to zero. Hardness
value was low in low water content samples. Sensory hardness value increased as
water content increased and showed a maximum at a water content value of 17 %
db, indicating that antiplasticizing effect of water was evidenced in sensory prop-
erties. At water contents higher than 17 %, sensory hardness decreased.
I II III
arbitrary units
m0 Tg fw 30
0 10 20
water content (%db)
hardness crispness
Tg curve Ed
Fig. 5 Scheme of variation of sensory and mechanical properties of cornflakes along water
content scale. The three zones of water sorption isotherm are indicated with roman numerals.
mo ¼ hydration limit; Tg ¼ water content corresponding to glass transition at 25 C; fw ¼ frozen
water
related to the occurrence of frozen water by DSC (fw in the graph). As indicated
previously, at a water content value of 25 (%db), a third proton population was
evidenced by RMN, and at a wc value near 29 (%db), this water had mobility
enough to freeze. In region III water can act as a solvent and crystallize during
cooling to 50 C.
Maximum of sensory hardness was shifted to higher water content compared to
Ed maximum. Deformation speed in sensory and mechanical tests was not coinci-
dent, and this difference could affect the correlation between those parameters and
the critical water content at which transitions occurred (Castro-Prada et al. 2009). A
possible explanation for the antiplasticization effect could be that at very low water
content below Tg, vibrational movements cause an increase in energy dissipation
during deformation, fracture speed propagation decreases, and therefore, total
energy required for deformation increases; thus, an increase in required stress is
observed.
4 Conclusions
For the first time, the antiplasticizing effect at water content below 16 (%db) was
not only instrumentally but also sensorially detected and should be taken into
account when assessing organoleptic changes of low-moisture brittle food. The
maximum hardness detected by the trained panel occurred in the glass transition
region at water contents at which neither frozen water nor the 1H NMR long
relaxation time T2 was observed. In this way, the characterization of water effects
on several properties was useful to predict changes detected by consumers in stored
low water cereal-based products.
References
AACC (1995) Approved methods of the American Association of Cereal Chemists. 44-16. The
Association, St. Paul, MN
Castro-Prada EM, Primo-Martı́n C, Meinders MBJ, Hamer RJ, Van Vliet T (2009) Relationship
between water activity, deformation speed, and crispness characterization. J Texture Stud 40
(2):127–156
Chang YP, Cheah PB, Ceow CC (2000) Plasticizing-antiplasticizing effects of water on physical
properties of tapioca starch films in the glassy state. J Food Sci 65(3):445–451
Chauvin MA, Younce F, Ross C, Swanson B (2008) Standard scales for crispness, crackliness and
crunchiness in dry and wet foods: relationship with acoustical determinations. J Texture Stud
39:345–368
Chen PL, Long Z, Ruan RR, Labuza TP (1997) Nuclear magnetic resonance studies of water
mobility in bread during storage. Lebensm-Wiss Technol 30(2):178–183
Chinachoti P, Vittadini E, Chatakanonda P, Vodovotz Y (2006) Characterization of molecular
mobility in carbohydrate food systems by NMR. In: Webb Graham A (ed) Modern magnetic
resonance. Springer, Dordrecht, Netherlands, pp 1703–1712
134 A.E. Farroni et al.
Choi SG, Kerr WL (2003) 1H-NMR studies of molecular mobility in wheat starch. Food Res Int
36:341–348
Chung HJ, Lim ST (2004) Physical aging of glassy normal and waxy rice starches: thermal and
mechanical characterization. Carbohydrate Polym 57(1):15–21
Civille GV, Szczesniak AS (1973) Guidelines to training a texture profile panel. J Texture Stud
4:204–223
Farroni AE, Matiacevich SB, Guerrero S, Alzamora S, Buera MP (2008) Multi-level approach for
the analysis of water effects in corn flakes. J Agric Food Chem 56(15):6447–6453
Farroni AE, Matiacevich SB, Buera MP (2010) Thermal transitions and molecular mobility in corn
flakes as affected by water content. In: Reid DS, Sajjaanantakul T, Lillford PJ, Charoenrein S
(eds) Water properties in food, health, pharmaceutical and biological systems: ISOPOW 10.
Wiley-Blackwell, Ames, IA, pp 583–590
Fiszman S, Durán L (1997) Reologı́a de Solidos y Textura. In: Aguilera JM (ed) Temas en
Tecnologı́a de Alimentos. Instituto Politécnico Nacional, Direccion de Publicaciones y Mate-
rial Educativo, México DF, p 229
Fontanet I, Davidou S, Dacremont C, Le Meste M (1997) Effect of water on the mechanical
behaviour of extruded flat bread. J Cereal Sci 25:303–311
Greenspan L (1977) Humidity fixed points of binary saturated aqueous solutions. J Res Natl Bur
Stand 81:89–96
Harris M, Peleg M (1996) Patterns of textural changes in brittle cellular cereal foods caused by
moisture sorption. Cereal Chem 73:225–231
Hicsasmaz Z, Rizvi SSH (2005) Effect of size and shape on modulus of deformability. LWT Food
Sci Technol 38(4):431–435
Hough G, Contarini A, Mu~noz A (1994) Training a texture profile panel and constructing standard
rating scales in Argentina. J Texture Stud 25:45–57
Jacoby D, King C (2001) Sensory evaluation in snack foods development and production. In:
Lucas EW, Rooney LW (eds) Snack foods processing. CRC Press, London
Labuza T, Tannenbaum S, Karel M (1970) Water content and stability of low-moisture and
intermediate moisture foods. Food Technol 24:543–550
Lewicki PP (2004) Water as the determinant of food engineering properties. A review. J Food Eng
61(4):483–495
Meilgaard MC, Strauss S, Carr BT, Civille GV (1999) Sensory evaluation techniques. CRC Press
LLC., Boca Raton, FL
Mohsenin NN, Mittal JP (1977) Use of rheological terms and correlation of compatible measure-
ments in food texture research. J Texture Stud 8(4):395–408
Peleg M (1998) Mechanical properties of dry brittle cereal products. In: Reid E (ed) The properties
of water in foods. CRC Press, London, pp 233–253
Peng Y, Lu R (2005) Modeling multispectral scattering profiles for prediction of apple fruit
firmness. Trans ASAE 48:235–242
Reid DS, Fennema OR (2008) Water and ice. In: Damodaran S, Parkin KL, Fennema OR (eds)
Food chemistry. CRC Press, Boca Raton, FL, pp 18–77
Roos YH (2010) Glass transition temperature and its relevance in food processing. Annu Rev Food
Sci Technol 1(1):469–496
Roos YH, Karel M, Kokini JL (1996) Glass transitions in low moisture and frozen foods: effects on
shelf life and quality. Food Technol 50(11):95–108
Ruan RR, Chen PL (1998) Nuclear magnetic resonance techniques. In: Ruan RR, Chen PL (eds)
Water in foods and biological materials: a nuclear magnetic resonance approach. Technomic,
Lancaster, pp 17–24
Suwonsichon T, Peleg M (1998) Instrumental and sensory detection of simultaneous brittleness
loss and moisture toughening in three puffed cereals. J Texture Stud 29:255–274
Yang J, Huang M, Peng J, Shi J (2010) Rapid determination of the moisture content of milk
powder by microwave sensor. Measurement. doi:10.1016/j.measurement.2010.08.007
Characterization of a Hydrate–Dehydrate
System with Critical Transitions
in the Typical Range of Processing
and Storage Conditions
Abbreviations
A Hydrate species
A Isolated site hydrates
B Channel hydrates
DSC Differential scanning calorimetry
K Constant
Ksp Constant
RH Relative humidity
TGA Thermogravimetric analysis
W Water species
1 Introduction
Hydrates are solid adducts where the parent compound (in this case a drug
molecule) and water come together to form a crystalline lattice (Khankari and
Grant 1995). The incorporation of water molecule(s) into the crystalline structure
produces a new and distinct unit cell, i.e., with its own mechanical and thermody-
namic properties, which are often significantly different from those of the anhy-
drous crystal form of the drug. Figure 1 depicts the transformation of the anhydrous
crystal into the hydrate by the presence of water. Figure 1b shows the formation of
the hydrate as observed in the corresponding water uptake isotherm. Hydrate
formation involves an equilibrium analogous to that of the activity of the hydrogen
Fig. 1 Formation of a hydrate. (a) Pictorial description of the transformation from crystalline
anhydrous to crystalline hydrate. (b) Theoretical diagram of hydrate formation as a function of
water activity
ion on the protonation/deprotonation of acids and bases. In the case of the hydrate,
the water activity (estimated by means of the relative humidity, RH), results in two
regions. There is a critical RH value below which the anhydrous form is the stable
solid. The hydrate is the stable form when the water activity is above the critical RH
value. The formation of hydrates is a subject of considerable interest in the
pharmaceutical industry. Water is often the solvent of choice for some industrial
processes such as wet granulation and aqueous based film coating. The small size of
the water molecule, and its ability to act as both a hydrogen donor and acceptor, are
factors that promoting incorporation into crystalline lattices. Solid-state transitions
from hydration or dehydration are possible during processing steps such as drying,
mixing, granulation, or tableting (Jorgensen et al. 2002).
The changes produced upon dehydration can affect further processing and
functionality. For example, the greater number of intermolecular hydrogen bonds
in theophylline monohydrate, in relation to the anhydrous solid form, results in a
Characterization of a Hydrate–Dehydrate System with Critical Transitions in. . . 137
AWs Ð AL þ WL ð1Þ
½AL ½WL
K¼ ð2Þ
AWs
K sp ¼ ½AL ½WL ð3Þ
Where AWs, A, and W denote the hydrate, and the drug and water species,
respectively, and the subscripts S and L denote solid and liquid (solution) state,
respectively. Since the concentration (activity) of water is a factor in the solubility
product (Eq. 3), by Le Chatelier’s principle, the result is that by increasing the
activity of water in the solvent (up to the end point where pure water is the solvent),
the concentration of the drug in solution is “pushed down.” If a situation where the
hydrate is more soluble than the anhydrous form is encountered, the explanation is
likely to be found in the particle properties of the materials involved, such as quality
of the crystals involved. The vast majority of organic drug molecules exist in
different crystal forms, including anhydrous polymorphs and hydrates, each with
different solubility (Pudipeddi and Serajuddin 2005). The differences in solubility
of crystal forms affect their dissolution rate and potentially their bioavailability.
Ampicillin trihydrate for example, exhibits lower solubility and lower plasma
concentration values than its anhydrous counterpart (Poole and Bahal 1968). It is
important to emphasize that while the potential effect of solubility on bioavailabil-
ity applies to all different (anhydrous and hydrated) crystal forms, hydrates tend to
produce more consistent and significant reductions on observed blood levels during
product development. From a bioavailability standpoint, all crystal forms of the
drug are important, hydrates are especially so. In pharmaceutical development, the
anhydrous form of a drug, with its higher aqueous solubility relative to the hydrate,
is often desired. However, since water is ubiquitous in the atmosphere, hydrate
formation is a constant potential issue, the relevance of which extends beyond solid
form selection for pharmaceutical processing; it can affect handling, storage, and
stability of pharmaceuticals.
Hydrates are typically classified according to two general criteria. One classifi-
cation is based on whether the number of water molecules in the crystal maintains a
fixed ratio in relation to the drug molecules. Accordingly, hydrates can be stoichio-
metric or non-stoichiometric. Stoichiometric hydrates are composed of a fixed
number of water molecules per host molecule, at a given temperature, pressure
and relative humidity. In non-stoichiometric hydrates water composition varies
with water activity, without significant alteration of crystal structure, relative to
increasing or decreasing water content (anisotropic expansion and contraction is
commonly observed). These water molecules are usually not firmly bound. Instead,
they fill interstitial spaces in the unit cell without forming strong hydrogen bonds.
138 A.D. Otte and R. Pinal
Fig. 2 Energetics of hydration. (a) High energy of hydration observed in hydrates of the isolated
site class. (b) Low energy hydration observed in channel hydrates
Characterization of a Hydrate–Dehydrate System with Critical Transitions in. . . 139
molecules interact only with the major component (drug), and are isolated from
direct contact with other water molecules. The hydrate on Fig. 2b, on the other
hand, corresponds to a channel hydrate, where the water molecules are loosely held
inside tunnel-like structures created by the drug molecules. The two types of
hydrate exhibit substantially different behavior. The isolated site hydrate (Fig. 2a)
maintains its integrity as temperature increases. This hydrate does not begin to lose
its water until the temperature approaches 100 C, the boiling point of water. Notice
that the dehydration endotherm is observed at a temperature greater than 100 C.
The caged-in nature of the water entrapment in this type of hydrate results in strong
energy of hydration, requiring temperatures on the order of the boiling point of
water (or even higher) for dehydration. In contrast, it is clear that the water in the
hydrate of Fig. 2b is loosely bound. Notice that dehydration begins to take place at
temperatures slightly above room temperature. Furthermore, dehydration in this
case gradually increases with temperature such that it is complete at a temperature
below the boiling point of water. One very significant difference between Fig. 2a
and b is the absence of a dehydration endotherm in the latter. In channel hydrates,
water can be so loosely bound that it does not exhibit a clearly identifiable
calorimetric dehydration event. In this case, the energy of dehydration is contained
in the upward drift of the DSC baseline throughout the thermogram.
Figure 2 also reveals information regarding the stoichiometry of the two hydrates.
The isolated site hydrate (a) can be considered stoichiometric, since its water content
is fixed and maintained upon heating until forced dehydration takes place at a
sufficiently high temperature. The channel hydrate (b), on the other hand, readily
loses water with temperature, such that the there is no fixed stoichiometry between
water and drug molecules. It should be pointed out that temperature is not the
only factor affecting the level of hydration (or stoichiometry) in channel hydrates.
The relative humidity has a similar effect. Specifically, for a channel
(non-stoichiometric) hydrate, the corresponding vertical step illustrated in Fig. 1a
would not be as sharp and often presents significant hysteresis between sorption and
desorption.
Dehydration of a crystalline hydrate is a complex process, possibly resulting in
a number of intermediaries and final states. Dehydration can result in an anhydrate
form either via direct transformation or through an intermediate amorphous phase
(Allen et al. 1978). Water loss can result in a melt, an amorphous form or
a crystalline form of lower hydration. Dehydration is also a dynamic process,
which can trigger and be accompanied by liquid-solid and solid-solid
transformations. An illustrative example is a pentahydrate converting to anything
from a tetrahydrate to monohydrate (Saleki-Gerhardt et al. 1995; Zhu and Grant
2001; Lester et al. 2006). Desolvation can also result in a “dehydrated hydrate,”
an isomorphic desolvate where the crystal lattice is retained even with the
absence of water molecules (Stephenson et al. 1998). Hydrates can also exhibit
140 A.D. Otte and R. Pinal
The diagram above shows that transformations among the solid phases occur
under temperature conditions typical of normal handling and storage. A system like
this can be considered to present high risk for transformations during processing or
shelf life. Therefore, this system represents an example where establishing the
relationship among the different solid phases is a critical task of development.
The results of this type of investigation are discussed below.
Characterization of a Hydrate–Dehydrate System with Critical Transitions in. . . 141
Fig. 3 X-ray powder diffraction of the three crystalline forms obtained with the drug of the study
Initial characterization data indicates that any of the three solid phases involved can
transform into either of the other two forms through a rather finely balanced
interaction between temperature and relative humidity. These observations in turn
suggest that the approach to establishing the relationship among different crystal
forms should be a study that exploits the differences in the energetics of hydration.
Accordingly, the hygroscopicity of the solid forms was carried out as a function of
temperature. Figure 4 shows the water uptake isotherms of form A at different
temperatures. The isotherms show the formation of the hydrate form and hysteresis
with a degree that is temperature dependent. It is noteworthy that at 10 and 15 C,
high RH results in an additional hydration step that is not observed at higher (near
room temperature conditions) temperatures. Figure 5 shows the water sorption
isotherms of form A at 30 and 40 C. At these temperatures, there is no hydrate
formation, and the solid exhibits low (less than 1 % uptake) hygroscopicity. The
results in Figs. 4 and 5 indicate that hydration of form A is highly sensitive to
temperature. The hydrate is readily formed at 25 C, as seen in Fig. 4. However,
30 C is too high for hydrate formation to occur, as shown in Fig. 5. The water
uptake isotherms of form B at 25 C are shown in Fig. 6.
Hydration of form B is also highly sensitive to temperature. The hydrated form is
readily produced at 25 C. However, initial characterization results (see diagram
above) show that at 40 C and 75 % RH, form B converts to anhydrous form A.
To enable use of the results presented in the isotherms shown in Figs. 4, 5, and 6
so the information could be used as a guide for product development, the informa-
tion was synthesized in the form of a phase diagram. Figure 7 presents a phase
142 A.D. Otte and R. Pinal
Fig. 5 Water uptake isotherms of Form A at the relatively high temperatures of 30 and 40 C
diagram delineating the stability regions of different forms of the drug, as a function
of temperature and relative humidity. Dots on the figure correspond to experimental
observations, whereas solid and dashed boundaries are interpolations and extrapo-
lations, respectively.
Characterization of a Hydrate–Dehydrate System with Critical Transitions in. . . 143
Fig. 7 Phase diagram of the three isolated crystalline forms of the drug of the study
Fig. 8 Pictorial representation of the relationship among the different solid forms represented in
the phase diagram of Fig. 7
Fig. 9 X-ray powder diffraction patterns of the final formulation of the drug in the study and the
placebo. Diffractograms for pure forms A and C are shown for comparison
Fig. 10 Change in X-ray pattern observed upon exposing the final formulation to high humidity
and low temperature (10 C) in order to induce the formation of Form C in the product
Table 1 Different crystal form dispositions in a hypothetical wet granulation, along with the
feasible events the different dispositions reflect
Granulation
containing Reflects
Form A + Form B A non-homogeneous wetting of the particles or A non-homogeneous
drying of the granules
Form I + Form X Water content very nonuniform in the granulation
Form II + Form X High and nonuniform water content in the granulation
Form I + Form II Non-intimate wetting of the particles, nonuniform drying of the granules,
+ Form X highly nonuniform final water content
to X-ray diffraction analysis. The results are shown in Fig. 10. Exposing the sample
to high humidity and low temperature, which induces the formation of the hydrate
(Form C) resulted in the virtual disappearance of the peak at 3 in the formulation.
This result indicates that the crystal form in the formulation is the anhydrous Form A.
One feature of systems in which it is possible to isolate a kinetically stable
intermediary solid such as Form B, is that the properties of the material are
dependent on the sample history. This feature therefore has potential use as a
process monitoring/diagnostic tool. Table 1 presents different crystal form dispo-
sitions in a hypothetical wet granulation, along with the feasible events the different
dispositions reflect.
Characterization of a Hydrate–Dehydrate System with Critical Transitions in. . . 147
4 Conclusions
References
Agbada CO, York P (1994) Dehydration of theophylline monohydrate powder: effects of particle-
size and sample weight. Int J Pharm 106:33–40
Allen PV, Rahn PD, Sarapu AC, Vanderwielen AJ (1978) Physical characterization of erythro-
mycin: anhydrate, monohydrate, and dihydrate crystalline solids. J Pharm Sci 67:1087–1093
Jorgensen A, Rantanen J, Karjalainen M, Khriachtchev L, Rasanen E, Yliruusi J (2002) Hydrate
formation during wet granulation studied by spectroscopic methods and multivariate analysis.
Pharm Res 19:1285–1291
Khankari RK, Grant DJW (1995) Pharmaceutical hydrates. Thermochim Acta 248:61–79
Lester C, Lubey G, Dicks M, Andol G, Vaughn D, Cambron RT, Poiesz K, Redman-Furey N
(2006) Dehydration of risedronate hemi-pentahydrate: analytical and physical characteriza-
tion. J Pharm Sci 95:2631–2644
McMahon LE, Timmins P, Williams AC, York P (1996) Characterization of dihydrates prepared
from carbamazepine polymorphs. J Pharm Sci 85:1064–1069
Morris KR (1999) Structural aspects of hydrates and solvates. In: Brittain HG (ed) Polymorphism
in pharmaceutical solids. Marcel Dekker, New York, NY, pp 126–179
Poole JW, Bahal CK (1968) Dissolution behavior and solubility of anhydrous and trihydrate forms
of ampicillin. J Pharm Sci 57:1945–1948
Pudipeddi M, Serajuddin ATM (2005) Trends in solubility of polymorphs. J Pharm Sci
94:929–938
Saleki-Gerhardt A, Stowell JG, Byrn SR, Zografi G (1995) Hydration and dehydration of crystal-
line and amorphous forms of raffinose. J Pharm Sci 84:318–323
Steiner T, Koellner G (1994) Crystalline β-cyclodextrin hydrate at various humidities: fast,
continuous, and reversible dehydration studied by X-ray diffraction. J Am Chem Soc
116:5122–5128
Stephenson GA, Groleau EG, Kleemann RL, Xu W, Rigsbee DR (1998) Formation of isomorphic
desolvates: creating a molecular vacuum. J Pharm Sci 87:536–542
Zhu HJ, Grant DJW (2001) Dehydration behavior of nedocromil magnesium pentahydrate. Int J
Pharm 215:251–262
Viscoelastic Sorption Behavior of Starch
and Gluten
Abbreviations
1 Introduction
The migration of gasses and liquids through (bio) polymers plays an important role
in numerous applications and processes like, e.g., drying and (re)wetting of foods
and food ingredients, packaging, and controlled release. A good understanding of
the transport mechanisms is therefore of great importance to control these pro-
cesses. The transport of solvents through (bio) polymers is a complex process and
involves diffusion due to water concentration gradients, swelling, shrinkage, and
relaxation of the polymeric matrix. Often also transitions between the glassy,
rubbery, and/or viscous states of the polymer system are involved.
Here we report on a theoretical and experimental study of water transport in
biopolymer films of starch and gluten. Relaxation effects of the polymer matrix
seem important for the migration of water in biopolymer systems like starch and
gluten (Oliver and Meinders 2011). This seems the case for a broad water activity
range, but especially when the system is in the neighborhood of the glass transition.
There are various models published that take into account the viscoelastic relaxa-
tion effects of the polymeric matrix (Wu and Peppas 1993; Vrentas and Vrentas
2001; Pawan et al. 2004).
These models are solved using as boundary condition a step change of the water
volume fraction at the sample boundary. Results indicate three regimes, depending
on the ratio of the diffusion time τD ¼ l2/D (with l the size of the sample and D the
diffusion coefficient) and matrix relaxation time τ1. This ratio is often called the
diffusion or sorption Deborah number and is defined as De ¼ τ1/τD. When De 1
(τ1 τD), the system behaves as an elastic solid, there is negligible time variation
of the polymer structure during the diffusion process, and water migration can be
described by Fickian diffusion (elastic diffusion). Similarly, when De 0
(τ1 τD), the system behaves as a viscous fluid and water migration can also be
described by Fickian diffusion (viscous diffusion). When De 1 (τ1 τD), the
rearrangement of polymer chains and the movement of solvent occur in comparable
Viscoelastic Sorption Behavior of Starch and Gluten 151
time scales, so the observed behavior is not consistent with the classical diffusion
theory (viscoelastic or anomalous diffusion).
We observed, however, that sorption curves of amorphous starch and gluten
showed non-Fickian features with characteristic relaxation times that are signifi-
cantly different than the characteristic diffusion time. This seems not in line with
the published results mentioned above. To understand and describe this behavior,
we extended the viscoelastic diffusion model and allowed variation of the water
volume fraction and stress of the sample boundary. The model attempts to describe
both the dynamical sorption behavior and the equilibrium values (isotherm) in a
single set of physical parameters of the polymer-water system: the Flory-Huggins
interaction parameter, the diffusion constant, and a concise set of rheological
parameters to describe the viscoelastic behavior of the polymer.
3 Model
Various models have been published that take into account the viscoelastic relax-
ation effects of the polymeric matrix (Wu and Peppas 1993; Vrentas and Vrentas
2001; Pawan et al. 2004); the models are all based on the assumption that the free
energy (and chemical potential) can be described as the sum of a solvent-polymer
mixing term and a stress term. Here we will discuss shortly the model of Singh
et al. (Pawan et al. 2004) that will be used as a starting point to describe the water
migration in biopolymers. Their model was derived using multiscale thermody-
namical theory for the swelling of biopolymers by modeling the solid phase as
152 M.B.J. Meinders and L. Oliver
viscoelastic and the fluid phase as viscous. For a one-dimensional sheet the water
transport equation can be expressed as
∂ϕw ∂ ∂ϕ ∂σ
¼ ð 1 ϕw Þ 2 Dð1 ϕw Þ w Bc ð1 ϕw Þ ð1Þ
∂t ∂X ∂X ∂X
where ϕw is the water volume fraction, ∂/∂t is the partial time derivative, D is the
water diffusion coefficient, Bc is a material constant, σ ¼ σ(ϕw,t) is the stress in the
system or osmotic pressure. The model is expressed in terms of the spatial
Lagrangian coordinate X in order to account for the swelling and shrinkage of the
polymeric material upon water adsorption and desorption, respectively. The
Lagrangian coordinate system is chosen to be that of the dry polymer material.
If we assume incompressibility, then the relation between the Eulerian and
Lagrangian coordinates x and X, respectively, can be approximated for a 1D
sheet by
∂X ¼ ð1 ϕw Þ∂X ð2Þ
The first term in Eq. (1) corresponds to Fickian diffusion while the second term
describes the matrix relaxation effects.
Pw , o
μw:o ¼ Rg Tlnαw, o ¼ Rg Tln ð3Þ
P
where aw,o and Pw,o/P are the water activity and the water partial pressure or
relative humidity (RH) of the environment, and Rg is the universal gas constant.
To describe the water chemical potential of the water-polymer system, which is
directly related to the isotherm, the approach of Leibler and Sekimoto (Leibler and
Sekimoto 1993) is followed. They considered the mixing of a solvent and polymer.
It is assumed that the chemical potential can be written as the sum of the chemical
potential due to the mixing, μw,FH and due to the stress in the system μw,σ
Viscoelastic Sorption Behavior of Starch and Gluten 153
The chemical potential due to the mixing of the polymer and the water (μw,FH) is
assessed by the Flory-Huggins (FH) theory for large polymers (like starch and
gluten)
μw, FH
¼ lnϕw þ ð1 ϕw Þ þ χ ð1 ϕw Þ2 ð5Þ
Rg T
dμw dμw, FH dσ
¼ V^w ð7Þ
dt dt dt
Together with the appropriate rheological model that gives an expression for dσ/dt
and
154 M.B.J. Meinders and L. Oliver
!
dμw, FH 1 1 dϕ p
¼ Rg T þ 1 þ 2χϕ p ð8Þ
dt ϕp 1 N dt
Many systems do not relax with a single relaxation time. For example, different
polymer segments of a polymer of different sizes and properties can contribute to
the relaxation process. The system cannot well be described by a single relaxation
time but a distribution of various relaxation times is needed. For polymer-water
mixtures often the stretched exponential is used to characterize the relaxation
function (Ediger et al. 1996; Le Meste et al. 2002). To describe the sorption
behavior of starch and gluten, we observed that more than one relaxation mode is
needed. We approximate their relaxation behavior by the so-called Wiechert model
with at most two Maxwell elements placed in parallel with a Hookean spring. The
constitutive relations are then given by
∂σ ∂ε 1 1 σ ke ε
¼ ðk e þ k 1 þ k 2 Þ σ1 þ ð9Þ
∂t ∂t τ1 τ2 τ2 τ2
∂σ 1 ∂ε σ 1
¼ k1 ð10Þ
∂t ∂t τ1
where ke, k1, and k2 are the spring constants; τi ¼ ki/ηi and ηi are the relaxation times
and viscosity of the two Maxwell elements; and ε is the strain. The latter can be
defined in various ways; here we used ε ¼ (ϕw ϕw,g)ϕw. The Wiechert model
reduces for k2 ¼ 0 to a Standard Linear Solid (SLS), ke ¼ k2 ¼ 0 to a Maxwell, and
for k1 ¼ k2 ¼ 0 to a fully elastic system.
This set of ODEs can be solved giving the appropriate initial values for the state
variable σ, σ 1, and ε.
The coupled set of partial differential equation (PDE), Eqs. (1) and (9), and
applying appropriate initial and boundary conditions can be solved using standard
PDE solvers. The model calculations and fitting with experimental data have been
performed using Matlab (Mathworks).
Viscoelastic Sorption Behavior of Starch and Gluten 155
4.1 Isotherms
Figure 1 shows the measured isotherms of gluten and starch films for different
thicknesses and duplicates. Also shown are fits of the Flory-Huggins model includ-
ing an extra stress term for the chemical potential in the glassy state (Eq. 4). Model
parameters are shown in Table 1 and Fig. 3. An elastic model is used to describe the
stress in the glassy state as a function of the volume fraction. It is seen that the
model seems to describe the measured isotherms of starch and gluten well with
reasonable parameters that are similar as found in literature (Oliver and Meinders
2011; Nicholls et al. 1995). The obtained values for the elastic modulus (about
5 108 N/m2) seem lower as those found in literature (about between 1 and
2 109 N/m2 (Nicholls et al. 1995; Chang et al. 2000)). The isotherm measure-
ments of the same material but differing in film thickness show a large variation.
We think that an important reason for this is that the samples have different
histories, like, e.g., different drying and cooling rates. When the water-polymer
system is going from the supercooled liquid to the glassy state, the molecular
motion of the polymer will be frozen. This also involves that stress will be built
up in the glassy system upon further cooling and/or drying. Below the glass
transition short-range molecular motions, involving water, side chains, and even
main chains, are still occurring (see, e.g., Ediger et al. 1996; Le Meste et al. 2002).
The precise process of immobilization of the polymer matrix occurs over a broad
temperature and/or water content range and depends on the history. This makes that
a number of properties of the glass like Tg, aw,g, isotherm, and relaxation phenom-
ena depend on the history of the sample.
1 1
Starch Gluten
0.8 0.8
0.6 0.6
aw
aw
0.4 0.4
0.2 0.2
0 0
0 0.1 0.2 0.3 0.4 0 0.1 0.2 0.3 0.4
ϕw ϕw
Fig. 1 Experimental isotherm of starch (left panel) and gluten films (right panel). Different
symbols indicate different film thicknesses. Also shown are the simulated isotherms (Eq. 4). The
dotted line corresponds to the FH isotherm (Eq. 5), while the continuous line is obtained by adding
the stress term (Eq. 6) for an elastic glassy state. Model parameters are shown in Table 1
156 M.B.J. Meinders and L. Oliver
Table 1 Parameters used to describe the sorption behavior of starch and gluten
Starch Gluten
ϕw,g 0.23 0.16 Glass transition water volume fraction
χ 0.9 1 Flory-Huggins interaction parameter
ke 5 108 6 108 N/m3 Spring constant
ρw 1,000 kg/m3 Density of the water
ρp 1,450 kg/m3 Density of the polymer
Mw 18 103 kg/mol Molecular mass water
Mp 1 1010 kg/mol Molecular mass polymer
T 298 K Temperature
Rg 8.31 J/K/mol Universal gas constant
Fig. 2 Typical examples of dynamical sorption curves of gluten (top panels) and starch (bottom
panels) films. Gluten and starch films are 75 μm and 104 μm thick, respectively. RH steps are
indicated. The figures show the water volume fraction φw, as a function of dimensionless time
θ ¼ t/0.11 h1. The dots are experimental data and the red lines correspond to the best fit of the
viscoelastic diffusion model. The blue line (only top left panel) corresponds to a best fit of the
Fickian diffusion model. The insets (top panels) show the evolution of the water volume fraction at
the boundary
-12 10 6
10 10 10
-13
10 9
10
5
10
-14
D , Bc [m2/s]
10
K [N/m 2]
8 ke Gluten
t [s]
10
k2 Gluten
10
-15
τ1 Gluten
k3 Gluten 4
D Gluten 10 τ2 Gluten
10
7 ke Starch
Bc Gluten
10
-16
k1 Starch
τ1 Starch
D Starch
Bc Starch k2 Starch τ2 Starch
-17 6 3
10 10 10
0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8
aw aw aw
Fig. 3 Values of the model parameters describing the sorption behavior of starch and gluten.
Other parameters are shown in Table 1
158 M.B.J. Meinders and L. Oliver
activities. Moreover, the glass transition is associated with motions of the polymer
backbone (especially measured with DSC), but also other molecular motions may
occur that affect water sorption.
Furthermore, this viscoelastic diffusion model can also describe the hysteresis
as always observed between adsorption and desorption (Meinders and van Vliet
2009). Different histories of the samples will cause that different stresses are
present in the sample so that a different isotherm and dynamical sorption behavior
is observed. For a desorption curve, which is measured shortly after the adsorption
isotherm, there is ample time for the system to relax the stresses that are built up
when going through the glass transition. This causes that at the same aw and in the
glassy state the water volume fraction of the desorption isotherm will be larger than
that of the adsorption isotherm.
5 Conclusions
A study to the dynamical and equilibrium sorption behavior of starch and gluten
films has been performed. A viscoelastic diffusion model is proposed to describe
the sorption behavior of these biopolymer films for the whole range of water
activities. Dynamical sorption curves of starch and gluten measured at different
water activities (ranging from 0.1 to 0.8) show that matrix relaxations play an
important role in the water sorption process, also far below the glass transition. The
model attempts to describe both the dynamical sorption behavior and the equilib-
rium values (isotherm) in a single set of physical parameters of the polymer, the
Flory-Huggins interaction parameter, the diffusion constant, and a concise set of
rheological parameters to describe the viscoelastic behavior of the polymer. It turns
out that the isotherms and all dynamical sorption curves can be very well described
by the model. However, the sorption behavior for all water activities could not be
described by only one single consistent concise set of parameters. This is believed
to be due to the fact that only one single aw,g is assumed for a compound. In practice
however, the glass transition seems to occur over a broad range of water activities
(at the same temperature). When the water-polymer system is going from the
supercooled liquid to the glassy state, the molecular motion of the polymer will
be frozen. This also involves that stress will be built up in the glassy system upon
further cooling and/or drying. Below the glass transition polymer relaxation con-
tinues. The precise process of immobilization and relaxation of the polymer occurs
over a broad temperature and/or water content range. Furthermore it will depend on
the history, like cooling and/or drying rate. This makes that a number of properties
of the glass like Tg, aw.g, isotherm, and relaxation phenomena depend on the history
of the sample. This viscoelastic diffusion model can therefore also explain and
describe the hysteresis that is often observed between adsorption and desorption.
Viscoelastic Sorption Behavior of Starch and Gluten 159
References
Chang YP, Cheah PB, Seow CC (2000) Plasticizing-antiplasticizing effects of water on physical
properties of tapioca starch films in the glassy state. J Food Sci 65(3):445–451
Ediger MD, Angell CA, Nagel SR (1996) Supercooled liquids and glasses. J Phys Chem 100
(31):13200–13212
Le Meste M, Champion D, Roudaut G, Simaktos D (2002) Glass transition and food technology: a
critical appraisal. J Food Sci 67(7):2444–2458
Leibler L, Sekimoto K (1993) On the sorption of gases and liquids in glassy polymers. Macro-
molecules 26(25):6937–6939
Meinders MBJ, van Vliet T (2009) Modeling water sorption dynamics of cellular solid food
systems using free volume theory. Food Hydrocoll 23:2234–2242
Nicholls RJ, Appelqvist IAM, Davies AP, Ingman SJ, Lillford PJ (1995) Glass transitions and the
fracture behaviour of gluten and starches within the glassy state. J Cereal Sci 21(1):25–36
Oliver L, Meinders MBJ (2011) Dynamic water vapour sorption in gluten and starch films.
J Cereal Sci 54:409
Pawan PS, Maier DE, Cushman JH, Haghighi K, Corvalan C (2004) Effect of viscoelastic
relaxation on moisture transport in foods. Part I: solution of general transport equation.
J Math Biol 49(1):1–19
Vrentas JS, Vrentas CM (2001) Viscoelastic diffusion. J Polym Sci B 39:1529–1547
Wu JC, Peppas NA (1993) Modeling of penetrant diffusion in glassy polymers with an integral
sorption Deborah number. J Polym Sci B 31(11):1503–1518
Molecular Weight Effects on Enthalpy
Relaxation and Fragility of Amorphous
Carbohydrates
Abbreviations
1 Introduction
relaxation during the aging time (t), respectively; ΔCp is a heat capacity change at
the glass transition temperature (Tg) and aging temperature (Ta).
The fragility concept was introduced by Angell (1985), which has practical
implications in predicting the physical and chemical stability of foods. The fragility
concept differentiates the amorphous systems as being either strong or fragile,
based on the temperature dependency of dynamic properties, structural changes,
and relaxation behavior near or above Tg. Differences between fragile and strong
systems are presented in Table 1. A fragility index (m) was introduced to better
quantify the fragility concept (Angell 1985). The fragility index is defined as:
ΔE
m¼ ð4Þ
2:303RT g
where ΔE is the activation energy for structural relaxations at Tg, and R is the gas
constant.
The objective of this study was to investigate the effects of molecular weight on
enthalpy relaxation and the fragility of amorphous carbohydrates. Anhydrous
glucose, maltose, and maltotriose were chosen to characterize amorphous carbo-
hydrates with increasing molecular weight.
Anhydrous glucose, maltose, and maltotriose were purchased from Fisher Scientific
(Pittsburg, PA). Glucose, maltose, and maltotriose solutions were prepared by
dissolving them in distilled water (10 g carbohydrates/100 g water). The carbohy-
drate solutions were dried in a laboratory freeze dryer (VirTis Freeze Mobile
24 with Unitop 600L, Virtis SP Industries Co., New York). Initially the carbohy-
drate solutions were frozen to 35 C. The frozen carbohydrate solutions were
transferred to a freeze dryer shelf maintained at 25 C. The shelf temperature was
increased to 20 C with a vacuum of 20 Pa and condenser temperature at 60 C.
After 48 h of freeze drying, glucose, maltose, and maltotriose were kept in a
164 R.M. Syamaladevi et al.
desiccator with P2O5 at room temperature (298 K) for 2 days. The moisture content
of dry amorphous carbohydrates was analyzed using the vacuum oven method
(~0.005 kg H2O/kg sample).
Temperature
Tg + 100°C
n
3°C/
/m i
Unaged experiment
3°C
min
Tg = Ta + 5°C
Ta
3°C/
n
/mi
Aged
min
3°C
experiment
25°C
Tg - 50°C
Time
ta
Fig. 1 Enthalpy relaxation experimental protocol for amorphous carbohydrates. Aged experi-
ment: Glucose, maltose, and maltotriose were scanned above the respective melting points of the
selected carbohydrates (~Tg + 100 C) to remove their thermal history after equilibration at 25 C
for 5 min. Then the carbohydrates were scanned to (Tgi 50) at 3 C/min to make a carbohydrate
glass. The carbohydrates were scanned to (Tgi 5) and annealed isothermally at (Tgi 5) for
selected periods of time (0.5–48 h). After aging, samples were cooled to (Tgi 50). After
equilibrating the carbohydrates for 1 min, they were reheated to (Tgi + 50) at a rate of 3 C/min.
Unaged experiment: Glucose, maltose, and maltotriose were scanned above the respective melting
points of the selected carbohydrates (Tg + 100 C) to remove their thermal history after equilibra-
tion at 25 C for 5 min. Then the carbohydrates were scanned to (Tgi 50) at 3 C/min to make a
carbohydrate glass. Then the samples were reheated to (Tgi + 50) at a rate of 3 C/min
width approach by identifying the onset (Tgi), end point glass transition tempera-
tures (Tge), and ΔCp.
The fragility of amorphous carbohydrates was determined by calculating the
activation enthalpy of structural relaxation (Moynihan 1993). Activation enthalpy
and glass transition width are related by:
ΔE 1 1
¼ Constant ð5Þ
R T gi T ge
where ΔE is the activation energy for structural relaxations at the glass transition
temperature and Tgi and Tge are the onset and end point glass transition tempera-
tures. By studying several material systems, the constant in the Eq. (5) was
identified as 5 0.5. In the current study, the value of the constant was selected
as 5. The activation enthalpy and fragility index (m) are related in Eq. (5) (Crowley
and Zografi 2001).
166 R.M. Syamaladevi et al.
Fig. 2 Identification of
onset and end point glass
Tgi
transition temperatures (Tgi,
and Tge respectively) and
enthalpy recovered (ΔH )
ΔH Nonreversible
heat flow
Temperature
Statistical significance of the results was analyzed using SAS 9.1 (SAS Institute,
Inc., Cary, NC, USA). A value of P
0.05 was selected as statistically significant
using the least square difference (LSD) method.
Molecular weights of glucose, maltose, and maltotriose were 180 g/mol, 342 g/mol,
and 504 g/mol, respectively. The obtained Tgi of unaged glucose, maltose, and
maltotriose was 34.5 0.20 C, 78.5 1.75 C, and 128.8 4.13 C, respectively.
Generally, no significant difference was observed between the Tgi of unaged and
aged carbohydrates, with the exception of maltotriose (Table 2). Wungtanagorn and
Schmidt (2001) mentioned that the increase or decrease of Tg with aging time is
dependent on the method used in the assignment of Tg. No statistically significant
difference in Tgi was observed for glucose aged for selected aging periods (0–48 h).
Significant difference in Tgi was observed between maltose aged for 12 and 48 h
(Table 2). The Tgi of maltotriose aged for 48 h was significantly different from
maltotriose aged for 0.5–12 h (Table 2). During the determination of Tgi, carbohy-
drates were heated above their melting points using MDSC, which could result in
thermal decomposition, thus influencing Tgi values (Lee et al. 2011a, b). Glass
transition temperature width (Tge Tgi) of glucose did not change significantly
during isothermal aging at (Tgi 5) for 0.5–48 h (Table 2). The (Tge Tgi) of
Molecular Weight Effects on Enthalpy Relaxation and Fragility. . . 167
Table 2 Onset (Tgi) glass transition temperatures of glucose, maltose, and maltotriose aged at
(Tgi 5) for selected times
Glucose Maltose Maltotriose
Time t, (Tge Tgi) (Tge Tgi) (Tge Tgi)
(h) Tgi ( C) ( C) Tgi ( C) ( C) Tgi ( C) ( C)
0 34.5e 6.3h,i (0.29) 78.5c,d 10.2g,h 128.8a 11.4f,g,h
(0.20)* (1.75) (1.18) (4.13) (3.09)
0.5 36.5e 5.0i (0.14) 76.8c,d 9.2g,h,i 131.7a 7.7h,i (0.71)
(0.50) (4.90) (0.59) (0.14)
1 36.2e 4.6i (0.23) 76.3c,d 10.2g,h 131.6a 7.8h,i (1.13)
(0.35) (3.70) (2.52) (4.51)
2 36.3e 5.0i (0.06) 79.1c,d 8.1h,i (0.13) 133.4a 6.3h,i (0.35)
(0.07) (1.90) (1.47)
6 36.4e 4.9i (0.05) 76.0c,d 11.0f,g,h 132.4a 7.9h,i (0)
(0.07) (2.60) (0.25) (4.20)
12 36.6e (0) 5.1i (0) 81.6c (0) 9.8g,h,i 132.7a 8.1h,i (0)
(0.42) (3.79)
48 36.1e (0) 5.2i (0) 72.7d (0) 13.8f,g (0) 113.7b (0) 15.4f (0)
Same superscripts within a column indicate there is no significant difference (P 0.05) in values
*Values in parentheses represent the standard deviation of at least two replicates
unaged glucose was significantly different from aged glucose (0.5–48 h). The
(Tge Tgi) of 2 h aged maltose was significantly different from maltotriose and
glucose (Table 2). The (Tge Tgi) of maltotriose aged for 48 h was significantly
different from maltose and glucose of aged maltotriose (Table 2).
The heat capacity change (ΔCp) at Tg of glucose, maltose, and maltotriose was
measured as the difference in Cp values between glassy and rubbery states in the
MDSC thermogram. The heat capacity change (ΔCp) at Tg of maltotriose was less
than that of other selected carbohydrates, which can be attributed to its higher chain
rigidity and molecular weight (McGonigle et al. 2005) (Fig. 3). A slight increase in
ΔCp at Tg with aging time was observed for maltose and maltotriose, while a
fluctuating trend in ΔCp was observed for glucose (Fig. 3). Since the equilibrium
state has a smaller heat capacity and free energy compared to glassy and rubbery
states, an increase in ΔCp is expected after aging (Chung and Lim 2003). In the
current study, the difference in the trends in ΔCp with aging time could be due to
thermal decomposition resulting from heating the carbohydrates above their melt-
ing points using the MDSC (Lee et al. 2011a, b).
Total enthalpy decreased, while relaxed enthalpy increased with increasing aging
time for glucose, maltose, and maltotriose (Fig. 4). The rate of enthalpy relaxation
168 R.M. Syamaladevi et al.
Fig. 4 Amount of
relaxation enthalpy after
aging of glucose, maltose,
and maltotriose at (Tg 5)
for selected aging times
Fig. 5 Variation of
(1 ΔHrelax/ΔH1) for
selected aging times during
aging of glucose, maltose,
and maltotriose at (Tgi 5)
170 R.M. Syamaladevi et al.
Table 3 Experimental values of ΔH1, τ, β, τφ(t)¼50%, and τφ(t)¼1% calculated for glucose,
maltose, and maltotriose
KWW equation
Total Relaxation time Relaxation time
enthalpy corresponding to corresponding to
available Relaxation 50 % of the 99 % of the
for Characteristic distribution maximum maximum
Amorphous relaxation, relaxation parameter, enthalpy, τφ(t)¼50% enthalpy, τφ(t)¼1%
system ΔH1 (J/g) time, τ (h) β (h) (h)
Glucose 3.94 16.9 0.23 50.9 338.4
Maltose 3.23 38.4 0.33 80.7 535.9
Maltotriose 2.54 128.6 0.24 371.4 2,467.6
100
10
1
100 200 300 400 500 600
Molecular weight (g/mol)
ΔH1 decreased with molecular weight for the selected carbohydrates (Table 3).
The greater ΔH1 of glucose is attributed to its higher ΔCp at Tg. Relaxation time
corresponding to 50 % of the maximum enthalpy (τφ(t)¼50%) and 99 % of the
maximum enthalpy (τφ(t)¼99%) during the aging process is determined at (Tgi 5)
from τ and β values (Liu et al. 2007). The τφ(t)¼50% values were 50.9 h, 80.7 h, and
371.4 h, while τφ(t)¼99% values were 338.4 h, 535.9 h, and 2,467.6 h for glucose,
maltose, and maltotriose, respectively.
Cp(glass) ratio (Table 4). Glucose, maltose, and maltotriose exhibited a small ΔCp at
Tg and are categorized as fragile. A greater ΔCp at Tg for maltotriose demonstrates
its stronger nature in comparison to glucose and fructose. One possible explanation
for this discrepancy in the fragility of glucose, maltose, and maltotriose could be the
thermal decomposition of carbohydrates to smaller components resulting by
heating them above their melting points, which may have influenced the fragility
determination (Lee et al. 2011a, b). Previous literature presents inconsistencies in
fragility predictions by different methods for various systems. No clear explanation
is available for the difference in fragility predictions by different methods
(Wungtanagorn and Schmidt 2001). Thus, more research is required for clarifica-
tion of this discrepancy.
The physical, chemical, and structural stability of amorphous food systems is
related to their fragility, as this will provide information on the variation in dynamic
properties and relaxation times of amorphous systems near or above Tg (Hancock
et al. 1995). Above Tg, strong systems exhibit greater relaxation times, while below
Tg a reversal of dynamic properties and relaxation times occurs (Mao et al. 2010).
Many physicochemical degradation reactions can be controlled by molecular mobil-
ity/relaxations in food systems where relaxation times provide useful information on
the rates of these reactions. By relating fragility to the relaxation times, the temper-
ature dependence of relaxations and thereby the rates of degradation reactions in
amorphous systems can be predicted (Hatley 1997). For instance, the rate of molec-
ular mobility of fragile systems is changed by approximately one order of magnitude
for every 10 C difference in temperature, whereas in strong systems, the rate of
molecular mobility changes approximately every 25 C (Hancock et al. 1995). Hatley
(1997) reported that fragile systems may be preferable in their glassy state, since
dynamic properties such as viscosity increase rapidly with small decreases in storage
temperature, which could increase the stability of glassy systems.
The enthalpy relaxation data of the current study can be used to predict the
enthalpy relaxation behavior of selected amorphous carbohydrates with molecular
weight between glucose and maltotriose. Also, the estimated relaxation times can
be used to predict the physicochemical stability of amorphous carbohydrates, which
could also reduce the cost and time of complex experiments. However, a relation-
ship between relaxation times and rates of physicochemical degradation reactions
should be established for such a comparison (Miller and Lechuga-Ballesteros
2006). Thus, enthalpy relaxation time and fragility information obtained may be
used to predict various physicochemical changes in selected carbohydrates during
their long-term glassy state storage.
4 Conclusions
glucose, maltose, and maltotriose. The enthalpy relaxation data of glucose, maltose,
and maltotriose might be used to predict the enthalpy relaxation behavior of specific
molecular weight materials. The fragility indices (m) determined using the glass
transition width approach presented fragile behavior for glucose, while maltose and
maltotriose were comparatively stronger in nature. A discrepancy in the prediction
of fragility of selected amorphous carbohydrates was observed in the research and
requires further study.
References
Angell CA (1985) Spectroscopy simulation and scattering, and the medium range order problem in
glass. J Non Cryst Solids 73:1–17
Arrighi V, Cowie JMG, Ferguson R, McEwen IJ, McGonigle EA, Pethrick RA, Princi E (2005)
Physical aging in poly(4-hydroxy styrene)/poly(vinyl methyl ether) blends. Polym Int
55:749–756
Christensen KL, Pedersen GP, Kristensen HG (2002) Physical stability of redispersible dry
emulsions containing amorphous sucrose. Eur J Pharm Biopharm 53(2):147–153
Chung HJ, Lim ST (2003) Physical aging of glassy normal and waxy rice starches: effect of aging
time on glass transition and enthalpy relaxation. Food Hydrocoll 17(6):855–861
Crowley KJ, Zografi G (2001) The use of thermal methods for predicting glass former fragility.
Thermochim Acta 380(2):79–93
Farahnaky A, Guerrero A, Hill SE, Mitchell JR (2008) Physical aging of cray fish flour at low
moisture contents. J Therm Anal Calorim 93(2):595–598
Gupta P, Kakumanu VK, Bansal AK (2004) Stability and solubility of Celecoxib-PVP amorphous
dispersions: a molecular perspective. Pharm Res 21(10):1762–1969
Hancock BC, Shamblin SL, Zografi G (1995) Molecular mobility of amorphous pharmaceutical
solids below their glass transition temperatures. Pharm Res 12(6):799–806
Hatley RHM (1997) Glass fragility and stability of pharmaceutical preparations-Excipient selec-
tion. Pharm Dev Tech 2(3):257–264
Lammert AM, Lammert RM, Schmidt SJ (1999) Physical aging of maltose glasses as measured by
standard and modulated differential scanning calorimetry. J Therm Anal Calorim 55:949–975
Lee JW, Thomas LC, Jerrell J, Feng H, Cadwallader KR, Schmidt SJ (2011a) Investigation of
thermal decomposition as the kinetic process that causes the loss of crystalline structure in
sucrose using a chemical analysis approach (Part II). J Agric Food Chem 59:702–712
Lee JW, Thomas LC, Schmidt SJ (2011b) Investigation of the heating rate dependency associated
with the loss of crystalline structure in sucrose, glucose and fructose using a thermal analysis
approach (Part I). J Agric Food Chem 59:684–701
Liu YT, Bhandari B, Zhou WB (2006) Glass transition and enthalpy relaxation of amorphous food
saccharides: a review. J Agric Food Chem 54(16):5701–5717
Liu Y, Bhandari B, Zhou W (2007) Study of glass transition and enthalpy relaxation of mixtures of
amorphous sucrose and amorphous tapioca starch syrup solid by differential scanning calo-
rimetry (DSC). J Food Eng 81:599–610
Mao C, Chamarthy SP, Byrn SR, Pinal R (2010) Theoretical and experimental considerations on
the enthalpic relaxation of organic glasses using differential scanning calorimetry. J Phys
Chem 114:268–279
McGonigle EA, Cowie JMG, Arrighi V, Pethrick RA (2005) Enthalpy relaxation and free volume
changes in aged styrene copolymers containing hydrogen bonding co-monomer. J Mater Sci
40:1869–1881
174 R.M. Syamaladevi et al.
Miller DP, Lechuga-Ballesteros D (2006) Rapid assessment of the structural relaxation behavior of
amorphous pharmaceutical solids: effect of residual water on molecular mobility. Pharm Res
23(10):2291–2305
Moynihan CT (1993) Correlation between the width of glass transition region and the temperature
dependence of the viscosity of high-Tg glasses. J Am Ceram Soc 76(5):1081–1087
Struik LCE (1978) Physical aging in amorphous polymers and other materials. Elsevier Scientific
Pub, Co., Amsterdam
Wang B, Pikal M (2010) The impact of thermal treatment on the stability of freeze-dried
amorphous pharmaceuticals: I. Dimer formation in sodium ethacrynate. J Pharm Sci 99
(2):663–682
Wungtanagorn R, Schmidt SJ (2001) Phenomenological study of enthalpy relaxation of amor-
phous glucose, fructose, and their mixture. Thermochim Acta 369:95–116
Effect of Dehydration Conditions on the Bulk
and Surface Properties of the Resulting
Dehydrated Products
Abbreviations
1 Introduction
2.1 Materials
X-ray powder diffraction (XRPD) patterns were carried out on a Bruker D5000
diffractometer (Bruker AXS Inc., Madison, WI) using copper radiation. The tube
voltage and amperage were set to 40 kV and 40 mA, respectively. The divergence
and scattering slits were set at 1 mm, and the receiving slit was set at 0.6 mm.
Diffracted radiation was detected by a Kevex PSI detector. A theta-two theta
continuous scan at 2.4 /min (1 s/0.04 step) from 3.0 to 40 2θ was used. An
alumina standard was analyzed to check the instrument alignment. Data were
collected and analyzed using Bruker axis software Version 7.0. Samples were
prepared by placing them in a quartz holder.
TA Q1000 DSC (TA instruments, New Castle, DE) was used for thermal studies up
to the melting point. Approximately 5 mg of sample was placed in the pan and
covered with a lid, and then a hole manually was punctured in the center. The
samples were analyzed from 0 to 50 C above the melting point of the sample at a
rate of 10 C/min. The experiments took place under nitrogen at a flow rate of
20 mL/min.
Samples were sprinkled onto sample holders covered with double-coated transpar-
ent adhesive tape. Excess was blown off and the samples were coated with Pt (60 s)
prior to imaging.
Samples were imaged with FEI Nova nanoSEM field emission SEM (FEI
Company, Hillsboro, OR) using Everhart-Thornley (ETD) or high-resolution
through-the-lens (TLD) detectors. Parameters were 5 kV accelerating voltage,
178 Y. Zhang and M.T. Carvajal
8000
7000
6000
5000
Lin (Cps)
4000
3000
2000
1000
0
4 10 20 30 40
2-Theta - Scale
Fig. 1 Anhydrate cytosine forms prepared under 60 C (black line) and 140 C (red line)
Effect of Dehydration Conditions on the Bulk and Surface Properties. . . 179
3900
3800
3700
3600
3500
3400
3300
3200
3100
3000
2900
2800
2700
2600
2500
2400
2300
Lin (Cps)
2200
2100
2000
1900
1800
1700
1600
1500
1400
1300
1200
1100
1000
900
800
700
600
500
400
300
200
100
0
4 10 20 30 40
2-Theta - Scale
Fig. 2 Anhydrate theophylline forms prepared under 60 C (black line) and 140 C (red line)
The morphology of dehydrated forms were also observed using SEM. Figure 4
shows the surface and cross section of the cytosine crystals of the resulting
dehydrated forms from the two temperatures. It can be seen that the surface of
crystals made under 140 C is slightly rougher than those made under 60 C. The
morphologies of the cross section from both are very similar.
Figure 5 shows the surface and cross section of the theophylline crystals for the
two dehydrated forms. The SEM pictures show that dehydration at a 140 C
produced smaller crystals along an axial of the crystal lattice from the original
hydrate crystals. This phenomenon could be explained that since the theophylline
hydrate is a tunnel hydrate, therefore, when water molecules leave the crystal, voids
are left in the tunnel. These voids could collapse by further heating; the original
large needle crystals then become smaller needle crystals along the tunnel. Also, it
can be observed that the surface of the dehydrated form of 140 C is rougher or
more corrugated than that at 60 C.
Figure 6 shows the surface and cross section of the thymine crystals in two
dehydrated forms. It is known that thymine shows a tunnel hydrate, and the
outcome of this compound is similar to that of theophylline. However, the tunnel
within the crystal is more noticeable than that in the theophylline crystal. The
crystals show obvious surface morphology changes: smooth, corrugated, and
lamella. Variability of the surfaces after processing will have an impact on the
functionality and performance of powders upon flow, compaction, and stability
(Galwey 2000).
An additional technique was used in an attempt to gain fundamental insight to
potentially further explain the resulting morphology and behavior of the two
dehydrated forms. A thermally stimulated polarized current (TSPC) experiment
was carried out for theophylline for the 60 and 140 C samples from 100 to 100 C
316 320.5
320
Melting Temperature, C
Melting Temperature, C
315.5
319.5
315 319
318.5
314.5 318
317.5
314
317
313.5 316.5
50 70 90 110 130 150 50 70 90 110 130 150
Dehydrated Temperature, C Dehydrated Temperature, C
Melting points of cytosine dehydrated forms prepared under 60°C and 140°C
271.1 273.5
271.05
Melting Temperature, C
Melting Temperature, C
273.4
271 273.3
270.95
273.2
270.9
273.1
270.85
270.8 273
270.75 272.9
270.7 272.8
270.65 272.7
50 70 90 110 130 150 50 70 90 110 130 150
Dehydrated Temperature, C Dehydrated Temperature, C
Melting points of theophylline thymine dehydrated forms prepared under 60°C and 140°C
320.3
Melting Temperature, C
Melting Temperature, C
319.4 320.25
319.2 320.2
319 320.15
318.8 320.1
320.05
318.6
320
318.4 319.95
318.2 319.9
318 319.85
50 70 90 110 130 150 50 70 90 110 130 150
Dehydrated Temperature, C Dehydrated Temperature, C
Melting points of thymine dehydrated forms prepared under 60°C and 140°C
Fig. 3 Melting points of thymine dehydrated forms prepared under 60 and 140 C
Fig. 4 SEM images of cytosine (Left: sample dehydrated at 60 C; Right: sample dehydrated at
140 C)
182 Y. Zhang and M.T. Carvajal
Fig. 5 SEM images of theophylline (Left: sample dehydrated at 60 C; Right: sample dehydrated
at 140 C)
(Fig. 7). Slight differences were observed in this case in the range below 100 C;
however, at higher temperature up to 200 C, there was a clear depolarization event
and differences between the two samples dehydrated at 140 and 60 C. This may
suggest that there is some mobility of molecules below 200 C.
Effect of Dehydration Conditions on the Bulk and Surface Properties. . . 183
Fig. 6 SEM images of thymine (Left: sample dehydrated at 60 C; Right: sample dehydrated at
140 C)
184 Y. Zhang and M.T. Carvajal
1.00E-12
0.00E+00
0 50 100 150 200 250
-1.00E-12
Current (Amps)
-2.00E-12
60C
140C
-3.00E-12
-4.00E-12
-5.00E-12
-6.00E-12
Temperature (C)
Fig. 7 TSPC profile for the 60 and 140 C theophylline samples (Yamauchi et al. 2011)
4 Conclusions
Acknowledgments We are grateful to the Dane O. Kildsig Center for Pharmaceutical Processing
Research (CPPR) for funding. Also, we thank Dr. Yinyan Zhao for conducting the TSPC analysis.
References
Abbreviations
C. Caballero-Cer
on
Centro de Biotecnologı́a FEMSA, Escuela de Biotecnologı́a y Alimentos,
Tecnol
ogico de Monterrey, Av. Eugenio Garza Sada 2501 Sur, Col. Tecnologico,
Monterrey, N.L. 64849, Mexico
J.A. Guerrero-Beltrán (*)
Departamento de Ingenierı́a Quı́mica, Alimentos y Ambiental, Universidad de las
Américas-Puebla, Sta. Catarina Mártir, Cholula, Puebla 72820, Mexico
e-mail: [email protected]
H. Mújica-Paz • J. Welti-Chanes (*)
Escuela de Ingenierı́a y Ciencias, Tecnologico de Monterrey, Av. Eugenio Garza Sada 2501
Sur, Col. Tecnologico, Monterrey, Nuevo Leon 64849, México
e-mail: [email protected]; [email protected]
J.A. Torres
Food Process Engineering Group, Department of Food Science and Technology,
Oregon State University, 100 Wiegand Hall, Corvallis, OR 97331, USA
1 Introduction
Fig. 1 Schematic
representation of moisture
sorption isotherms for food
Desorption
Adsorption
0 1
Water activity
Type I
Moisture content
Type IV
Type II
0 1 0 1
Water acvity Water acvity
et al. 2009). From a practical point of view, the monolayer value is a reference
parameter defining for dried products a moisture content value for maximal product
stability.
Many factors affect the sorption isotherm, but the most important ones are food
composition and temperature (Goula et al. 2008; Vega-Gálvez et al. 2009).
Food systems are complex matrices constituted by a continuous aqueous phase and
a small fraction of dispersed solids. Soluble compounds in the aqueous phase
consist of sugars, salts, and others, while the solid phase is constituted by insoluble
polymers such as starch, proteins, and insoluble fiber (Moreira et al. 2009). The
relationship of moisture and water activity depends on the food composition and
how water molecules interact with food compounds, i.e., how many polar sites are
available for the binding of water molecules (Iglesias and Chirife 1982; Kingsly and
Ileleji 2009; Moreira et al. 2009; Myhara et al. 1998). While water soluble mole-
cules such as sugars, salts, or soluble fibers bind water molecules contributing to
vapor pressure reduction, insoluble solids with no binding sites do not, thus higher
values of aw are reached at the same equilibrium moisture content (Goula
et al. 2008; Moreira et al. 2009).
Fig. 3 Effect of
temperature on moisture 20°C
30°C
sorption isotherms
Water Activity
Experimental methods are divided into those with the air relative humidity and/or
temperature continuously changing, and those known as dynamic methods, in
contrast with static methods where air properties are kept constant during aw
determination.
In static methods, samples in hermetic and closed vessels are kept in contact with
air at constant relative humidity (RH) and temperature until equilibrium between
the sample and air is reached. These methods can be classified as hygrometric,
manometric, and gravimetric (Iglesias and Chirife 1982). In hygrometric methods,
the surrounding food atmosphere vapor pressure is measured using dew point and
electrical sensors (Spiess and Wolf 1987). In manometric methods, air vapor
pressure in equilibrium with the food sample is measured (Goula et al. 2008; Spiess
and Wolf 1987). In gravimetric methods, equilibrium is evaluated by following
weight changes (Iglesias and Chirife 1982) in samples (2 g approx.) with known
moisture content in contact with air at constant RH and temperature. Weight
loss (desorption) or gain (adsorption) is evaluated until equilibrium is reached,
i.e., when the difference between two consecutive measurements is
approximately 0.0010 g. The measuring system consists of a series of hermetic
containers (glass jars or desiccators), a glass or plastic base (usually a petri dish),
and weighing bottles where samples are placed (Iglesias and Chirife 1982; Bell and
Labuza 2000; Spiess and Wolf 1987). Hermetic flasks (or desiccators) are placed in
a constant room temperature or in an incubator. Saturated salt solutions, sulfuric
acid, or glycerol solutions are used to control the RH. Saturated salt solutions are
Moisture Sorption Isotherms of Foods: Experimental Methodology. . . 193
preferred, as they are thermally stable and show small RH variation with temper-
ature changes, as well as being stable to changes in sample moisture content
(Greenspan 1977; Rockland 1960; Spiess and Wolf 1987). Table 1 shows the
most common salts employed in the 0.11–0.976 aw interval (Lopez-Malo
et al. 1994).
Static methods are simple to operate and allow multiple samples to be analyzed
simultaneously, but determination times are long, samples must be removed from
the system for weighing which may modify their equilibrium moisture, requiring
large sample quantities for a complete isotherm, and determination points are
discrete (Yu et al. 2008a, b; Lewicki 2000). Long determination times are required
to ensure that thermodynamic equilibrium between samples and air is reached
(Spackman and Schmidt 2010). However, these long times may result in mold
growth in samples with RH greater than 75 %, causing the loss of experimental data
(Bell and Labuza 2000; Yu et al. 2008a, b; Baucour and Daudin 2000). Automatic
and dynamic methods have been developed to avoid these experimental limitations.
Samples are placed on a microbalance and an air stream with known RH and
constant temperature passes continuously over them. Constant exposure to air
flow and the small sample weight reduce the determination time from days to
hours, making it possible to measure points at high aw levels without the risk of
microbial growth or physicochemical changes in the sample (Rahman and
194 C. Caballero-Cer
on et al.
Water activity and sample weight changes due to continuous flow of a saturated
vapor stream (adsorption) or dried air (desorption) over the sample are measured
directly (Schmidt and Lee 2012; Carter and Fontana 2008). Air flow stops after aw
changes of 0.015 units approximately, and weight and water activity are measured
for each point. Thermodynamic equilibrium may not be reached during measure-
ment, and thus results may differ from those obtained by other methods, particularly
for samples with a slow rate of water diffusion (Schmidt and Lee 2012; Carter and
Fontana 2008). This method has been used to evaluate time-depending phase
changes (Yao et al. 2011; Yuan et al. 2011). The sample chamber in the isotherms
generator consists of a precision balance, a standard mirror dew point sensor, inflow
ports for dry air and water saturated air, and an infrared temperature sensor. Dry air
is obtained by air circulation through a packed desiccant column, while the water
Moisture Sorption Isotherms of Foods: Experimental Methodology. . . 195
80
70
Moisture content (%, db)
60
50
40
30
20
10
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Water Activity
Fig. 4 Sorption isotherm at 32 C of apple obtained by DDI method (open square) and gravi-
metric static method ( filled diamond) (unpublished data)
saturated air is obtained by air circulation through a water reservoir (Schmidt and
Lee 2012; Carter and Fontana 2008). The DDI method is designed to obtain a
complete sorption isotherm (adsorption and desorption) with one sample in a short
time period of approximately 24 h, obtaining a high resolution isotherm (>50
points for isotherm). As more data become available, isotherms can be used to
obtain additional information on the adsorption of water molecules, as well as phase
changes due to moisture content changes. Apple moisture isotherms at 32 C
obtained by conventional static and DDI methods show the same behavior, but
the resolution is higher for the latter (Fig. 4). Studies on phase changes such as
crystallization, glass transition, and deliquescence points have been reported (Car-
ter and Fontana 2008; Schmidt and Lee 2012; Yao et al. 2011; Yuan et al. 2011;
Spackman and Schmidt 2010; Carter and Schmidt 2012). Yuan et al. (Yuan
et al. 2011) evaluated the polydextrose glass transition, obtaining Tg values similar
to those determined by differential scanning calorimetry (DSC). Yao et al. (2011)
measured the deliquescence point for crystalline sucrose at 15, 25, and 35 C.
Although Tg values obtained by the DDI method are restricted to the 15–40 C
equipment operation limits, these are the most common temperature storage con-
ditions; thus; results will be useful when setting food storage and shipping condi-
tions. Schmidt and Lee (2012) observed that sorption isotherms for corn starch,
isolated soy protein, microcrystalline cellulose, and crystalline sucrose obtained by
DDI and static gravimetric methods were similar. However, those for lactose and
corn flakes differed, which was assumed to be associated with phase transitions or
slow water diffusion rate, respectively (Carter and Fontana 2008). Diffusion rate
can be increased by decreasing the particle size or reducing air flow, yielding DDI
results in agreement with those obtained by other methods (Schmidt and Lee 2012).
196 C. Caballero-Cer
on et al.
The BET equation, a semiempirical 2-parameter model, describes the aw < 0.5
region (Al-Muhtaseb et al. 2002; Bhandari and Adhikari 2008) as follows:
Xm Caw
X¼ ð1Þ
ð1 aw Þ½1 þ ðC 1Þaw
where C is an adsorbent constant. This model assumes that the sorption heat in the
monolayer is constant and equal to the water vaporization heat plus a constant related
to the interaction site, the sorption heat for water molecules above the monolayer value
is equal to that for pure water, water sorption occurs only on specific food structure
sites, and the sorption surface is homogeneous. The monolayer moisture content
defines the moisture content for the maximum stability of dehydrated products.
H1 Hm
C ¼ C0 exp ð3Þ
RT
H1 Hq
K ¼ K 0 exp ð4Þ
RT
where H1, Hm, and Hq are, respectively, the condensation heat of pure water
(J/mol), total sorption heat of the monolayer (J/mol), and total sorption heat of
subsequent water layers (J/mol), T is the absolute temperature (K) and R is the ideal
gas constant (8.314 J/mol/K) (Talla et al. 2005). Xm is considered temperature
independent, but in some cases it has been shown to decrease when temperature
increases (Goula et al. 2008; Quirijns et al. 2005). This monolayer moisture content
decrease may be due to a reduction of the active binding water sites as a result of
physical or chemical changes induced by temperature. The temperature function for
Xm can be expressed as:
ΔHX
Xm ¼ Xm, 0 exp ð5Þ
RT
where Xm,0 and ΔHX (J/mol) are constants. The GAB model assumes that water
adsorption is a multilayer phenomenon without lateral interactions of water mole-
cules. The monolayer covers the complete sorbent surface with water molecules
bound very tightly in the monolayer, whereas the subsequent 2–9 multilayers have
less interaction with the sorbent (Quirijns et al. 2005). Timmermann et al. (2001)
observed that the GAB parameters are more representative than the corresponding
BET parameters for food products, whereas Singh and Singh (1996) showed that it
can describe different food isotherm types. The temperature, moisture content, and
aw intervals tested in this latter work were 4–140 C, 2–77 %, and 0.059–0.99,
respectively.
This model overcomes BET limitations (Rahman 1995) by assuming that the
potential energy of water molecules varies with the inverse of their distance from
the food surface and can describe a wide variety of food isotherms in the 0.1–0.8 aw
interval and is expressed as follows:
1=C
A
X¼ ð6Þ
RTln aw
198 C. Caballero-Cer
on et al.
where A and C are constants, and T is the absolute temperature (K). As the term RT
does not eliminate the temperature dependence of constants A or B, this expression
is simplified as follows (Iglesias and Chirife 1976; Basu et al. 2006):
1=B
A0
X¼ ð7Þ
ln aw
1=C
expðA þ BT Þ
X¼ ð8Þ
ln aw
1 aw ¼ expðkTXn Þ ð9Þ
where k and n are constants and T is the temperature ( C). The following expression
is a modification of the original Henderson equation considering k as temperature
function:
4.5 K€
uhn Equation (K€
uhn 1964)
This theoretical model, based on a capillary condensation concept and applied most
frequently to vegetable and animal products (Akanbi et al. 2006; Escobedo-
Avellaneda et al. 2011a, b), is expressed as follows:
Moisture Sorption Isotherms of Foods: Experimental Methodology. . . 199
k
X¼ þ B0 ð11Þ
ln aw
where A and B are constants. Applications of this equation include high starch
content products and also different vegetable products (Al-Muhtaseb et al. 2002).
Assuming that food moisture can be described as pure and hydration water, Lewicki
(2000) proposed the following 2-parameter model related to Raoult’s law:
b1
1
X¼A 1 ð13Þ
aw
where A and b are constants. This equation can yield better results than the GAB
model, especially when aw is close to 1.0. Although this equation is similar to the
Oswin model (Eq. 12), some differences between these two equations can be noted.
In contrast with the 2-parameter Lewicki expression (Eq. 13), the Oswin equation is
an empirical model, and thus its constants have no physical meaning. In addition,
the point aw ¼ 1 is excluded (Lewicki 2000; McMinn et al. 2004). Lewicki (1998)
proposed the following 3-parameter model based on a 2-function expression to
describe processes occurring in parallel, the first one predominating at high aw, with
the second being more important at low aw:
F F
X¼ þ ð14Þ
ð 1 aw Þ G 1 þ aw H
where F, G, and H are constants. This equation was tested using 31 sorption
isotherms reported for various food products yielding better results than those
obtained for the GAB, but not as good as those for the Peleg model (Eq. 16).
200 C. Caballero-Cer
on et al.
McMinn et al. (2004) showed that while the 2- and 3-parameter Lewicki equations
well describe experimental data for high starch products, the 3-parameter equation
gives better results for food products with Type II isotherms within the 30–60 C
temperature range, and the 2-parameter equation was superior to the GAB equation.
However, sorption isotherms of potato, high amylose, and high amylopectin starch
powders cannot be described using the Lewicki equations.
The Smith model, an empirical equation describing the high aw portion of sorption
isotherms for products containing high molecular weight solutes (Rahman 1995),
can be represented as follows:
X ¼ A þ Blogð1 aw Þ ð15Þ
where A and B are constants, A is the moisture fraction absorbed on the first
fraction, and B is the water in the multilayer fraction. This model, limited to the
0.5–0.95 aw interval (Al-Muhtaseb et al. 2004; Rahman 1995), assumes that there
are two fractions of water molecules adsorbed on the food surface. The first fraction
with a condensation heat higher than the one for pure water follows the Langmuir
model, while the second fraction consisting of a multilayer of condensed water
molecules formed only after the first is absorbed, preventing any evaporation from
the first fraction. The Smith model describes only the behavior of the second
fraction (Al-Muhtaseb et al. 2002).
where K1, K2, n1, and n2 are constants and n1 < 1 and n2 > 1. When aw ! 1,
X ! K1 + K1; however, the adsorbed moisture content values do not reach these
values at aw values close to 1.0. Therefore, this equation is used for aw < 0.95
(Lewicki 1998).
Moisture Sorption Isotherms of Foods: Experimental Methodology. . . 201
X
n 2
Xi, calc Xi, exp
i¼1
R2 ¼ 1 X
n ð17Þ
2
Xi, exp Xi, exp
i¼1
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
X n 2
RMSE ¼ Xi, calc Xi, exp ð18Þ
i¼1
n
1X Xi, calc Xi, exp
MRD ¼ ð19Þ
n i¼1 Xi, exp
100 X
n
aw, calc aw, exp
ASE ¼ abs ð20Þ
n i¼1 1 aw, exp
where aw,calc. and aw,exp, are the aw values estimated with the model and experi-
mental value, respectively, while Xi,exp, Xi,calc, Xi are the experimental, calculated,
and average moisture content values, respectively, and n is the number of experi-
mental data points. R2 quantifies the portion of the total variation explained by
regression, so when R2 is closer to 1, the model better fits the experimental data.
RMSE and MRD values are measurements of deviation between experimental and
predicted data, and low values of RMSE and MRD are expected for equations best
describing the experimental data. When MRD < 10 %, the equation can be used to
describe experimental data (Ruiz-Lopez and Herman-Lara 2009; Basu et al. 2006).
Plotting the residuals (X Xcalc) against the independent variable is also a measure
of error distribution, i.e., when the model suitably describes the experimental data,
the residuals should be randomly independent errors with a zero mean, constant
variance and arranged in a normal distribution. If residual plots have a clear pattern,
the model should be rejected (Basu et al. 2006).
202 C. Caballero-Cer
on et al.
then compared to those obtained with the GAB model. The reported values of MRD
above 15 % for the Halsey, Henderson, and Oswin equations show that these
equations are not suitable for high sugar content products such as fruits.
The mathematical description of sorption isotherms for vegetable products,
particularly spray dried tomato pulp (Table 3), shows that in the 20–60 C temper-
ature range the GAB model best describes the experimental data (MRE < 4 %)
followed by the Peleg equation (MRE < 11 %). The Halsey, Henderson, and Oswin
equations were not successful in showing MRE values reaching 86 % for the Halsey
equation. In contrast, for green pepper, red pepper, and samples with high starch
content such as potato, the Halsey and Oswin equations showed better results with
MRE values in the 3.7–10.0 % range, while for mushrooms and carrots, the
Henderson model was the most suitable with MRE of 0.42 % and 4.7 %, respec-
tively. Table 4 summarizes equations describing the moisture isotherm for other
204 C. Caballero-Cer
on et al.
Table 4 Mathematical equation used to describe diverse food products moisture sorption
isotherm
Equation Samples
GAB Cocoa beans (Menkov 2000; Sá and Sereno 1994), millet flour (Jamali et al. 2006),
Basundi ready to use (Schmidt and Lee 2012), Chenopodium ambrosioides
(pazote) (Johnson and Brennan 2000), cowpea and its protein insolated (Ayranci
and Duman 2005), bitter orange leaves (Moraga et al. 2004), curd powder
(Varghese et al. 2008), rice crackers (Singh and Singh 1996), maltodextrin (Lim
et al. 1995), safflower petals and tarragon (Kaymak-Ertekin and Gedik 2003), fish
flour (Lemus-Mondaca et al. 2009), cornmeal (Gabas et al. 1999), freeze-dried
milk products (Katekawa and Silva 2007), starch gels (McMinn et al. 2004), potato
starch (Al-Muhtaseb et al. 2002), bean flour (Mohamed et al. 2005), amaranth
grains (Peleg 1993), macaroni (Arslan and To grul 2005), rice (Tonon et al. 2009),
lentil seeds (Menkov et al. 2004), corn starch (Pott et al. 2005)
BET Cocoa beans (Menkov 2000; Sá and Sereno 1994), millet flour (Jamali et al. 2006),
Basundi ready to use (Schmidt and Lee 2012), cowpea and its protein insolated
(Ayranci and Duman 2005), walnut flour (Menkov et al. 2005), curd powder
(Varghese et al. 2008), safflower petals and tarragon (Kaymak-Ertekin and Gedik
2003), fish flour (Lemus-Mondaca et al. 2009), cornmeal (Gabas et al. 1999),
freeze-dried milk products (Katekawa and Silva 2007), bean flour (Mohamed
et al. 2005), macaroni (Arslan and Togrul 2005), rice (Tonon et al. 2009), lentil
seeds (Menkov et al. 2004), corn starch (Pott et al. 2005)
Halsey Cocoa beans (Sá and Sereno 1994), cowpea and its protein insolated (Ayranci and
Duman 2005), curd powder (Varghese et al. 2008), safflower petals and tarragon
(Kaymak-Ertekin and Gedik 2003), freeze-dried milk products (Jamali et al. 2006),
starch gels (McMinn et al. 2004), potato starch (Al-Muhtaseb et al. 2002), corn
starch (Pott et al. 2005)
Henderson Curd powder (Varghese et al. 2008), cocoa beans (Sá and Sereno 1994), cowpea
and its protein insolated (Ayranci and Duman 2005), freeze-dried milk products
(Katekawa and Silva 2007), starch gels (McMinn et al. 2004), macaroni (Arslan
and To grul 2005), rice (Tonon et al. 2009), potato starch (Al-Muhtaseb et al. 2002),
corn starch (Pott et al. 2005)
Künh Cocoa beans (Sá and Sereno 1994), cowpea and its protein insolated (Ayranci and
Duman 2005), freeze-dried milk products (Katekawa and Silva 2007)
Lewicki Corn starch (Pott et al. 2005)
Oswin Cowpea and its protein insolated (Ayranci and Duman 2005), curd powder
(Varghese et al. 2008), safflower petals and tarragon (Kaymak-Ertekin and Gedik
2003), freeze-dried milk products (Katekawa and Silva 2007), macaroni (Arslan
and To grul 2005), rice (Tonon et al. 2009)
Peleg Starch gels (McMinn et al. 2004), macaroni (Arslan and To grul 2005), rice (Tonon
et al. 2009), potato starch (Al-Muhtaseb et al. 2002)
Smith Cocoa beans (Sá and Sereno 1994), cowpea and its protein insolated (Ayranci and
Duman 2005), curd powder (Varghese et al. 2008), starch gels (McMinn
et al. 2004), potato starch (Al-Muhtaseb et al. 2002), macaroni (Arslan and To grul
2005), rice (Tonon et al. 2009), corn starch (Pott et al. 2005)
food products. For amaranth grains and bean flour, the Henderson, Halsey, and
Oswin models had MRD < 10 %, confirming that these equations can be used to
describe high starch products. In conclusion, the models best describing moisture
isotherms vary with the product type and composition.
Moisture Sorption Isotherms of Foods: Experimental Methodology. . . 205
Q 1
ln aw ¼ ð21Þ
R T
ΔH ΔS
ln aw ¼ ð22Þ
RT R
A linear relationship between the enthalpy and entropy of the moisture sorption
forces has been shown for several products where the slope represents the isokinetic
temperature (TB) and the intercept is the Gibb’s free energy value (ΔGB) (Aguerre
et al. 1986; Beristain et al. 1996; Ferro-Fontan et al. 1982). This relationship
(Eq. 23) is known as the enthalpy-entropy compensation or isokinetic theory.
This relationship determines if sorption phenomena are controlled by enthalpic or
entropic processes. TB, defined as the temperature at which all reactions in the
series proceed at the same rate, can be compared to the harmonic temperature
(Thm, Eq. 24) to confirm the existence of enthalpy-entropy compensation.
When TB > Thm, the sorption process is enthalpy controlled, while Thm > TB indi-
cates that the process is entropy controlled (Aguerre et al. 1986; Gabas et al. 2000).
ΔH ¼ T B ΔS þ ΔGB ð23Þ
206 C. Caballero-Cer
on et al.
Fig. 5 Schematic
representation of isosteric
λvap
0 1.0
Moisture content (g/g, db)
n
T hm ¼ X
n ð24Þ
1= TÞ
i¼1
Sorption isotherms are used to predict the stability of foods sensitive to moisture
changes during storage (Bell and Labuza 2000), including microbial activity pre-
diction, evaluation of chemical changes due to moisture gain or loss, enzymatic
Moisture Sorption Isotherms of Foods: Experimental Methodology. . . 207
reactions, lipid oxidation, and nonenzymatic browning (Iglesias and Chirife 1982).
Physical changes may take place during the adsorption process, including sticki-
ness, caking, crystallization, and other alterations caused by aw changes. Knowl-
edge of moisture content or aw value leading to undesirable changes in product
quality, defined as the critical moisture sorption point (MSP), could improve food
transport and storage to prevent undesirable product changes.
Product quality losses have also been related to the glass transition point of
dehydrated products. An amorphous state is reached in foods with high sugar
content when they have been dehydrated; however, when phase changes to a
rubbery state, quality loss can occur within days and even hours. The viscosity of
amorphous material is around 1012 Pa s, and the mobility of water molecules will be
low enough to prevent microbial and enzymatic degradation reactions (Xie
et al. 2010; Rahman et al. 2003; Liu et al. 2010). This phase change from a glassy
state to a rubbery state is known as glass transition, which is a temperature, moisture
content, and time dependent process.
Phase transitions can be accelerated by increasing temperature or adding a
plasticizer. In high sugar products, water can serve as the plasticizer; thus, the
increase in moisture content may decrease the glass transition temperature to the
storage temperature (Yuan et al. 2011). The moisture content at which a glass
transition takes place at a given storage temperature is a critical moisture content
value. Estimations of the critical moisture content and aw at the glass transition
point are presented in Fig. 6 for structural or texture changes in the final product due
to the gain of water (Tonon et al. 2009).
Moisture content (g/g, db)
Storage
temperature
Temperature (K)
Xc
aw,c
0 1
WaterActivity
Fig. 6 Graphical representation of glass transition (dashed line) and moisture sorption isotherms
(solid line) used to predict critical moisture content by dried products in order to prevent changes
on structural or texture properties
208 C. Caballero-Cer
on et al.
Moisture isotherms can be used to select the most appropriate packaging material to
achieve a desired shelf life (Welti-Chanes et al. 2008; Escobedo-Avellaneda
et al. 2011a). Since the adsorption and desorption processes are not the same due
to the hysteresis phenomenon, it is important to use adsorption isotherms when
analyzing moisture gain and desorption isotherms. An adequate predictive model is
crucial to accurately predict the shelf life of products, such as the one proposed by
Labuza and Altunakar (2007):
X e Xi k A p0v
ln ¼ θ ð25Þ
Xe X c x ds m
where X is moisture content (g H2O/g dried solids) with the subindices i and c
corresponding to the initial and critical values, respectively, while e describes the
equilibrium with the storage RH; k/x (g/m2 dia mmHg) is the permeance of the
packaging film, A (m2) is the packaging area, ds (g solids) are the dried solids of the
food, pv0 (mm Hg) is the water vapor pressure at the sorption isotherm temperature,
m is the linearized portion slope of the sorption isotherm in the range of interest
(Fig. 7), and θ (days) is the estimated shelf life (Bourlieu et al. 2009). Equation (24)
allows estimations of shelf life time by considering initial moisture content, mois-
ture content at which the product becomes unacceptable to the consumer, temper-
ature and RH of storage, and packaging material. This equation can be used in a
deterministic mode, i.e., when parameter variability is not considered in the calcu-
lation and therefore the predicted value does not include a shelf life confidence
Moisture content (g/g, db)
X,e
X,c
X,i
Fig. 7 Sorption isotherm (solid line) used to calculate parameters in shelf life equation for dried
foods, where Xc, Xi, and Xe are the critical, initial, and equilibrium moisture content, respectively
Moisture Sorption Isotherms of Foods: Experimental Methodology. . . 209
assessment. On the other hand, a Monte Carlo procedure was recently used to
consider the variability of the moisture isotherm to estimate frequency distributions
for the shelf life of dehydrated tomato slices, chopped onions, and green beans
slices (Escobedo-Avellaneda et al. 2011b). A Monte Carlo procedure is a statistical
method that allows the generation of frequency distributions describing all possible
values for a given outcome (Chotyakul et al. 2011, 2012). This procedure takes into
account the statistical variability of all parameters used. Escobedo-Avellaneda
et al. (2011b) showed a shelf life of 175.5, 62, and 57 days for tomato, onion, and
green beans of intermediate and low moisture content when the variability of GAB
parameters and aw values was not taken into account. On the other hand, when the
standard deviation of GAB parameters and aw values was considered, shelf life
frequency distributions of about 130–270 days, 55–69 days, and 51–65 days for
tomato, onion, and green beans were obtained, respectively. If the accepted prob-
ability that the shelf life would fail before the time specified were 5 %, the
recommended shelf life for tomato, onion, and green beans would be 150 days,
58 days, and 53 days, respectively. On the other hand, if the average values for the
prediction model parameters were used to estimate shelf life, the probability of
product failure before this value would be equal to or higher than 50 %. Therefore,
the Monte Carlo method proposed by these authors is an effective tool to determine
the shelf life of food products with an acceptable risk of product failure before the
shelf life declared by the producer on its label.
10 Final Remarks
calculation procedures such as the Monte Carlo method represent a new opportu-
nity to improve shelf life estimations.
Acknowledgments The authors acknowledge the financial support from Tecnol ogico de Mon-
terrey (Research Chair Funds CAT-00200 and Nutrigenomica), and author Claudia Caballero-
Ceron thanks for the financial support for her graduate studies to CONACYT and Tecnol
ogico de
Monterrey.
References
Carter BP, Schmidt SJ (2012) Developments in glass transition determination in foods using
moisture sorption isotherms. Food Chem 132:1693. doi:10.1016/j.foodchem.2011.06.022
Chotyakul N, Velazquez G, Torres JA (2011) Assessment of the uncertainty in thermal food
processing decisions based on microbial safety objectives. J Food Eng 102:247–256
Chotyakul N, Pérez Lamela C, Torres JA (2012) Effect of model parameter variability on the
uncertainty of refrigerated microbial shelf-life estimates. J Food Process Eng 35:829. doi:10.
1111/j.1745-4530.2010.00631.x
Corey ME, Kerr WL, Mulligan JH, Lavelli V (2011) Phytochemical stability in dried apple and
green tea functional products as related to moisture properties. LWT Food Sci Technol
44:67–74
Duarte Goneli AL, Corrêa PC, Horta De Oliveira GH, Ferreira Gomes C, Mendes Botelho F (2010)
Water sorption isotherms and thermodynamic properties of pearl millet grain. Int J Food Sci
Technol 45:828–838
Eim VS, Rossell o C, Femenia A, Simal S (2011) Moisture sorption isotherms and thermodynamic
properties of carrot. Int J Food Eng 7(3):1. doi:10.2202/1556-3758.1804
Escobedo-Avellaneda Z, Pérez-Pérez MC, Bárcenas-Pozos ME, Welti-Chanes J (2011a) Moisture
adsorption isotherms of freeze-dried and air-dried Mexican red sauce. J Food Process Eng
34:1931–1945
Escobedo-Avellaneda Z, Velazquez G, Torres JA, Welti-Chanes J (2011b) Inclusion of the
variability of model parameters on shelf-life estimations for low and intermediate moisture
vegetables. LWT Food Sci Technol 47:364–370
Ferro-Fontan C, Chirife J, Sancho E, Iglesias HA (1982) Analysis of a model for water sorption
phenomena in foods. J Food Sci 47:1590–1594
Furmaniak S, Terzyk AP, Gołembiewski R, Gauden PA, Czepirski L (2009) Searching the most
optimal model of water sorption on foodstuffs in the whole range of relative humidity. Food
Res Int 42:1203–1214
Gabas AL, Telis-Romero J, Menegalli FC (1999) Thermodynamic models for water sorption by
grape skin and pulp. Dry Tech 17:962–974
Gabas AL, Menegali FC, Telis-Romero J (2000) Water sorption enthalpy-entropy compensation
based on isotherms of plum skin and pulp. J Food Sci 65:680–684
Goula AM, Karapantsios TD, Achilias DS, Adamopoulos KG (2008) Water sorption isotherms
and glass transition temperature of spray dried tomato pulp. J Food Eng 85:73–83
Greenspan L (1977) Humidity fixed points of binary saturated aqueous solutions. J Res Natl Bur
Stand 81a:89–112
Halsey G (1948) Physical adsorption in non-uniform surfaces. J Chem Phys 16:931
Henderson SM (1952) A basic concept of equilibrium moisture. Agric Eng 33:29–35
Hossain MD, Bala BK, Hossain MA, Mondol MRA (2001) Sorption isotherms and heat of sorption
of pineapple. J Food Eng 48:103–107
Iglesias HA, Chirife J (1976) Equilibrium moisture contents of air dried beef. Dependence on
drying temperature. Int J Food Sci Technol 11:565–573
Iglesias HA, Chirife J (1982) Handbook of food isotherms. Academic, New York, NY
Jamali A, Kouhila M, Mohamed LA, Jaouhari JT, Idimam AI, Abdenouri N (2006) Sorption
isotherms of Chenopodium ambrosioides leaves at three temperatures. J Food Eng 72:77–84
Johnson P-NT, Brennan JG (2000) Moisture sorption isotherm characteristics of plantain (Musa,
AAB). J Food Eng 44:79–84
Jouppila K, Roos YH (1997) Water sorption isotherms of freeze-dried milk products: applicability
of linear and non-linear regression analysis in modeling. Int J Food Sci Technol 32:459–471
Katekawa ME, Silva MA (2007) On the influence of glass transition on shrinkage in convective
drying of fruits: a case study of banana drying. Dry Tech 25:1659–1666
Kaya S, Kahyaoglu T (2007) Moisture sorption and thermodynamic properties of safflower petals
and tarragon. J Food Eng 78:413–421
Kaymak-Ertekin F, Gedik A (2003) Sorption isotherms and isosteric heat of sorption for grapes,
apricots, apples and potatoes. LWT Food Sci Technol 37:429–438
212 C. Caballero-Cer
on et al.
Moreira R, Chenlo F, Vázquez MJ, Cameán P (2005) Sorption isotherms of turnip top leaves and
stems in the temperature range from 298 to 328 K. J Food Eng 71:193–199
Moreira R, Chenlo F, Torres MD, Vallejo N (2008) Thermodynamic analysis of experimental
sorption isotherms of loquat and quince fruits. J Food Eng 88:514–521
Moreira R, Chenlo F, Torres MD (2009) Simplified algorithm for the prediction of water sorption
isotherms of fruits, vegetables and legumes based upon chemical composition. J Food Eng 94
(3–4):334–343
Myhara RM, Taylor MS, Slominski BA, Al-Bulushi I (1998) Moisture sorption isotherms and
chemical composition of Omani dates. J Food Eng 37:471–479
Oswin CR (1946) The kinetics of package life. III. The isotherm. J Soc Chem Ind 65(12):419–421
Peleg M (1993) Assessment of a semi-empirical four parameter general model for sigmoid
moisture sorption isotherms. J Food Process Eng 16:21–37
Pott I, Neidhart S, Mühlbauer W, Carle R (2005) Quality improvement of non-sulphited mango
slices by drying at high temperatures. Innov Food Sci Emerg Technol 6:412–419
Quirijns EJ, van Boxtel AJB, van Loon WKP, van Straten G (2005) Sorption isotherms, GAB
parameters and isosteric heat of sorption. J Sci Food Agric 85:1805–1814
Rahman MS (1995) Food properties handbook. CRC, Boca Raton, FL
Rahman MS (2009) Food stability beyond water activity and glass transition: macro-micro region
concept in the state diagram. Int J Food Prop 12:726–740
Rahman MS, Al-Belushi RH (2006) Dynamic isopiestic method (DIM): measuring moisture
sorption isotherm of freeze-dried garlic powder and other potential uses of DIM. Int J Food
Prop 9:421–437
Rahman MS, Kasapis S, Guizani N, Al-Amri OS (2003) State diagram of tuna meat: freezing curve
and glass transition. J Food Eng 57:321–326
Rockland LR (1960) Saturated salt solutions for static control of relative humidity between 5 and
40 C. Anal Chem 32:1375–1376
Roman-Gutierrez AD, Guilbert S, Cuq B (2002) Distribution of water between wheat flour
components: a dynamic water vapor adsorption study. J Cereal Sci 36:347–355
Ruiz-L opez II, Herman-Lara E (2009) Statistical indices for the selection of food sorption
isotherm models statistical indices for the selection of food sorption isotherm models. Dry
Tech 27:726–738
Sá MM, Sereno AM (1994) Glass transition and state diagrams for typical natural fruits and
vegetables. Thermochim Acta 246:285–297
Schmidt SJ, Lee JW (2012) Comparison between water vapor sorption isotherms obtained using
the new dynamic dewpoint isotherm method and those obtained using the standard saturated
salt slurry method. Int J Food Prop 15:236–248
Sharma P, Singh RRB, Singh AK, Patel AA, Patil GR (2009) Sorption isotherms and thermody-
namics of water sorption of ready-to-use Basundi mix. LWT Food Sci Technol 4:441–445
Singh PC, Singh RK (1996) Application of GAB model for water sorption isotherms of food
products. J Food Process Preserv 20:203–220
Smith SE (1947) The sorption of water vapor by high polymers. J Am Chem Soc 69:646–651
Spackman CCW, Schmidt SJ (2010) Characterizing the physical state and textural stability of
sugar gum pastes. Food Chem 119:490–499
Spiess WEL, Wolf W (1987) Critical evaluation of methods to determine moisture sorption
isotherms. In: Rockland LR, Beuchat LR (eds) Water activity: theory and applications to
food. Institute of Food Technologists, Chicago, IL, pp 215–233
Talla A, Jannot Y, Elambo Nkeg G, Puiggali J-R (2005) Experimental determination and modeling
of sorption isotherms of tropical fruits: banana, mango, and pineapple experimental determi-
nation and modeling of sorption isotherms of tropical. Dry Tech 23:1477–1498
Timmermann EO, Chirife J, Iglesias HA (2001) Water sorption isotherms of foods and foodstuffs:
BET or GAB parameters? J Food Eng 48:19–31
214 C. Caballero-Cer
on et al.
Tonon RV, Baroni AF, Brabet C, Gibert O, Pallet D, Hubinger MD (2009) Water sorption and
glass transition temperature of spray dried açai (Euterpe oleracea Mart.) juice. J Food Eng
94:215–221
van den Berg C, Bruins S (1981) Water activity and its estimation in food systems: theoretical
aspects. In: Rockland LR, Stewart GF (eds) Water activity: influence on food quality. Aca-
demic Press, Inc., New York, NY, pp 1–61
Varghese KS, Ramachandrannair SV, Mishra HN (2008) Moisture sorption characteristics of curd
(Indian yogurt) powder. Int J Dairy Technol 62:85–92
Vega-Gálvez A, L opez J, Miranda M, DiScala K, Yagnam F, Uribe E (2009) Mathematical
modeling of moisture sorption isotherms and determination of isosteric heat of blueberry
variety O’Neil. Int J Food Sci Technol 44:2033–2041
Viollaz PE, Rovedo CO (1999) Equilibrium sorption isotherms and thermodynamic properties of
starch and gluten. J Food Eng 40:287–292
Viswanathan R, Jayas DS, Hulasare RB (2003) Sorption isotherms of tomato slices and onion
shreds. Biosyst Eng 86:465–472
Welti-Chanes J, Pérez E, Guerrero-Beltrán JA, Alzamora SM, Vergara-Balderas F (2008) Appli-
cations of water activity management in the food industry. In: Barbosa-Canovas GV, Fontana
AJ, Schmidt AR, Labuza TP (eds) Water: activity in foods. Blackwell Publishing Ltd., Oxford,
UK, pp 341–357
Xie F, Liu W-C, Liu P, Wang J, Halley PJ, Yu L (2010) Starch thermal transitions comparatively
studied by DSC and MTDSC. Starch 62:350–357
Yan Z, Sousa-Gallagher J, Oliveira FAR (2008a) Sorption isotherms and moisture sorption
hysteresis of intermediate moisture content banana. J Food Eng 86:342–348
Yan Z, Sousa-Gallagher MJ, Oliveira FAR (2008b) Effect of temperature and initial moisture
content on sorption isotherms of banana dried by tunnel drier. Int J Food Sci Technol
43:1430–1436
Yao W, Yu X, Lee JW, Yuan X, Schmidt SJ (2011) Measuring the deliquescence point of
crystalline sucrose as a function of temperature using a new automatic isotherm generator.
Int J Food Prop 14:882–893
Yu X, Kappes SM, Bello-Pérez LA, Schmidt SJ (2008a) Investigating the moisture sorption
behavior of amorphous sucrose using a dynamic humidity generating instrument. J Food Sci
73(1):25–35
Yu X, Martin SE, Schmidt SJ (2008b) Exploring the problem of mold growth and the efficacy of
various mold inhibitor methods during moisture sorption isotherm measurements. J Food Sci
73:69–81
Yu X, Schmidt AR, Bello-Pérez LA, Schmidt SJ (2008c) Determination of the bulk moisture
diffusion coefficient for corn starch using an automated water sorption instrument. J Agric
Food Chem 56:50–58
Yuan X, Carter BP, Schmidt SJ (2011) Determining the critical relative humidity at which the
glassy to rubbery transition occurs in polydextrose using an automatic water vapor sorption
instrument. J Food Sci 76(1):78–89
Understanding Cryopreservation of Oyster
Oocytes from a Physical Chemistry
Perspective
Abbreviations
CF Carboxyfluorescein
CPA Cryoprotection agents
DPPC Dipalmitoylphosphocholine
EPC Egg yolk phosphatidylcholine
IIF Intracellular ice formation
LUV Large unilaminar vesicles
TEM Transmission electron microscopy
1 Introduction
300
RH = 0%
250 RH = 7.0 %
150
RH = 32.8 %
100
RH = 43.2 %
50 RH = 52.9 %
Endo.
↓
RH = 57.6 %
0
270 290 310 330 350 370 390 410 430
Temperature (°C)
Fig. 1 Schematic representation of a cell undergoing a freezing process. Water and solute content
of the cell and its extracellular medium are at equilibrium before freezing. As temperature goes
below the freezing point of the medium, ice is usually formed extracellularly, initially. As a result,
solutes are extracellularly being frozen in concentrated form. An osmotic gradient created in this
way causes the intracellular water to diffuse out of the cell. Intracellular ice may be formed if there
is remaining water in the cell when the temperature goes below its freezing point. s ¼ solutes,
w ¼ water, blue triangles ¼ ice crystals
The rate of cooling of the cellular materials affects the rate of ice formation, which
usually occurs spontaneously in the extracellular matrix. The rate of extracellular
freezing, in turn, affects the rate of cell dehydration, concentration of intra- and
Understanding Cryopreservation of Oyster Oocytes from a Physical Chemistry. . . 217
Oyster oocytes were collected from Pacific oysters (Crassostrea gigas) from
marine farms throughout New Zealand. Oocyte viability (% fertilized) was assessed
at each of the following steps of the cryopreservation process, where a subsequent
step is an additional step to the previous ones: (a) after loading with cryoprotection
agents (CPA) and equilibration for 10 min at 20 C, (b) after cooling to 10 C at
1 C min1 and before seeding, (c) after a 5 min hold at 10 C and seeding, (d) after
cooling to 35 C at 0.3 C min1, and (e) after plunging into liquid nitrogen. A
control treatment was included by fertilizing fresh oocytes.
The results of the ability of oocytes to fertilize after each of the steps in the
standard protocol show that the highest fertilization rate was recorded for the
control with fresh oocytes (84.0 2.3 %); steps a, b, and c were not significantly
different from the control (P > 0.05). Lowest fertilization was recorded for step e,
where oocytes were plunged into liquid nitrogen (36.1 5.0 %), and step d was not
significantly from step e. Steps d and e were both significantly different from the
control.
100
90
80
Cumulative frequency (%)
70
60 30 ºC min-1
50 5 ºC min-1
40 0.3 ºC min-1
30
20
10
0
0 -10 -20 -30 -40 -50 -60 -70
Temperature (ºC)
Fig. 2 Effect of cooling rate on the temperature of intracellular ice formation in oyster oocytes.
Oocytes were loaded with 10 % ethylene glycol in deionized water and cooled to 10 C, then
cooled to 70 C at 30, 5 or 0.3 C min1 (data represent a mean of SEM; n ¼ 3)
Understanding Cryopreservation of Oyster Oocytes from a Physical Chemistry. . . 219
For oocytes cooled at 30 C min1, cell darkening was observed between 38.0
and 52.0 C, with a median value of 46.6 0.2 C. For oocytes cooled at
5 C min1, darkening occurred between 36.4 and 60.3 C, with a median
value of 53.9 0.1 C. No cell darkening was observed when oocytes were
cooled at 0.3 C min1. These results indicated that the oocytes did not form
intracellular ice when they were cooled at the very slow rate of 0.3 C min1.
Verification of the absence of IIF in oocytes cooled at 0.3 C min1 was carried out
in the TEM (Salinas et al. 2008). Under the TEM, ice crystals were observed as
white, round or oval structures homogeneously spread throughout the cell (Fig. 3a, b)
and were distinct from the abundant lipid droplets, which appeared as black and dark
gray, round structures. Although it was found that all oocytes contained ice crystals in
the cytoplasm, only some of them contained ice in the nucleus. This observation led
to the development of the cooling programs (Table 1) to target IIF reduction.
As a result of the above studies, the cooling rate to the plunging temperature (step d)
and the plunging temperatures (step e) were further investigated. Variables such as
a slower post-extracellular ice formation-cooling rate (0.1 C min1 versus
0.3 C min1); longer holding time at the plunge temperature (30 min versus
0 min) and lower plunge temperature (60 C versus 35 C) were assessed as
outlined in Table 1. Samples corresponding to Programs A and H were analyzed in
detail on the TEM.
The average post-thaw fertilization rates are given in Table 2. These rates were
plotted against the total time of exposure to extracellular ice formation (Fig. 4).
Table 2 shows that programs A, B E, and F, which are oocytes with plunging
temperature at 35 C have significantly higher post-thaw fertilization rates than
those plunged at 60 C (programs C, D, G, and H). This result can be better
illustrated in Fig. 4, where the overall post-thaw fertilization rate of oocytes
plunged at 35 C (data clustered in ovals) is higher than that of oocytes plunged
at 60 C (data clustered in rectangles), indicating it was critical that the plunging
temperature of the samples into liquid nitrogen cannot be too low. Groups that were
plunged at 60 C (programs G and H) yielded the lowest post-thaw fertilization
rates. The figure also shows a definitive lower fertilization rate of the samples
plunged at 60 C and cooled at 0.1 C min1 as opposed to the rest of the
treatments, which could be an indication of the oocytes being in a freeze concen-
trated condition for a much longer period than the other treatments. On the other
hand, the holding time for 30 min at the plunging temperature was not significant in
affecting the fertilization rate.
220 M.H. Lim et al.
Fig. 3 Transmission electron micrographs of oyster oocytes after freeze substitution. Oocytes
were frozen using the protocol of Tervit et al. (2005). Instead of thawing the samples, sections of
the straws were cut under liquid nitrogen, freeze substituted, and prepared for electron microscopy.
For visual simplicity, only the relevant structures are labeled in (a) (IC ice crystals, N nucleus, LD
lipid droplets, PM plasma membrane). Scale bar on the bottom right measures 5 μm
A further detailed TEM analysis of oocytes from programs A and H shows that
despite the fertilization rate in program A being significantly higher than H, the
mean diameter of the ice crystals in the cytoplasm was found not to differ signif-
icantly between the two programs (P > 0.05). More interestingly, the amount of ice
in the cytoplasm and the proportion of oocytes with ice crystals in the nucleus were
significantly higher in the A program than in the H program (P < 0.05). Therefore,
it was evident that IIF in oyster oocytes was not detrimental for their survival during
cryopreservation.
Table 1 Cooling programs used to reduce intracellular ice formation in oyster oocytes
Cooling rate Holding time Plunge temperature
Program ( C min1) (min) ( C)
Aa 0
35
B 30
0.3
C 0
60
D 30
E 0
35
F 30
0.1
G 0
60
H 30
a
Standard protocol
60
50 A
B
Fertilization (%)
40 C
D
30
E
20 F
G
10 H
0
0 100 200 300 400 500 600
Time post-EIF (min)
Fig. 4 Effect of time of exposure post-extracellular ice formation (EIF) on the post-thaw
fertilization rate of oyster oocytes cooled with different programs (solid figures indicate a cooling
rate of 0.3 C min1; clear figures indicate 0.1 C min1; same shape of figures indicate same
holding time; data clustered in ovals indicate a 35 C plunge into liquid nitrogen, data clustered
in rectangles indicate a 60 C plunge)
222 M.H. Lim et al.
In summary, despite Pacific oocytes being cooled with the standard protocol
undergoing IIF, they still maintain a high post-thaw fertilization rate. This indicates
that IIF was not detrimental to the viability of the Pacific oyster oocytes. These
results support previous reports that state that intracellular ice is not necessarily a
main cause of cell death (Muldrew et al. 2004). The current study also shows that
reducing the concentration of solutes in the intracellular and extracellular unfrozen
matrix by cooling at a slower cooling rate, and to lower plunging temperatures,
decreased the fertilization most significantly. This could probably be because of the
prolonged detrimental effect of the highly concentrated unfrozen matrix on the
oocytes, suggesting that Pacific oyster oocytes could survive IIF more than solute
effects and cell shrinkage.
Liposomes were used as a simplified membrane model system for studying freeze
injury at the plasma membrane level. As phosphatidylcholine is the major compo-
nent in cell membranes, large unilaminar vesicles (LUV) made from egg yolk
phosphatidylcholine (EPC) and dipalmitoylphosphocholine (DPPC) were used in
this study. EPC consisted of a mixture of acyl chain lengths, which resemble the
composition of biological membranes. DPPC was also used, as a comparison, since
intraliposomal water was retained in the DPPC LUV and froze around 40 C
(Kristiansen and Westh 1991; Kristiansen 1992), which mimicked intracellular
freezing at fast cooling rates. Carboxyfluorescein (CF) was encapsulated as a
marker for monitoring LUV stability. Details of methods and results in this section
can be found in Siow et al. (2007).
The study showed that there is an abrupt increase in leakage of EPC LUV between
5 and 10 C (Fig. 5), which was demonstrated to coincide with the temperatures
of extraliposomal ice formation, as observed at around 7 C from a DSC scan
(results not shown).
In addition to the detrimental effect that could arise from the extra-liposomal ice
formation (Sterling 1968; Rubinsky et al. 1990), numerous previous studies have
shown that leakage increases at the gel–liquid-crystal phase transition temperature
of the phospholipid bilayer (Siminovitch and Chapman 1971; Higgins et al. 1986).
In the current study, the abrupt increase between 5 and 10 C coincided with the
EPC undergoing phase transition at 8 C during cooling from room temperature to
60 C and at 15 C during heating from 60 to 20 C, as determined by DSC
Understanding Cryopreservation of Oyster Oocytes from a Physical Chemistry. . . 223
Fig. 5 Effect of cooling rates on the CF leakage of EPC LUV. Approximately 20 mg of the EPC
LUV was cooled from 20 C to the respective temperatures at 0.5 and at 10 C min1. Samples
were held for 1 min at the final temperature before they were heated to 20 C at 10 C min1. n ¼ 3
(results not shown). This result was in accordance with previous studies (Ladbrook
and Chapman 1969; Chapman 1975).
As water forms ice, the phase volume of the unfrozen matrix decreases. The
reduction of the phase volume of the unfrozen matrix freeze-concentrate the EPC
LUV, eventually leading to the aggregation of vesicles (Fig. 6b). The effect of
freeze-concentration of the EPC LUV, however, did not seem to lead to a higher
leakage of the EPC LUV as temperatures decreased to lower temperatures at
10 C min1 (Fig. 5). At a slower cooling rate of 0.5 C min1, the leakage pattern
was the same as for 10 C min1. However, the osmotic pressure difference across
the bilayers caused an additional stress to the LUV, and therefore higher leakage
was observed.
The increase in leakage of EPC LUV with increased holding time at 15 C (Fig. 7)
was probably due to the fluid and deformable nature of the bilayer of EPC LUV.
EPC consists of phospholipids with a mixture of hydrocarbons of different chain
lengths, resulting in a broad phase transition temperature (Ladbrook and Chapman
1969; Chapman 1975). As mentioned above, the phase transition of the EPC
dispersion was observed at 8 C when the sample was cooled from room temper-
ature to 60 C, and at 15 C during heating from 60 to 20 C (results not
shown). We suggest that the CF leaked out from the fluid and deformable vesicles at
15 C, and that its bilayer was in dynamic and fluctuating gel–fluid phases that
coexist close to the phase transition. This observation was also made by Leidy
et al. (2001). The fluid and deformable nature of the EPC LUV at 15 C can be
noted in Fig. 6a, in which some of the EPC LUV were flattened instead of retaining
their spherical shape. At lower temperatures, the EPC LUV were likely to become
224 M.H. Lim et al.
Fig. 6 Scanning electron micrographs of EPC LUV and DPPC LUV that were cooled to 15,
40, and 55 C at 10 C min1. (a) EPC LUV at 15 C; arrows show the squashed and
flattened EPC LUV, (b) EPC LUV at 55 C, (c) DPPC LUV at 15 C, (d) DPPC LUV at
40 C; arrows show the freeze-concentrated DPPC LUV at the unfrozen channels. The inset in
the micrographs shows a magnified unfrozen channel
Fig. 7 Effect of holding time on the CF leakage of EPC LUV. EPC LUV (approximately 20 mg)
were cooled from 20 C to 15 C or 30 C at 10 C min1. Samples were held at the defined
temperatures (15 or 30 C) for the respective holding time before they were heated to 20 C at
10 C min1. n ¼ 3
Understanding Cryopreservation of Oyster Oocytes from a Physical Chemistry. . . 225
more solid, hence preventing more leakage at temperatures below 30 C. This
inference is obtained from the electron micrographs, where the vesicles appeared to
be less deformable at 30 C than at 15 C (Fig. 6a and b, respectively).
As for the DPPC LUV, leakage that started at 10 C, followed by the subsequent
rapid increase in leakage between 10 and 25 C (Fig. 8), was also near the
temperature of extraliposomal ice formation that occurred at around 9 C (from
DSC scans, not shown). A similar observation has been reported in a previous study
in which a significant leakage of the hydrogenated soybean phosphatidylcholine
liposomes was detected at the extraliposomal ice formation temperature of between
6 and 8 C (Fransen et al. 1986). Since nucleation is a stochastic event, a broad
temperature range of the extraliposomal ice formation was observed in the current
study.
In contrast to EPC, DPPC undergoes the phase transition at 41 C (Chapman
et al. 1967; Mason and Huang 1981; Crowe et al. 1989). Therefore, extraliposomal
ice formation was believed to be solely responsible for the leakage of DPPC LUV
that started at 10 C and the subsequent rapid increase of leakage between 10
and 25 C (Fig. 8). The detrimental effect of the extracellular ice has been
previously reported (Sterling 1968; Rubinsky et al. 1990). Such deleterious effects
of the extraliposomal ice was apparent for the DPPC LUV, but not for the
EPC LUV.
The freeze-concentrated, compressed, and aggregated DPPC LUV in the unfro-
zen matrix (Fig. 6c, d) contributed to the gradual increase in leakage of DPPC LUV
as the increase in ice phase volume as temperatures were lowered to 40 C at
10 C min1 (Fig. 8). It should be noted that the reduction of the phase volume of
25
CF leakage (%)
20
15
10
0
0 -10 -20 -30 -40 -50 -60
Temperature (°C)
Fig. 8 Effect of subzero temperatures on the CF leakage of the dispersion of DPPC LUV.
Approximately 20 mg of the EPC LUV was cooled from 20 C to the respective temperatures at
10 C min1. Samples were held for 1 min at the final temperature before they were heated to
20 C at 10 C min1. n ¼ 3
226 M.H. Lim et al.
the unfrozen matrix also caused an increase of salt concentration in the unfrozen
matrix (Meryman 1974; Mazur 1984). The increase of salt concentration and the
subsequent increase in osmotic pressure in the unfrozen matrix had been ascribed to
the promotion of cellular dehydration, which accounted for the slow freeze-injury
involving freezing rates of below 1 /min (Meryman 1974).
In the current study, extraliposomal ice formation of the DPPC LUV was
observed at 9 C, and intraliposomal freezing was observed at 43 C when
cooled at a relatively fast cooling rate of 10 C/min (Fig. 8). Consequently, the
DPPC LUV could have been exposed to osmotic stress across the bilayer upon
extraliposomal ice formation and before intraliposomal ice formation. Exposure of
DPPC LUV to osmotic stress, however, was short, since the cooling rate was
relatively fast. In Fig. 6c, d, the DPPC LUV appeared to be solid spherical at
15 and 40 C. Thus, lesion and leakage of the DPPC LUV was mainly caused by
the freeze-concentration and compression effects, which were a result of the phase
volume reduction of the unfrozen matrix. The decrease and leveling off in leakage
of DPPC LUV at temperatures below 40 C coincided with the temperature of
intraliposomal freezing (from DSC scans, not shown). This result suggested that
intraliposomal freezing prevented CF diffusion, as the DPPC LUV were thawed
from subfreezing temperatures. The decrease of leakage, in particular, was due to
the DPPC LUV being both intra- and extraliposomally frozen, thus preventing the
diffusion of CF before both the intra- and extraliposomal ice were melted. In
comparison, DPPC LUV that were only frozen in the extraliposomal space at
temperatures above 43 C allowed CF diffusion from the intraliposomal to the
extra-liposomal space before the melting of the extraliposomal ice.
There was relatively low leakage for DPPC LUV held at 15 C compared to the
other subzero temperatures (Fig. 9). The leakage was also lower than for the EPC
LUV that were held at the same temperature. DPPC are phospholipids with
16 carbons in their acyl chain. Therefore, DPPC is at the gel phase at room
temperature and changes into the liquid-crystal phase at a much higher temperature
(41 C). At 15 C, DPPC LUV were in the gel phase and were likely to be more
solid than the EPC LUV. This was supported by the electron micrographs in which
the DPPC LUV appeared to be spherical and less deformable than the EPC LUV at
15 C (Fig. 6a, c). Thus, CF could remain trapped and, as a result, low leakage
was observed.
The percentage of CF leakage of DPPC LUV that was held at 50 C was lower
compared to those held at 40 C (Fig. 9). Intraliposomal freezing of the DPPC
LUV found at 43 C (Fig. 8) may prevent further CF diffusion at 50 C.
Supporting the results found in the current study, Kristiansen (1992) found that
below 40 C, leakage of DPPC multilamellar vesicles leveled off because the
matrix solidification was complete. Similar observations were also made for both
egg lecithin liposomes (Higgins et al. 1986) and multilamellar liposomes consisting
Understanding Cryopreservation of Oyster Oocytes from a Physical Chemistry. . . 227
Fig. 9 Effect of holding time on the CF leakage of DPPC LUV. DPPC LUV (approximately
20 mg) were cooled from 20 C to 15 C or 40 C or 50 C at 10 C min1. Samples were
held at the defined temperatures (15 or 30 C) for the respective holding time before they were
heated to 20 C at 10 C min1. n ¼ 3
3 Overall Summary
The use of large unilamellar, EPC and DPPC LUV offers suitable membrane model
systems for the study and improvement of cryopreservation procedures. For exam-
ple, when a slower rate of freezing was used, a higher leakage was observed and this
relates to the cellular dehydration induced by the osmotic stress at slow cooling
rate, which was conducive to prevent intracellular freezing if cells could withstand
the resultant osmotic shrinkage. Conversely, since we learned from the Pacific
oyster oocytes study that they could survive IIF more than solute effects and cell
shrinkage, an appropriate freezing rate could be investigated to produce an optimal
rate using the model system.
The variations in the phase transition of both the EPC and DPPC LUV illustrated
that mechanical damage at low subfreezing temperatures due to the extracellular ice
could be avoided by having a more fluid membrane. On the other hand, if the cells
228 M.H. Lim et al.
have a more fluid membrane structure, they are more permeable or “leaky” at
higher subfreezing temperatures. Therefore, this study suggests that manipulating
the cell membrane fluidity/solidity of oocytes by changing the diet of oysters could
be one way to control the rate of water diffusion in and out of the cells under a
controlled freezing condition.
To have a closer representation of a biological membrane, liposomal model
systems need to show increased complexity by including other cell membrane
components, such as more heterogeneous phospholipids, proteins, and sugars.
Such increased complexity would allow a better understanding of the dynamics
and prediction of the kinetics of cell freezing. In this study, both the cellular and
liposomal studies showed that high phase volume of ice could cause cell distortion
and compression with the ice channels in the concentrated unfrozen matrix. There-
fore, a careful study of the ice dynamics could be a future focus to improve the
cryopreservation procedures.
References
Acker JP, McGann LE (2003) Protective effect of intracellular ice during freezing. Cryobiology
46:197–202
Chapman D (1975) Phase transitions and fluidity characteristics of lipids and cell membranes.
Q Rev Biophys 8:185–235
Chapman D, Williams RM, Ladbrooke BD (1967) Physical studies of phospholipids.
VI. Thermotropic and lyotropic mesomorphism of some 1,2-diacyl-phosphatidylcholines (lec-
ithins). Chem Phys Lipids 1:445–475
Crowe JH, McKersie BD, Crowe LM (1989) Effects of free fatty acids and transition temperature
on the stability of dry liposomes. Biochim Biophys Acta 979:7–10
Fahy GM, Lilley TH, Linsdell H, John Douglas MS, Meryman HT (1990) Cryoprotectant toxicity
and cryoprotectant toxicity reduction: in search of molecular mechanisms. Cryobiology
27:247–268
Fransen GJ, Salemink PJM, Crommelin DJA (1986) Critical parameters in freezing of liposomes.
Int J Pharm 33:27–35
Gao D, Critser JK (2002) Mechanisms of cryoinjury in cells. ILAR J 41:187–196
Hagedorn M, Peterson A, Mazur P, Kleinhans FW (2004) High ice nucleation temperature of zebra
fish embryos: slow-freezing is not an option. Cryobiology 49:181–189
Higgins J, Hodges NA, Olliff CJ, Phillips AJ (1986) Factors influencing cryoprotective activity
and drug leakage from liposomes after freezing. J Pharm Pharmacol 38:259–263
Kristiansen J (1992) Leakage of a trapped fluorescent marker from liposomes: effects of eutectic
crystallization of NaCl and internal freezing. Cryobiology 29:575–584
Kristiansen J, Westh P (1991) Freezing behavior of multilamellar vesicles in 0.9 percent sodium
chloride. Cryo-Lett 12:167–176
Ladbrook BD, Chapman D (1969) Thermal analysis of lipids, proteins and biological membranes.
A review and summary of some recent studies. Chem Phys Lipids 3:304–367
Leidy G, Wolkers WF, Jorgensen K, Mouritsen OG, Crowe JH (2001) Lateral organization and
domain formation in a two component lipid membrane system. Biophys J 80:1819–1828
Lovelock JE (1953) The haemolysis of human red blood cells by freezing and thawing. Biochim
Biophys Acta 10:414–426
Mason JT, Huang CH (1981) Chain-length dependent thermodynamics of saturated symmetric
chain phosphatidylcholine bilayers. Lipids 16:604–608
Understanding Cryopreservation of Oyster Oocytes from a Physical Chemistry. . . 229
N.E. Zaritzky
Abbreviations
1 Introduction
Plant genetic resources are vitally important for human beings and the sustainability
of the planet. Biodiversity conservation is the practice of protecting and preserving
the abundance and variety of all species (Lambardi et al. 2004; Walters 2006).
Sensitivity to desiccation and low temperatures limits the storage potential of seeds
and their genetic conservation. Research on over 9,000 plant species has demon-
strated that seeds can be grouped according to their storage behavior in both
orthodox seeds and nonorthodox seeds. A great number of tropical species are
nonorthodox seeds, including recalcitrant (Roberts 1973) and intermediate seeds
(Ellis et al. 1990, 1991). The main characteristics of these seeds, according to
Bonner (2008) include the following:
– True orthodox seeds can be stored for long periods at seed moisture contents of
5–10 % and subfreezing temperatures.
– Sub-orthodox seeds can be stored under the same conditions, but for shorter
periods due to high lipid content or thin seed coats.
– Temperate recalcitrant seeds cannot be dried at all, but can be stored for 3–5
years at near-freezing temperatures.
– Tropical recalcitrant seeds cannot be dried, and they do not survive at temper-
atures below 10–15 C.
Orthodox seeds acquire desiccation tolerance during development, can be dried
to low water contents and retain viability in the dry state for predictable periods
(Vertucci and Roos 1990). Examples of orthodox seeds include bean (Phaseolus
vulgaris L.), pea (Pisum sativum), mungbean (Vigna radiata), sunflower
(Helianthus annuus), and soybean (Glycine max). Most agricultural crops produce
orthodox seeds, since they undergo a drying phase as they mature on the plant and
usually contain around 20–30 % water when mature and are ready to harvest. After
collection, orthodox seeds survive further drying to at least 5 % water content.
Mature orthodox seeds can be dehydrated without damage to very low levels of
moisture. The stability of orthodox seeds in the dry state has been a crucial factor in
the development of agriculture and human civilization. Low water content is known
to slow deteriorative chemical reactions in dry seeds. Studies of seed storage over
two centuries have confirmed this observation and thus the preservation of seeds at
low or ultra-low water content has been advocated.
Intracellular glasses have been detected in a number of seeds. High levels of
sugars and other biopolymers in seeds result in a rapid increase in cytoplasmic
viscosity during drying, which prevents the cellular biological system from
reaching physical and chemical equilibrium in a measurable time frame. As a
result, cytoplasmic components become vitrified, and intracellular glasses form
during drying. It has been suggested that the formation of intracellular glasses may
be a strategy for seeds to survive desiccation (Burke 1986; Williams and Leopold
1989; Sun 1997). Cytoplasmic vitrification minimizes major changes in molecular
organization and cellular structures during dehydration, thus leading to the preser-
vation of biological structures. The high viscosity of the glassy state immobilizes
cellular constituents, thus inhibiting diffusion and slowing deleterious reactions or
234 N.E. Zaritzky
100
80
60
40 Viscous liquid
Temperature ( °C)
20
-20
-40
-80
-100
0 0.1 0.2 0.3 0.4 0.5
water content g g-1 (dry weight)
Fig. 1 Schematic representation of glass transition curve for different citrus seeds, based on data
of three seed species: Glycine max, Phaseolus vulgaris, and Pisum sativum, as reported by Bruni
and Leopold (1991, 1992), Sun and Leopold (1994), Sun et al. (1994), Leprince and Walters-
Vertucci (1995), Williams and Leopold (1995), and Sun (1997)
The Role of Water in the Cryopreservation of Seeds 235
Sun and Leopold (1994); Sun et al. (1994); Leprince and Walters-Vertucci (1995);
and Williams and Leopold (1995). The effect of water content on storage stability
appeared to be largely related to the plasticization effect of water on intracellular
glasses in orthodox seeds.
In contrast, recalcitrant seeds do not undergo maturation drying maintaining a
high water content; that is, they do not experience a reduced cellular metabolism
and are sensitive to desiccation and to low temperatures (Pammenter et al. 1991;
Pammenter and Berjak 2000). Recalcitrant seeds cannot be dried and cannot be
stored at subzero temperatures because they are damaged by freezing injury
resulting from ice formation. Examples of recalcitrant seeds included avocado
(Persea Americana), mango (Mangifera indica L.), cacao (Theobroma cacao),
and tea (Camellia sinensis). Due to the active metabolism in the hydrated state,
viable recalcitrant seeds cannot be stored for long-term periods. The longevity of
recalcitrant seeds is short, from a few weeks to a few months for species adapted to
tropical environments, and up to about 3 years for several species adapted to
temperate environments. Recalcitrant seeds neither tolerate nor survive desiccation.
The dehydration of recalcitrant tissues produces membrane deterioration
(plasmalema and mitochondria), protein denaturation, reduction of respiratory
rate, and reduction in ATP level. When freshly harvested recalcitrant seeds are
dried, viability is reduced considerably at certain moisture content (“critical mois-
ture content” or “lowest safe moisture content”). If drying continues further,
viability is eventually reduced to zero. The sensitivity of recalcitrant seeds to low
temperatures is due to their high water content. Critical moisture contents for loss of
viability on desiccation vary greatly among recalcitrant species, as well as among
cultivars and seed lots, depending on the stage of seed maturity at time of collec-
tion. Recalcitrant seeds are present in at least 70% of tropical trees. Oxidative
processes and free radical reactions seem to be involved in cellular and molecular
deterioration; seeds show a strong resistance to rehydration, and loss of cellular
integrity leads to a loss of viability. The water content at which recalcitrant seeds
start to lose viability is relatively high; it is far above that at which a glassy state can
exist at room temperature and nonfreezable water is lost.
Recalcitrant seeds are characterized by high water content at maturity 0.43–
4.0 gH2O g1 (dry basis, db), which is 30–80 % on a wet mass basis (wb); they
cannot survive drying below around 20–30 % water content (wb). For example,
Avicennia marina seeds are unable to survive water content lower than 0.5 gH2O g1
(db) (33 % wb); cacao (Theobroma cacao) seeds rapidly lose germination when
they are dried to 0.26 gH2O g1 db. Highly recalcitrant seeds are unable to survive
water content lower than 0.5 gH2O g1 db (33 % wb). Recalcitrant seeds are unable
to survive below 0.42–0.25 gH2O g1 (30–20 % wb).
Recalcitrant and orthodox seeds differ greatly in their ecology and morphology.
Recalcitrant seeds primarily come from perennial trees in the moist tropics. In some
cases, they also come from temperate tree or aquatic species. Most orthodox seeds
come from annual species grown in open fields. With respect to morphology,
recalcitrant seeds differ from orthodox seeds not only in size, but also complexity
and viability. Generally, recalcitrant seeds are covered with fleshy or juicy layers
and impermeable testa. These structures maintain the seeds in a high-moisture
236 N.E. Zaritzky
0.7
Intermediate
0.6
0.5
0.4 Orthodox
0.3
0.2
0.1
0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
-1 )
water content (g g wet basis
Fig. 2 Schematic diagram showing approximate values of safe moisture levels (minimum water
content) for recalcitrant, intermediate, and orthodox seeds, expressed on a dry or a wet basis
but at shedding was still sensitive; and Phaseolus vulgaris, an orthodox seed that
increased in tolerance with development, and during maturation drying became
even more tolerant.
Moisture level below 0.5 g g1 db (33 % wb) was lethal for the highly recalci-
trant tropical species Avicennia marina, a tropical wetland species commonly
known as gray mangrove or white mangrove, a species of mangrove tree that
develops in the zones of estuarine areas. The moderately recalcitrant Aesculus
hippocastanum is a temperate terrestrial species that can tolerate dehydration
between 0.42 and 0.25 g g1 db (30–20 % wb). The orthodox seed Phaseolus
vulgaris tolerates water contents as low as 0.08 g g1 db (9 % wb) without
viability loss.
Study of these seeds was conducted at three developmental stages: stage
1, immediately post histo-differentiation; stage 2, in the middle of the reserve
accumulation phase; and stage 3, at the end of reserve accumulation. Different
processes or mechanisms are proposed to confer protection against desiccation in
seeds; their deficiency or absence could contribute to relative degrees of desiccation
sensitivity.
The most important intracellular physical characteristics of dehydration resis-
tance in seeds were determined according to Pammenter and Berjak (2000):
– Reduction of the degree of vacuolation that increases the mechanical resilience
of cells to dehydration. Avicennia marina is a highly recalcitrant seed and one of
the species most sensitive to desiccation. It is highly vacuolated, and this
condition does not change with development, supporting the concept that drying
of highly vacuolated material leads to mechanical damage. In contrast, the more
tolerant species showed a decline in vacuolation with development.
– Integrity of the cytoskeleton, which is formed by microtubules and microfila-
ments; it is an integrated intracellular support system, and contributes to the
organization of the cytoplasm and the nucleus.
– Conformation of DNA, chromatin, and nuclear architecture: maintenance of the
integrity of genetic DNA material in the desiccated condition in orthodox seeds,
and/or its rapid repair when seeds are dehydrated, is considered to be a funda-
mental requirement for desiccation tolerance.
– Intracellular de-differentiation is a characteristic of maturing desiccation-tolerant
seeds. Intracellular structures are simplified and minimized (minimization of
surface areas of membranes and mitochondria) in maturing desiccation-tolerant
seeds. Mitochondria of the root meristem cells of recalcitrant seeds were highly
differentiated and had the appearance of active mitochondria, while those in
orthodox seeds were de-differentiated and appeared inactive. In orthodox
seeds, there was a decline in the contribution of mitochondria to cell volume
with development, and by the end of reserve accumulation, the degree of differ-
entiation of mitochondria was very low.
– “Switching off” of metabolism: Decrease in respiratory rate is an essential event
enabling an orthodox seed to withstand rapid loss of water. Recalcitrant seeds
are, in contrast, metabolically active. Measured respiration rates were high in the
238 N.E. Zaritzky
Species that are freezing or desiccation tolerant have been observed to accumulate
solutes, especially sucrose and trehalose. For nonorthodox seed species, cryopreser-
vation is the only technique available for long-term germplasm conservation. In the
case of intermediate seed-propagated species, seeds are partially desiccation tolerant
and, therefore, the whole seed cryopreservation is the first option to be tested.
Seed moisture content and germination conditions need to be optimized to
maximize freezing tolerance and recovery following the cryopreservation process
(Dussert et al. 2001; Hor et al. 2005).
240 N.E. Zaritzky
Graiver et al. (2011), analyzed the optimal moisture content hydration status for
cryopreservation of different Citrus seeds: Citrus sinensis (sweet orange), Citrus
paradise (grapefruit), Citrus reticulata var. Criolla and Citrus reticulata var. Dancy
(mandarin).
The tolerance (viability) of these seeds to desiccation at different relative
humidities was compared to the viability of the seeds that were desiccated prior
to the LN treatment.
Seeds were extracted manually from freshly harvested mature fruits of the
different Citrus species. After extraction, seed were surface-sterilized by immersion
in ethanol aqueous solution, followed by treatment in sodium hypochlorite aqueous
solution. Seeds were then rinsed twice in tap water and in distilled water and
immediately surface dried. To analyze tolerance to desiccation, seeds without
testa (endocarp) were placed under desiccation conditions by equilibration at
20 C over seven saturated salt solutions (LiCl, MgCl2, K2CO3, NaNO2, NaCl,
KCl, KNO3) with constant equilibrium relative humidity (ERH%) of 11, 32, 43,
64, 75, 85, and 95 %, respectively. Water content of the seeds (WC) was determined
gravimetrically after oven-drying the seeds at 103 C until constant weight and
expressed on dry basis (gH2O g1 db). A desiccation period of 35 days was applied
for the different assayed ERH, in order to achieve equilibrium conditions (constant
weight). Water sorption isotherms were obtained from the data of WC and the
corresponding ERH of the saturated salt solutions; the curves were modeled using
Guggenheim-Anderson-de Boer (GAB) equation and the simplified D´arcy and
Watt model proposed by Dussert et al. (2001).
Seed viability was analyzed using germination experiments. Seeds were sown in
hermetic controlled germination conditions (humid sand in covered plastic box and
kept in a growth chamber at 25 C in the dark). The percentage of normal seedlings
was evaluated 4–6 weeks after sowing.
Figure 3 shows the effect of the desiccation ERH on the viability of one of the
tested citrus seeds (grapefruit, Citrus paradise). Similar results were reported by
Graiver et al. (2011) for the other tested seeds. As can be observed, as the ERH% of
the desiccation stage increased, the number of germinated seeds was higher. A
decline in the germination percentage was observed for all of the tested Citrus seeds
when desiccation was conducted at ERH < 75 %. Seed desiccation sensitivity
(WC50) was quantified by the quantal response model of Dussert et al. (1999).
In order to analyze the feasibility of cryopreservation of citrus seeds, the
viability of seeds (germinability) submitted to different levels of desiccation,
followed by LN treatment was measured. To study the tolerance to LN exposure,
seeds were previously desiccated by equilibration at 20 C over the seven saturated
salt solutions, wrapped in aluminum foil and then immersed in LN 1 h. After the
cooling period, seeds were immersed for 5 min in a water-bath at 37 C and directly
placed under germination conditions. Figure 3 shows the germinability percentages
of seeds previously desiccated at different ERH, and then submitted to LN; as can
be observed, viability was significantly lower than in the case of seeds that were
only desiccated. A similar pattern of sensitivity to LN exposure after desiccation
was observed for all of the tested citrus seeds. No survival was achieved at the
The Role of Water in the Cryopreservation of Seeds 241
100.0
90.0
80.0
70.0
Seed germinability %
60.0
50.0
40.0
30.0
20.0
10.0
0.0
0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0
Desiccation equilibrium relative humidity %
200
1
180
0 c
160 Heat flow ( W g-1) -1
b
140
-2
a
Enthalpy (J g-1 db)
-3
120 -4
-5
100
-6
-150 -100 -50 0 50 100 150
80 Temperature °C
60
40
20
0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65
Seed moisture content (g H2O g-1 db)
Fig. 4 Total enthalpy measured by DSC as a function of moisture content of seed samples for
C. paradisi ( filled diamond) and C. sinensis (open square). Inset: DSC heating thermograms of
Citrus seed samples: (a) untreated seed, (b) seed desiccated on silica gel, (c) seed lipid extract.
(Adapted from Graiver et al. 2011)
can be noted that the endotherm of the silica gel dehydrated seed was similar to that
of the lipid extract. In addition, lipid and ice melting events are overlapping;
therefore, to calculate the unfrozen water fraction, the lipid enthalpy must be
subtracted from the total enthalpy values. Additionally, in Fig. 4 the peak areas of
the DSC thermograms (total enthalpy) for C. paradisi and C. sinensis seeds that
were dehydrated at different ERH previous to DSC cooling and warming programs
were plotted as a function of the moisture content of the seed samples. It can be
observed that at low moisture contents in the tissue, asymptotic values of enthalpy
(20 J g1 dry basis) can be attributed to the presence of lipids in the seeds were
reached. Unfrozen water content (WCu) was calculated as follows:
ΔH T ΔHL
WCu ¼ WC ð1 þ WCÞ
λ
where: WC ¼ total water content (dry basis); ΔHT ¼ total enthalpy measured by
DSC (J g1 dry basis ); λ ¼ latent heat of ice melting; ΔHL ¼ lipid melting enthalpy
(J g1 dry basis ). The obtained average values of unfrozen water content (WCu)
expressed as g H2O g1 db were 0.14 and 0.13 for C. sinensis and C. paradise,
respectively. The values of WCu in Citrus sp. were found to be negatively corre-
lated to seed lipid content (Hor et al. 2005; Graiver et al. 2011). Finally, it is
important to relate the unfrozen water content in the seeds with LN tolerance.
Using the water sorption isotherms of each citrus seed, the equilibrium relative
humidity (ERH%) of the desiccation atmosphere leading to the WCu determined by
DSC was obtained. For the measured values of WCu (0.14 g H2O g1 db for
The Role of Water in the Cryopreservation of Seeds 243
C. sinensis and 0.13 g H2O g1 db for C. paradisi) the obtained ERH% were 81 %
for C. sinensis and 67 % for C. paradisi . These ERH% are included in the optimum
ranges for seed cryopreservation determined by the germinability tests, resulting in
75–85 % for C. sinensis and 64–75 % for C. paradisi. Therefore, seed survival was
maximized when dehydration previous to LN treatment was performed at the
highest ERH%, leading to the absence of frozen water in the tissue.
6 Conclusions
References
Bonner FT (2008) Storage of seeds. Chapter 4. In: The woody plant seed manual. USDA
Agricultural Handbook 727, USA, pp 85–96
Bruni F, Leopold AC (1991) Glass transitions in soybean seeds: relevance to anhydrous biology.
Plant Physiol 96:660–663
Bruni F, Leopold AC (1992) Pools of water in anhydrobiotic organisms: a thermally stimulated
depolarization current study. Biophys J 63:663–672
Burke MJ (1986) The glass state and survival of anhydrous biological systems. In: Leopold AC
(ed) Membranes, metabolism and dry organism. Cornell University Press, New York,
pp 358–364
Dussert S, Chabrillange N, Engelmann F, Hamon S (1999) Quantitative estimation of seed
desiccation sensitivity using a quantal response model: application to nine species of the
genus Coffea L. Seed Sci Res 9:135–144
Dussert S, Chabrillange N, Rocquelin G, Engelmann F, Lopez M, Hamon S (2001) Tolerance of
coffee (Coffea spp.) seeds to ultra-low temperature exposure in relation to calorimetric
properties of tissue water, lipid composition, and cooling procedure. Plant Physiol
112:495–504
Ellis RH, Hong TD, Roberts EH (1990) An intermediate category of seed storage behaviour?
I. Coffee. J Exp Bot 41:1167–1174
Ellis RH, Hong TD, Roberts EH (1991) An intermediate category of seed storage behaviour?
II. Effects of provenance immaturity, and imbibition on desiccation-tolerance in coffee. J Exp
Bot 42:653–657
244 N.E. Zaritzky
Graiver N, Califano A, Zaritzky N (2011) Partial dehydration and cryopreservation of citrus seeds.
J Sci Food Agric 91(14):2544–2550
Hamilton KN, Ashmore SE, Drew RA (2008) Desiccation and cryopreservation tolerance of near
mature seeds of Citrus garrawayae. Seed Sci Technol 36:157–161
Hamilton KN, Ashmore SE, Pritchard HW (2009) Thermal analysis and cryopreservation of seeds
of australian wild Citrus species (Rutaceae): Citrus australasica, C inodora and C garrawayi.
CryoLetters 30:268–279
Hor YL, Kim YJ, Ugap A, Chabrillange N, Sinniah UR, Engelmann F, Dussert S (2005) Optimal
hydration status for cryopreservation of intermediate oily seeds: citrus as a case study. Ann Bot
95:1153–1161
Lambardi M, De Carlo A, Biricolti S, Puglia AM, Lombardo G, Siragusa M, De Pasquale F (2004)
Zigotic and nucellar embryo survival following dehydration/cryopreservation of Citrus intact
seeds. CryoLetters 25:81–90
Leprince O, Walters-Vertucci C (1995) A calorimetric study of the glass transition behaviours in
axes of bean seeds with relevance to storage stability. Plant Physiol 109:1471–1481
Makeen MA, Normah MN, Dussert S, Clyde MM (2007) The influence of desiccation and
rehydration on the survival of polyembryonic seed of Citrus suhuiensis cv. Limau madu. Sci
Hortic 112:376–381
Pammenter NW, Berjak P (2000) Aspects of recalcitrant seed physiology. R Bras Fisiol Veg
12:56–69
Pammenter NW, Vertucci CW, Berjak P (1991) Homeohydrous (recalcitrant) seeds: dehydration,
the state of water and viability characteristics in Landoiphia kirkii. Plant Physiol 96:1093–1098
Roberts EH (1973) Predicting the storage life of seeds. Seed Sci Technol 1:499–514
Sacandé M, Buitink J, Hoekstra FA (2000) A study of water relations in neem (Azadirachta indica)
seed that is characterized by complex storage behaviour. J Exp Bot 51:635–643
Sun WQ (1997) Glassy state and seed storage stability: the WLF kinetics of seed viability loss at
T > Tg and the plasticization effect of water on storage stability. Ann Bot 79:291–297
Sun WQ, Irving TC, Leopold AC (1994) The role of sugar, vitrification and membrane phase
transition in seed desiccation tolerance. Physiol Plant 90:621–628
Sun WQ, Leopold AC (1993) Acquisition of desiccation tolerance in soybean. Physiol Plant
87:403–409
Sun WQ, Leopold AC (1994) Glassy state and seed storage stability: a viability equation analysis.
Ann Bot 74:601–604
Vertucci CW (1989a) Effects of cooling rates on seeds exposed to liquid nitrogen temperature.
Plant Physiol 90:1478–1485
Vertucci CW (1989b) The effects of low water contents on physiological activities of seeds.
Physiol Plant 77:172–176
Vertucci CW (1989c) Relationship between thermal transitions and freezing injury in pea and
soybean seeds. Plant Physiol 90:1121–1128
Vertucci CW (1990) Calorimetric studies of the state of water in seed tissues. Biophys J
58:1463–1471
Vertucci CW, Farrant JM (1995) Acquisition and loss of desiccation tolerance. In: Galili G,
Kigel J (eds) Seed development and germination. Marcel Dekker, Inc., New York
Vertucci CW, Leopold AC (1987) Water binding in legume seeds. Plant Physiol 85:224–231
Vertucci CW, Roos EE (1990) Theoretical basis of protocols for seed storage. Plant Physiol
94:1019–1023
Walters C (2006) Water properties and cell longevity. In: Buera M, Welti Chanes J, Lillford P,
Corti H (eds) Water properties of food, pharmaceutical and biological materials. PUSA, CRC,
Boca Raton
Williams RJ, Leopold AC (1989) The glassy state in corn embryos. Plant Physiol 89:977–981
Williams RJ, Leopold AC (1995) Changes in glass transition temperatures in germinating pea
seeds. Seed Sci Res 5:117–120
Water Activity and Microorganism Control:
Past and Future
A. L
opez-Malo and S.M. Alzamora
Abbreviations
1 Introduction
spoilage, such as drying, salting, and adding sugars. In some cases, the spoilage of
products subjected to these procedures has also been referred to in the following
ways: halophilic bacteria can grow in salted products; osmophilic yeasts may ferment
sugar in preserved foods; and some dried foods can be spoiled by xerophilic molds
(Mossel 1975). Substantial interest in the influence of water activity (aw) on food
product quality and stability, promoted by empirical observations between total
moisture content and product stability, began during the middle of the last century.
Scott (1957) introduced the concept of aw as a quantitative approach to define the
influence of moisture content on microbial response in foods. Microbiologists rec-
ognized that aw, rather than moisture content, controlled microbial response, as well
as sporulation and/or toxin production (Jay et al. 2005). The relationship between aw
and food-borne microorganisms has been the topic of study by food researchers over
the past several decades (Christian 2000; Chirife 1995; Lenovich 1987). Microbiol-
ogists have investigated how microorganisms respond under different conditions of
temperature, pH, additives, atmosphere composition, and aw (Hocking and Pitt 1987;
Beuchat 1996). The influence of aw in microbial death, survival, sporulation, and
toxin production in food has been extensively studied by food microbiologists
(Lenovich 1987; Beuchat 1983, 1987; Hocking and Christian 1996; Gutierrez
et al. 1995). The aw principle has been included in various government regulations,
with the recognition that control of aw as an important critical control point for risk
analysis, as defined by the HACCP concept, sets up aw limits on food products.
However, trends in food consumption and safety during the last few decades
have significantly impacted the way in which the aw concept is primarily used in
food preservation design. Consumers increasingly demand fresh-like food products
that are less heavily preserved/processed (e.g. with high aw and lower added pre-
servatives), and more convenient and nutritious. To correspond these trends without
compromising safety, a number of combinations of several stress factors, such as
reduced aw, in a multifactorial food preservation approach (“hurdle” technology),
have been developed in the past 20 years. Targeted application of the hurdle
concept has aimed to improving the quality and safety of foods (Alzamora
et al. 2000a, b; Leistner and Gould 2002; Raso and Barbosa-Cánovas 2003; Ross
et al. 2003; Allende et al. 2006; Tapia and Welti-Chanes 2012).
are some extrinsic factors that influence food microbial spoilage and therefore
influence the aw-microbial response. The use of combinations of extrinsic and
intrinsic factors together with lowered aw levels is common in the food industry
(Alzamora et al. 2003). Generally, as the food aw moves toward the microbial
minimal aw for growth, changes in other environmental factors will have a greater
impact on death or survival.
Taking into account aw values related to possible microbial responses, minimum
aw values that allow microbial growth for different types of microorganisms are
essential for product design and stability. Figure 1 displays aw range and its
influence on some microbial responses; similar tables can be consulted elsewhere
(Corry 1973; Beuchat 1983, 1987; Gould 1989a) and can be used to classify
osmosensitive and osmotolerant microorganisms, as well as to find minimum aw
values for growth or toxin production of several pathogenic and spoilage microor-
ganisms. Table 1 shows minimal aw for growth of selected food-borne pathogens
when pH and temperature are optimum for growth (i.e., aw is employed as the only
preservation factor to inhibit microbial growth). At aw < 0.90–0.92, bacteria are
usually inhibited, including pathogenic bacteria, with the exception of Staphylo-
coccus aureus, which can grow at aw 0.86. Bacteria rarely spoil foods with aw
values below 0.85, the exceptions being brines and salted foods that may be spoiled
by moderate and extreme halophilic bacteria (Hocking and Christian 1996). Yeast
and molds are more tolerant to reduced aw, but usually no growth exists below aw of
0.62 (Scott 1957; Hocking and Pitt 1987).
248 A. L
opez-Malo and S.M. Alzamora
D. hansenii death at aw 0.95 or 0.90 independently of solute type. The use of sodium
chloride to adjust aw caused yeast inhibition or death in more cases than the use of
glucose or sucrose. Citric and phosphoric acids allowed yeast growth at aw 0.95
(sodium chloride, glucose, or sucrose) or aw 0.90 (glucose or sucrose). Acid and
solute type exhibited an important effect on D. hansenii growth response, demon-
strating the importance of their selection when used to adjust pH and aw in foods.
As an example of water activity effects on bacterial growth and the effect of
different solutes used to adjust aw, Fig. 2 displays the growth curves of
Enterobacter sakazakii (ATCC 51329) at selected water activity values when
using sodium chloride or glycerol as aw depressing solutes (Delgado-Portales
et al. 2007). Growth curves were affected by aw and by the solute used to adjust
aw. By using the Gompertz equation, parameters of the sigmoideal growth curves
were obtained (Table 3). The resulting values presented the same tendency as the
two solutes tested. A decrease in the growth rate and higher lag times were observed
as aw of the system decreased. A slight reduction in food aw indicates that
microorganisms surviving processing in such foods, or contaminating the food
after processing, will be osmoregulating at some level. Since homeostatic responses
often require the expenditure of energy by the stressed cells, restriction of avail-
ability would be expected to result in lower growth rates. Regarding the effect of
solute type, Brewer (1999) indicated that, in general, sodium chloride is more
inhibitory than glycerol and sucrose for bacteria that are considered halo-sensible
(most spore-formers, Enterobacteriaceae, Pseudomonas fluorescens) and less inhib-
itory than glycerol to halotolerant bacteria (Micrococcaceae, Vibrio). At aw 0.96,
Fig. 2 Enterobacter
sakazakii (ATCC 51329)
growth curves of selected
water activity values when
using sodium chloride or
glycerol as aw depressing
solutes. (Adapted from
Delgado-Portales
et al. 2007)
Table 3 Gompertz parameters for Enterobacter sakazakii (ATCC 51329) growth at selected
water activity values when using sodium chloride or glycerol as aw depressing solutes
Growth rate (μm) Maximum population density
Lag time λ (h) Log (N/No)/h Log (N/No)
Control (aw 0.996) 0.64 0.11 2.64 0.12 4.7 0.1
TSB + NaCl
aw 0.98 1.22 0.13 1.61 0.11 4.8 0.1
aw 0.97 2.20 0.21 1.51 0.09 4.8 0.1
aw 0.96 2.69 0.20 0.90 0.06 4.4 0.1
TSB + glycerol
aw 0.98 0.93 0.11 1.77 0.08 5.8 0.1
aw 0.97 1.52 0.13 1.55 0.07 5.8 0.1
aw 0.96 1.11 0.10 1.18 0.10 5.6 0.1
Water Activity and Microorganism Control: Past and Future 251
a decrease in growth rate and an increase in lag time was observed when media were
adjusted with NaCl. It can be seen that evaluated aw values allow growth and that
minimum aw for growth of this strain is lower than 0.96. Breeuwer et al. (2003)
evaluated the behavior of different strains of E. sakazakii in brain–heart infusion
(BHI) broth adjusted to aw 0.934 with sorbitol and found that minimum aw for
growth was between 0.93 and 0.96, although it should be noted that profound
differences may exist between strains of this organism (Breeuwer et al. 2003;
Buchanan 2003; Edelson-Mammel et al. 2006). Gould (1989a) recognized that in
some instances solute effects may depend on the ability of the solute to permeate the
cell membrane, as in the case of glycerol, which readily permeates the membrane of
many bacteria, and therefore has a lower inhibitory water activity. Chirife (1994)
discussed in detail the “specific solute effect” for S. aureus. He concluded that the
inhibitory effects of solutes most often present in low aw—preserved foods, such as
NaCl and sucrose, are primarily related to their aw lowering capacity. But for other
solutes such as ethanol, propylene glycol, butylene glycol, and various polyethylene
glycols, antibacterial effects (attributed mainly to the effects of these molecules on
membrane enzymes responsible for peptidoglycan synthesis) are important.
The sequence of hurdles has been shown to be crucial for the survival of
pathogens. Tiganitas et al. (2009) studied the effect of lethal and sublethal pH
(adjusted with lactic acid to 4.0 and 4.5) and aw (adjusted with NaCl, 15 and 20 %
NaCl) stresses on the inactivation or growth of Listeria monocytogenes and Sal-
monella Typhimurium, simulating various processing conditions (e.g. ripening of
cheese, fermentation of meat). Sequential exposure to the stresses resulted in faster
reductions than exposure to double stresses applied simultaneously. The pH then
NaCl sequence was more detrimental to pathogens than the reverse sequence. Acid
tolerance of L. monocytogenes was higher than osmotic tolerance, while the
opposite was observed for Salmonella. These results suggest that during food
processing, the sequence and severity of the stresses applied significantly affect
inactivation of L. monocytogenes and Salmonella. These findings are not in agree-
ment with previous reports in which adaptation to acid was shown to confer cross-
tolerance to other stresses, such as heat and salt (Leyer and Johnson 1993). Rather,
such protection was linked to HCl and acetic acid adaptation, instead of adaptation
to lactic acid. These results evidence the necessity of further research in situations
in which pathogens undergo exposure to multiple hurdles.
Better knowledge of the relation of water in foods has led to the rediscovery and
optimization of older preservation techniques and to a renewed interest in foods that
are shelf-stable due to control of water activity (Alzamora et al. 2003), including
dried, intermediate and high moisture foods with inherent empiric preservation
factors, and also to novel products (especially food products with high water
activity), for which the hurdles are intentionally selected and applied. In the search
for foods with superior quality, the importance of considering the combined action
of reduced aw with other preservation factors as a system to develop new food
products is a current trend in food processing. Minimally processed foods represent
a new variety of products that has responded to these demands. Therefore, the
control of aw for food design is being used in many ways (Alzamora et al. 2003),
252 A. L
opez-Malo and S.M. Alzamora
i.e., at various stages of the food distribution chain, during storage, processing,
and/or packaging as a “back-up” hurdle in existing minimally processed products
with short shelf-life to diminish microbial pathogenic risk and/or increase their
shelf-life; to obtain products with long shelf-life (fully dehydrated and intermediate
moisture products); and as a preservative factor (together with other emerging
and/or traditional preservative factors) to obtain high moisture novel foods by
hurdle techniques.
There are two main categories of foods with reduced aw whose stability is based
on a combination of factors, i.e. intermediate moisture-available (IM) and high
moisture-available (HM) foods. IM foods generally range from 0.60 to 0.90 aw and
10–50 % moisture content (Alzamora et al. 1995). Additional preservation factors
provide a margin of safety against resistance to aw spoilage microorganisms
(mainly molds and yeasts, and some bacterial species). Lowering aw is often
combined with chemical preservatives (i.e., sorbates, sulphites, benzoates, and
antimicrobials of natural origin), a reduction of pH, and sometimes with compet-
itive microorganisms. Other IM products receive thermal treatment during
manufacturing process that inactivates heat-sensitive microorganisms, while the
subsequent hot filling in sealed containers further improves microbial stability
(Leistner and Gould 2002). Most IM foods are designed in such way that they
can be stored at ambient temperature for several months (Karel 1973, 1976). Many
IM products, due to the addition of very high amounts of solutes (such as sugar or
salt) to reduce aw to the desired level, are too sweet or too salty, becoming
undesirable from a nutritional and sensory point of view. Therefore, this category
of products has been subjected in the last decade to continuous revision and
discussion.
High-moisture foods have aw values above 0.90; in this category, the reduction
of aw is a microbial stress factor of less relative importance because most micro-
organisms are able to proliferate (Leistner and Gould 2002). Stability at ambient
temperature can be reached by applying a designed hurdle technology approach.
HM fresh-like fruits and cooked meat products, preserved by the interaction of
aw—mild heat treatment—pH—preservatives, and that can be stored without
refrigeration, represent a rational application of the combined approach (Alzamora
et al. 1995, 2000a, b; Leistner and Gould 2002). Most shelf-stable foods do not rely
solely on aw for microbial control, but on other preservation factors. The combi-
nation of aw and pH acts as a relevant factor in many of these products by
preventing proliferation of pathogenic microorganisms, while other approaches
(antimicrobials, thermal treatment, etc.) play a secondary role, mainly against
spoilage flora (Tapia et al. 1994). Different approaches have been explored for
obtaining shelf-stability and fresh-like quality in fruit products. Commercial, min-
imally processed fruits are fresh (with high moisture), and are prepared for conve-
nient consumption and distribution to the consumer in a fresh-like state. Minimum
processing includes minimum preparation procedures such as washing, peeling
and/or cutting, and packing, after which the fruit product is usually placed in
refrigerated storage where its stability varies depending on the type of product,
processing, and storage conditions. If fresh-like fruit is the goal, dehydration should
Water Activity and Microorganism Control: Past and Future 253
It is important to note that low aw inhibits growth, but does not kill bacteria (or does
so very slowly). On the contrary, nonsporing bacteria that may contaminate food
before or after processing steps seem to be protected by the low aw and survive with
little or no reduction in the amount of cfu for long periods of time (Lund
et al. 2000). Thus, aw reduction is a very effective method to eliminate the risk of
toxigenic bacteria (i.e., bacteria that grow in food and produce toxins) such as
Staphylococcus aureus, Clostridium botulinum, and Bacillus cereus, but it does not
eliminate the risk posed by infectious bacteria, i.e. Salmonella and some E. coli
strains
The effects of temperature on survival of microorganisms have been widely
documented, with the heat resistance of vegetative cells and spores as influenced by
aw being probably the most broadly studied area in terms of microbial inactivation
(Gould 2000; Leistner 2000). In general, vegetative cells and spores of fungi are
more resistant as aw of the heating menstrum is reduced. The type of solute used to
adjust aw to the same value may result in significant differences in the heat
resistance of a given microorganism. Small molecular weight compounds such as
sodium chloride and glycerol were shown to decrease the resistance of heat resistant
strains of Salmonella, whereas the heat resistance of heat-sensitive strains was
increased. Larger molecular weight solutes, such as sucrose, exerted a more pro-
tective effect against heat inactivation.
Commercial and legal requirements regarding the safety of food products have
focused attention on the development and improvement of decontamination
methods. Due to the presence of a specific microflora adapted to low water content,
the decontamination of dried powders, such as milk powder, flour, herbs and spices,
egg products, algal powders, and carrrageenans, is difficult. The heat resistance of
dried vegetative cells is many times higher than the resistance of the same micro-
organisms in an aqueous solution, but much more research is needed to understand
the relation aw-microorganism thermotolerance. Fine and Gervais (2005) devel-
oped a process for heat treatment of seeds and food powders based on very short
254 A. L
opez-Malo and S.M. Alzamora
Fig. 3 Decimal reduction times (D) for Zygosaccharomyces bailii inactivation when subjected to
thermal (TT) or thermo-ultrasonic (TTU) treatments at selected aw values
order inactivation kinetics, and D values were calculated. D values were lower for
aw 0.99 than for 0.97 or 0.95, being significantly ( p < 0.05) lower for thermo-
ultrasonic (TUT) than for thermal treatments (TT). Z. bailii D values at 45 C were
reduced from 15.4 (TT) to 7.4 (TTU) min (aw 0.99), from 26.8 (TT) to 8.6 (TTU)
min (aw 0.97), and from 43.5 (TT) to 12.9 (TTU) min (aw 0.95). TTU significantly
( p < 0.05) reduced the heat-protective effect of reduced aw, demonstrating that a
combination of heat and ultrasound can be applied to inactivate yeast cells, reduc-
ing heat treatment intensity, especially in aw reduced media (Fig. 3).
Palou et al. (1998) evaluated the effect of aw of the suspension media (0.98 or
0.95) on the survival of Zygosaccharomyces bailii subjected to high hydrostatic
pressure treatments (Fig. 4). Their findings demonstrate that aw has an important
influence on yeast survival after high pressure treatments, becoming more resistant
as aw decreases.
Fig. 4 Survival curves of Zygosaccharomyces bailii suspended in media formulated with selected
aw values and exposed to high-pressure treatments. (Adapted from Palou et al. 1998)
and volume of fluids) is vital for survival and growth. Cell homeostasis is
maintained through reaction mechanisms that act in response to relative minor
changes in physiological variables, leading to a series of events that will in turn
restore the altered variable to its original value (Gould 1996). These homeostatic
mechanisms act to ensure that microbial cell physiological activities remain rela-
tively unchanged, even when the environment around the cell is different and
greatly perturbed (Leistner and Gould 2002). In the case of vegetative microorgan-
isms, the homeostatic mechanisms are energy-dependent because the cell must
expend energy to resist the stress factors, e.g., to repair damaged components and
to synthesize new cell components. In the case of spores, the homeostatic mecha-
nisms is not energy-consuming but they are built in prior to being exposed to an
environmental stress (Leistner and Gould 2002). Preservation factors must over-
come microbial homeostatic resistance in order to be effective.
Regarding aw status, all microorganisms must deal with changes in water
availability in their surroundings, since the concentration of solutes within the
cell is higher than that in the environment (Kempf and Bremer 1998). Supporting
this difference on opposite sides of the semipermeable cytoplasmic membrane is
essential, since positive turgor is considered as the driving force for cell expansion.
As a result of changes in the osmolality of the medium (decreasing aw), passive flow
of water across microbial cell membranes occurs, resulting in a loss of turgor.
Microorganisms do not possess active transport mechanisms for water; hence,
turgor is adjusted by controlling osmotically active solutes in the cytoplasm
Water Activity and Microorganism Control: Past and Future 257
(Kempf and Bremer 1998). The general mechanism managed by microbial cells
that are able to overcome the initial loss following a hyperosmotic shock is the
accumulation of cytoplasmatic solute(s) so as to increase the internal osmotic
pressure, which can restore turgor (Gould 1989a; Gutierrez et al. 1995). Osmotic
regulation involves the expression of a number of genes to optimize growth under
the stress condition, allowing cells to modulate the rate of acquisition of compatible
solutes. A great variety of solutes can be accumulated by various microorganisms
(i.e. cations, amino and imino acids, polyols, and sugars). These solutes allow
continued activity of cytoplasmic enzymes at lower aw values; are soluble to high
concentrations; and the cell membrane exhibits controlled permeability to them.
Compatible solutes may be transported from the medium or synthesized de novo in
the cytoplasm. The accumulation process is energy dependent and causes an
increase in maintaining metabolism, with consequent reduction in the microbial
growth rate. Control of microbial survival or growth in foods by modification of aw
is attained when the target microorganism is unable to carry out these effects; in
other words, the mechanisms exceed the osmoregulatory capacity of the cell. In
terms of microbial control, different strategies can be employed: using aw as the
only microbial stress factor and, therefore, aw must be reduced to very low levels; or
using other stress factors in combination with lowering of aw (“hurdle concept” or
combined methods), thus reducing the amount of energy available for
osmoregulation.
Osmotic induction of general stress systems results in the consequent develop-
ment of multitolerances towards other environmental stresses when subjected to
hypertonic environments (Pichereau et al. 2000; O’Byrne and Booth 2002).
New food preservation strategies can be developed on a sound scientific base if
combination of hurdles is considered, taking into account their different modes of
action over microbial cells and the mechanisms mediating microbial adaptive
responses (Ross et al. 2003; Gould 1996). Our knowledge of these subjects is,
however, far for complete. Nowadays, the advent of new methodic developments
may provide a significant conceptual advance in the understanding of responses in
microorganisms to a variety of environmental stresses (Brul et al. 2006). Many
advances in recent years have been made to understand the behavior of stressed
microorganisms and food preservation. McMahon et al. (2007) found that sublethal
stresses (high/low temperature, osmotic and pH stress) which food pathogens
encounter in modern food preservation systems increase the inter- and intra-specific
horizontal transmission of selected antibiotic resistance (ABR) plasmids. So,
increased use of bacteriostatic (sublethal) rather than bactericidal (lethal) food
preservation systems may contribute to the dissemination of ABR among important
food pathogens.
Other important concept of enormous importance for the food industry has to do
with heterogeneity within bacterial populations (Booth 2002). In traditional analy-
sis, the homogeneity of the population of cells in terms of their biochemistry and
responsiveness has been assumed. But homogeneity is neither required nor desir-
able for survival; moreover, heterogeneity is an intrinsic factor in organisms which
assures that, when a stress is imposed, some cells will survive and become the new
258 A. L
opez-Malo and S.M. Alzamora
colonists of the newly encountered environment. In this sense, recent studies have
pointed out the variability in the growth limits of individual microbial cells as aw is
reduced. Assuming that each colony was derived from a single cell, and considering
the number of cells of Salmonella enteritidis inoculated, Koutsoumanis (2008)
determined that as NaCl concentration increased and pH decreased, the number
of cells able to grow decreased. As the inoculums size increased, NaCl growth
limits of S. enteritidis increased and became less variable. The author noted the
need for stochastic approaches in quantitative microbiology, especially at environ-
ments close to the boundary of growth. As contamination of foods with pathogens
occurs at very low levels, it is clear that the behavior of single cells should be taken
into account through stochastic models.
A new scientific approach to study the survival strategies of microorganisms of
concern in various foods by using “genomics” technology (genomics,
transcriptomics, proteomics, and metabolomics) is being explored by various
groups (Brul et al. 2006, 2008). Integration of molecular biology and classical
physiology (the so-called “functional genomics”) could lead to a molecular mech-
anistic basis of the survival strategies of pathogens of concern. In this way,
preservation factors could be optimally combined to obtain synergistic or additive
effects on the inhibition of unwanted microorganisms.
Even though there exist restrictions and sensory problems that may cause the
reduction of aw in some food products, the employment of this preservation factor
continues to be useful in process design, food formulation, food quality, and in the
selection of storage conditions. Microbiological relation to water activity has an
important impact and practical implications in the food industry. It is clear that aw
plays a significant role in the survival and death of microorganisms in foods and, in
combination with other traditional preservation factors, aw is being used nowadays
as the basis for chemical and microbial stability in a wide range of products.
Food safety and stability have been controlled by the food processing industry using
a number of methods. Vegetative microbial cells and spores are destroyed or restricted
by the individual or combined use of the following traditional preservation methods:
heat sterilization, pasteurization, refrigeration, freezing, water activity control,
cooking, acidification, fermentation, vacuum packaging, and/or adding antimicrobial
agents. The extensive use of these procedures has maintained the low-risk status of the
food supply. However, the new generation of minimally processed, fresher, antimi-
crobial, and additive-free food products has less resistance to spoilage and allows
greater risks of food-borne illness if any abuse occurs. Traditional preservation pro-
cedures generally base their intensity on microbial control, so after destroying, elim-
inating, or substantially reducing initial microbial counts, measures to prevent
recontamination should be taken, which often results in a safe and stable product
(Baird-Parker 1995). If microbial hazard or spoilage cannot be totally eliminated from
Water Activity and Microorganism Control: Past and Future 259
the food, microbial growth and toxin production must be inhibited. Microbial growth
can be inhibited by combining intrinsic food characteristics with extrinsic product
storage and packaging conditions. If confident judgment about food quality, stability,
and safety is to be made, it is essential to rely on a better understanding of survival to aw
stressor, accurate quantitative data and mechanistic predictive models about combined
factors affecting growth or survival of unwanted microorganisms are needed
(Gould 1989b; Buchanan 1992; Alzamora and Lopez-Malo 2002).
References
Allende A, Tomás-Barberán FA, Gil MI (2006) Minimal processing for healthy traditional foods.
Trends Food Sci Technol 17:513–519
Alzamora SM, Cerrutti P, Guerrero S, Lopez Malo A (1995) Minimally processed fruits by
combined methods. In: Welti Chanes J, Barbosa Cánovas GV (eds) Food preservation by
moisture control fundamentals and applications. Technomic Pub Co, Lancaster, pp 463–492
Alzamora SM, Fito P, Lopez-Malo A, Tapia MS, Parada-Arias E (2000a) Minimally processed
fruit using vacuum impregnation, natural antimicrobial addition and/or high hydrostatic
pressure techniques. In: Alzamora SM, Tapia MS, Lopez Malo A (eds) Minimally processed
fruits and vegetables Fundamental aspects and applications. Aspen Publishers, Inc., Gaithers-
burg, pp 293–315
Alzamora SM, Gerschenson LN, Cerrutti P, Rojas AM (1989) Shelf stable pineapple for long term
non refrigerated storage. Lebensm Wiss Technol 22:233–236
Alzamora SM, L opez-Malo A (2002) Microbial behavior modeling as a tool in the design and
control of processed foods. In: Welti-Chanes J, Barbosa-Canovas GV, Aguilera JM (eds)
Engineering and food for the 21st century. CRC, Boca Raton, pp 631–650
Alzamora SM, L opez-Malo A, Tapia MS (2000b) Minimally processed fruits and vegetables:
fundamentals and applications. Aspen Publishers, Inc., Gaithersburg
Alzamora SM, Tapia MS, Argaiz A, Welti J (1993) Application of combined methods technology
in minimally processed fruits. Food Res Int 26(125):130
Alzamora SM, Tapia MS, Lopez-Malo A, Welti-Chanes J (2003) The control of water activity. In:
Zeuthen P, Bogh-Sorensen L (eds) Food preservation techniques. CRC, Cambridge, England,
pp 126–143
Baird-Parker AC (1995) Development of industrial procedures to ensure the microbiological
safety of food. Food Control 6:29–36
Beuchat LR (1983) Influence of water activity on growth, metabolic activities and survival of
yeasts and molds. J Food Prot 46:135–140
Beuchat LR (1987) Influence of water activity on sporulation, germination, outgrowth, and toxin
production. In: Rockland LB, Beuchat LR (eds) Water activity: theory and applications to food.
Marcel Dekker, Inc., New York
Beuchat LR (1996) Detection and enumeration of microorganisms in hurdle technology foods,
with particular consideration of foods with reduced water activity. In: Barbosa-Cánovas GV,
Welti-Chanes J (eds) Food preservation by moisture control. Fundamentals and applications.
Technomic Publishing Company, Lancaster
Booth IR (2002) Stress and single cell: Intrapopulation diversity is a mechanism to ensure survival
upon exposure to stress. Int J Food Microbiol 78:19–30
Breeuwer P, Lardeau A, Peterz M, Joosten HM (2003) Desiccation and heat tolerance of
Enterobacter sakazakii. J Appl Microbiol 95:967–973
Brewer MS (1999) Traditional preservatives. Sodium chloride. In: Robinson RK, Batt C, Patel P
(eds) Encyclopedia of food microbiology-2. Academic, London, pp 1723–1728
260 A. L
opez-Malo and S.M. Alzamora
Brul S, Mensonides FIC, Hellingwerf KJ, Teixera de Matos MJ (2008) Microbial system biology:
new frontiers open to predictive microbiology. Int J Food Microbiol 128:16–21
Brul S, Schuren F, Montijn R, Keijser BJF, van der Spek H, Oomes SJCM (2006) The impact of
functional genomics on microbiological food quality and safety. Int J Food Microbiol
112:195–199
Buchanan RL (1992) Predictive microbiology: mathematical modeling of microbial growth in
foods. In: Finley JW, Robinson SF, Armstrong DJ (eds) Food safety assessment, ACS series.
American Chemical Society, Washington, DC, pp 250–260
Buchanan RL (2003) Resistance-thermal and other DHHS. Food and Drug Administration. Center for
Food Safety and Applied Nutrition. www.fda.gov/ohrms/dockets/ac/03/slides/3939s1_buchanan_
Capozzi V, Fiocco D, Amodio ML, Gallone A, Spano G (2009) Bacterial stressors in minimally
processed food. Int J Mol Sci 10:3076–3105
Cerrutti P, Alzamora SM, Vidales SL (1997) Vanillin as antimicrobial for producing shelf stable
strawberry purée. J Food Sci 62:608–610
Chirife J (1994) Specific solute effects with special reference to Staphylococcus aureus. In: Fito P,
Mulet A, McKenna B (eds) Water in foods. Fundamental aspects and their significance in
relation to processing of foods. Elsevier Applied Science, Great Britain, pp 409–419
Chirife J (1995) An update on water activity measurements and prediction in intermediate and high
moisture foods: the role of some non-equilibrium situations. In: Barbosa-Cánovas G, Welti-
Chanes J (eds) Food preservation by moisture control. Fundamentals and Applications.
Technomic Publishing Company, Lancaster, pp 169–189
Christian JHB (2000) Drying and reduction of water activity. In: Lund BM, Baird-Parker TC,
Gould GW (eds) The microbiological safety and quality of foods. Aspen Pub., Marylandz,
pp 146–174
Corry JEL (1973) The water relations and heat resistance of microorganisms. Prog Ind Microbiol
12:73–80
Delgado-Portales RE, Palou E, Lopez-Malo A (2007) Probabilistic modeling of Enterobacter
sakazakii survival to desiccation after grown at selected water activities. International associ-
ation for food protection annual meeting
Edelson-Mammel S, Porteous MK, Buchanan RL (2006) Acid resistance of twelve strains of
Enterobacter sakazakii, and the impact of habituating the cells to an acidic environment.
J Food Sci 71:201–207
Fine F, Gervais P (2005) A new high temperature short time process for microbial decontamina-
tion of seeds and food powers. Powder Technol 157:108–113
Gould GW (1989a) Drying, raised osmotic pressure and low water activity. In: Gould GW
(ed) Mechanisms of action of food preservation procedures. Elsevier Applied Science,
New York
Gould GW (1989b) Predictive mathematical modelling of microbial growth and survival in foods.
Food Sci Technol Today 3(2):89–92
Gould GW (1996) Methods for preservation and extension of shelf life. Int Food Microbiol
33:51–64
Gould GW (2000) Induced tolerance of microorganisms to stress factors. In: Alzamora SM, Tapia
MS, Lopez-Malo A (eds) Minimally processed fruits and vegetables. Fundamental aspects and
applications. Aspen Publishers Inc, Gaithersbug, pp 29–37
Gruzdev N, Pinto R, Sela Saldinger S (2012) Persistence of Salmonella enterica during dehydra-
tion and subsequent cold storage. Food Microbiol 32:415–422
Guerrero S, Alzamora SM, Gerschenson LN (1994) Development of a shelf stable banana purée by
combined factors: microbial stability. J Food Prot 57:902–907
Gutierrez C, Abee T, Booth IR (1995) Physiology of the osmotic stress response in microorgan-
isms. Int J Food Microbiol 28:233
Hocking AD, Christian JHB (1996) Microbial ecology interactions in the processing of foods. In:
Barbosa-Cánovas G, Welti-Chanes J (eds) Food preservation by moisture control. Fundamen-
tals and applications. Technomic Publishing Company, Lancaster
Water Activity and Microorganism Control: Past and Future 261
Hocking AD, Pitt JI (1987) Media and methods for enumeration of microorganisms with
consideration of water activity. In: Rockland LB, Beuchat LR (eds) Water activity: theory
and applications to food. Marcel Dekker, Inc., New York
IFT (2003) Analysis of microbial hazards related to time/temperature control of foods for safety.
Compr Rev Food Sci Food Safety 2:33–41
Jay JM, Loessner MJ, Golden DA (2005) Modern food microbiology. Springer Science + Business
Media, Inc., New York
Karel M (1973) Recent research and development in the field of low moisture and intermediate
moisture foods. CRC Crit Rev Food Technol 3:329–373
Karel M (1976) Technology and application of new intermediate moisture foods. In: Davies R,
Birch GG, Parker KJ (eds) Intermediate moisture foods. Applied Science Publishers Ltd,
London, pp 4–31
Kempf B, Bremer E (1998) Uptake and synthesis of compatible solutes as microbial stress
responses to high-osmolality environments. Arch Microbiol 170:319–330
Koutsoumanis K (2008) A study on the variability in the growth limits of individual cells and its
effect on the behavior of microbial populations. Int J Food Microbiol 128:116–121
Leistner L (2000) Hurdle technology in the design of minimally processed foods. In: Alzamora
SM, Tapia MS, Lopez-Malo A (eds) Minimally processed fruits and vegetables. Fundamental
Aspects and Applications. Aspen Publishers Inc., Gaithersbug, pp 13–27
Leistner L, Gould GW (2002) Hurdle technologies. Combination treatments for food stability,
safety and quality. Kluwer Academic, New York
Lenovich LM (1987) Survival and death of microorganisms as influenced by water activity. In:
Rockland LB, Beuchat L (eds) Water activity: theory and applications to food. Marcel Dekker,
Inc., New York, pp 119–136
Leyer GJ, Johnson EA (1993) Acid adaptation induces cross-protection against environmental
stresses in Salmonella typhimurium. Appl Environ Microbiol 59:1842–1847
Lopez-Malo A, Palou E, Franco-Corzo A (2001) Yeast inactivation kinetics during thermo-
ultrasonication treatments at selected water activities and pHs. International association for
food protection annual meeting
Lund BM, Baird-Parker TC, Gould GW (2000) The microbial safety and quality of food. Aspen
Publishers, Inc., Gaithersburg, pp 387–388
McMahon MAS, Blair IS, Moore JE, Mc Dowell DA (2007) The rate of horizontal transmission of
antibiotic resistance plasmids is increased in food preservation-stressed bacteria. J Appl
Microbiol 103:1883–1888
Mossel DAA (1975) Water and microorganisms in foods—a synthesis. In: Duckworth RB
(ed) Water relations in foods. Academic, New York
O’Byrne C, Booth IR (2002) Osmoregulation and its importance to food—borne microorganisms.
Int J Food Microbiol 74:203–216
Palou E, L opez-Malo A (2001) Solute type effects on Debaryomyces hansenii growth response in
selected water activity and pH media. Institute of Food Technologists (IFT) annual meeting
Palou E, Lopez-Malo A, Barbosa-Canovas GV, Welti-Chanes J (2000) High hydrostatic pressure
and minimal processing. In: Alzamora SM, Tapia MS, L opez-Malo A (eds) Minimally
processed fruits and vegetables. Fundamental aspects. Aspen Publishers, Gaithersburg,
pp 205–222
Palou E, Lopez-Malo A, Barbosa-Canovas GV, Welti-Chanes J, Davidson PM, Swanson BG
(1998) High hydrostatic pressure come-up time and yeast viability. J Food Prot 61:1657–1660
Palou E, Lopez-Malo A, Welti-Chanes J (2002) Innovative fruit preservation methods using high
pressure. In: Welti-Chanes J, Barbosa-Canovas GV, Aguilera JM (eds) Engineering and food
for the 21st century. CRC, Boca Raton, pp 715–725
Pichereau V, Hartke A, Auffray I (2000) Starvation and osmotic stress induced multiresistances
influence of extracellular compounds. Int J Food Microbiol 55:19–25
Raso J, Barbosa-Cánovas GV (2003) Nonthermal preservation of foods using combined
processing technique. Crit Rev Food Sci Nutr 43:265–285
262 A. L
opez-Malo and S.M. Alzamora
Ross AIV, Griffiths MW, Mittal GS, Deeth HC (2003) Combining nonthermal technologies to
control foodborne microorganisms. Int J Food Microbiol 89:125–138
Scott WJ (1957) Water relations of food spoilage microorganisms. Adv Food Res 6:549–556
Tapia MS, Aguilera JM, Chirife J, Parada E, Welti J (1994) Identification of microbial stability
factors in traditional foods from Iberoamerica. Rev Esp Cienc Tecnol Alim 34:145–163
Tapia MS, Welti-Chanes J (2012) Hurdle technology principles applied in decontamination of
whole and fresh-cut produce. In: Gomez Lopez VM (ed) Decontamination of fresh and
minimally processed produce, 1st edn. Wiley, New York, pp 417–449
Tiganitas A, Zeaki N, Gounadaki AS, Drosinos EH, Skandamis PN (2009) Study of the effect of
lethal and sublethal pH and aw stresses on the inactivation or growth of Listeria
monocytogenes and Salmonella Typhimurium. Int J Food Microbiol 134:104–112
On Modeling the Effect of Water Activity
on Microbial Growth and Mortality Kinetics
Abbreviations
aw Water activity
k Growth rate parameter
λ Lag time
ODE Ordinary differential equation
P Probability
Pd Momentary probability rate of cell division
Pm Momentary probability rate of cell mortality
t Time
tc Location of the inflection point
Y(t) Linear or logarithmic growth ratio
Yasymp Asymptotic linear or logarithmic growth ratio
μmax Maximum specific growth rate - the slope of the growth curve at tc
1 Introduction
The effect of moisture on the growth of different kinds of bacteria, yeasts, and molds
in or on foods has traditionally been depicted in the form of a rate vs. water activity
(aw) plot such as that shown in Fig. 1. The rate curve of each organism type begins
(or ends) at a point along the aw-axis. At lower aw levels, presumably, the particular
type of organism does not grow, i.e., its growth rate is zero. While such plots can and
have been successfully used to identify safe water activities for food storage, for
example, they do not always tell the whole story. One reason is that the borderline
between growth and no-growth is rarely sharp. Another reason is that a plot of this
kind does not fully reveal what happens above the marked water activity threshold.
Moreover, in many foods, water activity control is accompanied by a drying process,
in which case the temperature history may affect the type or number of organisms
remaining in the food, and/or the introduction of a complementary agent such as salt
or chemical preservative. To deal with these aspects of microbial growth inhibition, it
is necessary to develop a kinetic model of growth whose parameters account for the
role of all pertinent factors that affect a food’s biological stability and that of their
interactions. The same can be said about chemical and physical changes that inevi-
tably occur during food preservation, especially when heat or drying is involved.
Such changes can consist of two kinds: degradation, notably the loss of freshness and
nutrients, or synthesis, notably the accumulation of Maillard reaction products, but
also other compounds that impart off flavor. Water activity, and factors such as pH
and salt concentration, affect the heat resistance not only of microorganisms, includ-
ing bacterial spores, but also of enzymes. The subject has been amply studied and
there is a large body of literature covering its various aspects.
The objective of this chapter is not to summarize the results of the numerous
studies; this task is better left to professional microbiologists who have much better
insight into the biology and biochemistry of microbial growth; they also possess the
tools to evaluate what has been published in the literature and to assess their
practical implications in food preservation. The main goal here is to discuss ways
to introduce the role of water activity into existing kinetic rate models of microbial
growth and inactivation, and to present a new class of probabilistic models that can
account for simultaneous growth and mortality that sometimes ensue.
A typical microbial growth curve has four regions representing the lag phase,
exponential growth phase, stationary phase, and mortality phase, shown schemat-
ically in Fig. 2. Most of the studies of microbial growth kinetics have focused on the
sigmoid part of the curve, i.e., the first three phases. This is a reasonable approach
because most foods become inedible or unsafe to eat long before the mortality
phase sets in. The Gompertz model and several variants of the logistic (Verhulst)
equation are the most commonly used growth models in food microbiology
(McMeekin et al. 1993; McKellar and Lu 2004; van Boekel 2009). All of these
models deal with net growth and therefore do not account for cell mortality, which
at least theoretically can occur not only during the stationary phases but also during
the lag and exponential phases. A notable exception is the quasi-chemical model
developed at the US Army Natick R&D laboratory, which takes into account the
release of polluting metabolites by the cells, and hence for simultaneous division
and mortality at any phase (Taub et al. 2003). Since the conventional growth
models do not account for mortality, none of them can be considered as being
fundamental in the strict sense of the term, and any mechanistic proposed interpre-
tation of their parameters must be based on independent experimental evidence. In
other words, the shape of the growth curve alone does not contain enough infor-
mation to account for the continuously changing roles of cell physical growth,
division, and mortality during a microbial population’s growth cycle.
The sigmoid part of a more or less symmetric isothermal growth curve should be
characterized by no less than three parameters, as demonstrated in Fig. 3, in order to
separately account for the asymptotic growth level, the location of the inflection
point of the growth curve, and the slope of the curve at this point. At least in
principle, these three characteristics need not vary in a predetermined manner as a
result of changes in food temperature, water activity, pH, salt concentration, oxygen
tension, etc. Much of the published research on microbial growth has focused on the
effect of temperature, water activity and pH on the maximum specific growth rate
(μmax), i.e., the growth curve’s slope at its inflection point. However, to extrapolate
accelerated storage data to ambient conditions, it is necessary to determine the
effect of these factors not only on the rate (Fig. 4) but also on all three parameters as
shown schematically in Figs. 5, 6, and 7, in order to produce the whole growth
curve, as described below. Moreover, in order to predict a dynamic growth pattern,
Fig. 3 Schematic view of a growth curve described by the shifted logistic model (Eq. 1) and its
parameters
On Modeling the Effect of Water Activity on Microbial Growth and Mortality. . . 267
Fig. 5 Schematic view of the effect of temperature and water activity combinations on asymptotic
growth level (Yasymp)
268 M. Peleg et al.
Fig. 6 Schematic view of the effect of temperature and water activity combinations on maximum
growth rate (μmax or kYasymp/4 in terms of the shifted logistic model). Left—when water activity
does not affect the organism’s heat resistance; right—when dryness increases heat resistance
Fig. 7 Schematic view of the effect of temperature and water activity combinations on inflection
point location (tc). Similar relationship will emerge for the “lag time” (λ )
i.e., when some or all of the factors that affect the growth vary simultaneously, each
of the three parameters’ dependency on the pertinent factors should be determined
experimentally. The hypothetical temperature and aw dependencies of the three
growth parameters shown in Figs. 5, 6, and 7 give a clue as to how the three
On Modeling the Effect of Water Activity on Microbial Growth and Mortality. . . 269
Y
Y ðt Þ ¼ asymp
1 þ exp k T aw t tc T aw t t
Y
asymp
ð1Þ
1 þ exp k T aw t tc T aw t
This equation describes the relationship between the linear or logarithmic growth
ratio Y(t), Y(t) ¼ [N(t) N0]/N0 or log[N(t)/N0] vs. time, and contains the already
mentioned three parameters—Yasymp, the asymptotic linear or logarithmic growth
ratio, tc, the growth curve inflection point location, and k, a representative of the
growth rate—which are all functions of the particular temperature–water activity
conditions under which the growth curve has been determined. The maximum
specific growth rate (μmax) according to this model is kYasymp/4.
In principle, and in the case of temperature alone in practice (Corradini and
Peleg 2006a), this model could be used to extrapolate isothermal accelerated
270 M. Peleg et al.
Fig. 8 Schematic view of the extrapolation of accelerated storage data of microbial growth using
the shifted logistic equation as a model
where
1
t* ðtÞ ¼ loge
k T aw t
"
#
exp k T aw t tc T aw t Y asymp T aw t þ Y ðtÞ 1 þ exp k T aw t tc T aw t
Y asymp T aw t exp k T aw t tc T aw t Y ðtÞ 1 þ exp k T aw t tc T aw t
ð3Þ
When a chemical reaction or microbial inactivation follows first or other fixed order
kinetics, determining the exponential rate constant, or other characteristic rate
constant, as a function of temperature, aw and/or other factor(s) will be theoretically
sufficient to construct the progress curve of the process under dynamic conditions.
This has been the basis of the conventional method for estimating changes in
processed, stored, or shipped foods, by extrapolating accelerated storage data
(Fig. 4) and constructing time–temperature integrators for in situ measurements.
In all these applications, it has been assumed, almost universally, that the temper-
ature dependence of the characteristic rate constant of the process follows the
Arrhenius equation, and in some cases the log-linear model, and its hallmark, the
z value. This traditional approach has serious limitations when it comes to nonlinear
biological processes and chemical and biochemical reactions, especially when they
272 M. Peleg et al.
where Y(t) here can be the survival ratio of an organism, i.e., N(t)/N0, or of a
chemical component like a vitamin, i.e., C(t)/C0 (Corradini and Peleg 2004a), and
b[T(aw(t))] and n[T(aw(t))] are temperature and water activity-dependent coeffi-
cients. Notice that first-order kinetics is just a special case of the Weibullian model
(Eq. 4) where n[T(aw(t))] ¼ 1.
In principle, the Weibullian model can be used to extrapolate accelerated storage
data and generate entire survival or degradation curves for stationary ambient
temperature–water activity conditions. This requires the extrapolation of the
model’s two parameters, as shown schematically in Fig. 9. The same applies to
alternative models of nonlinear decay or synthesis kinetics. They may have two or
more parameters, all of which need to be extrapolated in order to produce the entire
curve. The potential of this approach has been demonstrated with vitamin C
degradation during frozen storage, albeit with the temperature alone being the
accelerating factor (Corradini and Peleg 2006b).
If the momentary logarithmic inactivation or degradation rate is indeed the
logarithmic rate at the momentary temperature–water activity conditions at the
time that corresponds to the system’s momentary state, Y(t), the Weibullian model
can be converted into the dynamic rate equation:
n½Tðaw ðtÞÞ1
dlogY ðtÞ logY ðtÞ n½Tðaw ðtÞÞ
¼ b½T ðaw ðtÞÞn½T ðaw ðtÞÞ ð5Þ
dt b½T ðaw ðtÞÞ
The rate model (Eq. 5) is an ODE and can be easily solved numerically by modern
mathematical software for almost any conceivable temperature–water activity
history. But here, too, the main hurdle in the model implementation is not the
mathematical solution, but the amount of experimental work needed to determine
the temperature–water activity of the coefficients and/or the effect of other factors’
dependence. Like the growth model described in the previous section, the predic-
tive ability of Eq. (5) has been demonstrated with experimental data, albeit with
only temperature variability (e.g., Corradini and Peleg 2004b; Pardey et al. 2005).
On Modeling the Effect of Water Activity on Microbial Growth and Mortality. . . 273
c(t)/c0
Time (arbitrary scale) Time (arbitrary scale)
b(T)
n(T)
extrapolated
extrapolated
Temperature Temperature
Fig. 9 Schematic view of the extrapolation of accelerated storage data of microbial inactivation
or chemical degradation using the Weibullian model
4 Probabilistic Models
All kinetic models used to describe microbial growth or mortality are continuous
and deterministic, as they should be, and therefore only apply to large populations
of cell or spores. Yet, especially when it comes to pathogens, the number of cells
that affect the safety of a food item can sometimes be small, and hence the
frequently sporadic and haphazard incidence of food poisoning episodes. Similar
patterns can also be observed in microbial spoilage of food products. There are
cases where some, but not all, units of the same shipment spoil early in a pattern that
cannot be attributed to faulty packaging, different storage conditions, or
recontamination. Scenarios involving only a few microbial cells can be produced
274 M. Peleg et al.
DOMINANT GROWTH
t=0
t = Dt
t = 2 Dt
t = Dt
t = 2 Dt
DOMINANT MORTALITY
t=0
t = Dt
t = 2 Dt
Viable/Growing/Dividing Cell
Division Daughter Cells
Dead Cell
Fig. 10 Schematic view of the probabilistic model of cell division and mortality (first two
iterations). Notice that the probabilities of division and mortality change with time. For mathe-
matical formulation of the model, see Horowitz et al. (2010)
On Modeling the Effect of Water Activity on Microbial Growth and Mortality. . . 275
0.3
Pd (t)
Probability Rate Functions
Pm(t)
0.1
0
0 10 20 30 40 50 0 10 20 30 40 50
Time (arbitrary scale) Time (arbitrary scale)
Fig. 11 Generation of a sigmoid growth curve with the discrete version of the probabilistic model.
Left—division and mortality probability rate functions; right—total number of cells
0.3
Pm(t)
Probability Rate Functions
0.2
Pd (t)
0.1
0
0 10 20 30 40 50 0 10 20 30 40 50
Time (arbitrary scale) Time (arbitrary scale)
Fig. 12 Generation of growth followed by mortality curve with the discrete version of the
probabilistic model. Left—division and mortality probability rate functions; right—total number
of cells
against time the growth curve, N(t) vs. t. It can be shown that as the initial number
of followed cells increases, the growth curve becomes smoother and more determin-
istic (Horowitz et al. 2010). According to this model, a sigmoid growth curve is
produced when the probability rate functions, Pd(t) and Pm(t), converge (Fig. 11).
When Pd(t) is surpassed by Pm(t) (Fig. 12), the result is growth followed by mortality.
Or, conversely, if Pd(t) surpasses Pm(t) (Fig. 13), the result is inactivation followed by
renewed growth. According to this model, pure inactivation occurs when Pd(t) ¼
0 and pure growth when Pm(t) ¼ 0. In other words, almost all of the commonly
276 M. Peleg et al.
0.2
Pd (t)
Probability Rate Functions
0.15
0.1
0.05
0 0
0 10 20 30 40 50 0 10 20 30 40 50
Time (arbitrary scale) Time (arbitrary scale)
Fig. 13 Generation of inactivation and resumed growth curve with the discrete version of the
probabilistic model. Left—division and mortality probability rate functions; right—total number
of cells
growth and mortality. Microbial growth and mortality rarely follow first order or
any other fixed order kinetics. Therefore, most, if not all, traditional models, which
are based on a single rate constant and its temperature and water activity depen-
dence, are ill-suited for extrapolation of accelerated storage data and predicting
dynamic patterns. A more promising approach would be to determine the
temperature–water activity dependence of all pertinent kinetic parameters and
incorporate it in the growth or inactivation rate equation coefficients. Doing so
may require considerable experimental work, and it will be a challenge to future
research to minimize the effort and cost.
Regardless of whether the growth/mortality model is derived from kinetic or
probabilistic considerations, inclusion of parameters that account for the combined
role of several factors in its mathematical structure will inevitably render it cum-
bersome. Nevertheless, the model, if kinetic, will most likely be an ODE that can be
easily solved numerically by modern mathematical software. The growth/mortality
model itself need not be unique. This has been demonstrated with existing models.
Therefore, a new model’s form could be chosen in light of simplicity and conve-
nience considerations only. Although not generally recognized by users, all current
microbial growth and inactivation models are empirical (or phenomenological),
claims to the contrary concerning some of them not withstanding. Thus, the only
requirements for any new class of models are mathematical consistency and
demonstrated predictive ability; prediction here refers to the ability of a model to
estimate survival and growth data not used in the model’s parameters determina-
tion. In other words, good fit and successful interpolation are not considered
prediction in the strict scientific sense of the term.
One reason that none of the traditional models of microbial growth can be
considered fundamental is that they only deal with the net change in the microbial
population size and hence overlook mortality, which can occur at any growth phase.
While this might not be a handicap when large populations are concerned, it can
become a serious flaw when only a few cells are involved. At least in principle, a
fully probabilistic model based on varying division and mortality probability rates
can remedy this shortcoming of the current kinetic models. Fully probabilistic
models might be particularly suitable to the analysis, and perhaps prediction, of
the fate of individual or small groups of cells, such as those of very virulent but not
widely spread pathogens. A stochastic and discrete model can be converted into a
deterministic and continuous one. In this form it will apply to large populations,
which will make it much easier to determine its parameters and their temperature–
water activity dependence experimentally, using standard laboratory and data
analysis procedures. Once determined, the parameters can be incorporated in the
stochastic version and be used to simulate and perhaps predict spoilage and food
poisoning patterns. Full development of the concept and its applications are still in
the future. Yet, it already suggests ways to resolve outstanding issues in predictive
microbiology. A great challenge to future researchers will be to combine the rate
and stochastic models with heat and mass transfer theories in order to account for
spatial as well as temporal changes in microbial populations in foods. Another
challenge will be to develop predictive models for growth and inactivation on food
278 M. Peleg et al.
surfaces. These models will most probably be very different from those developed
for organisms evenly dispersed in the bulk, because they will need to account for
biofilms formation and its kinetics, as well as for local variations in moisture and
chemical composition along the food’s surface during the microbial growth.
References
Corradini MG, Peleg M (2004a) A model of non-isothermal degradation of nutrients, pigments and
enzymes. J Sci Food Agric 84:217–226
Corradini MG, Peleg M (2004b) Demonstration of the Weibull-Log logistic survival model’s
applicability to non isothermal inactivation of E. coli K12 MG1655. J Food Prot 67:2617–2621
Corradini MG, Peleg M (2006a) Shelf-life estimation from accelerated storage data. Trends Food
Sci Technol 18:37–47
Corradini MG, Peleg M (2006b) Prediction of vitamin loss during non isothermal heat processes &
storage with non linear kinetics models. Trends Food Sci Technol 17:24–34
Horowitz J, Normand MD, Corradini MG, Peleg M (2010) A probabilistic model of growth,
division, and mortality of microbial cells. Appl Environ Microbiol 76:230–242
McKellar RC, Lu X (2004) Primary models. In: McKellar RC, Lu X (eds) Modeling microbial
responses on foods. Boca Raton, CRC
McMeekin TA, Olley JN, Ross T, Ratkowsky DA (1993) Predictive microbiology: theory and
application. Wiley, New York
Pardey KK, Schuchmann HP, Schubert H (2005) Modelling the thermal inactivation of vegetative
microorganisms. Chem Ing Tech 77:841–852
Peleg M (2006) Advanced quantitative microbiology for food and biosystems: models for
predicting growth and inactivation. CRC, Boca Raton
Ross T, Dalgaard P (2004) Secondary models in modeling microbial responses. In: Lu X,
McKellar RC (eds) Food. CRC, Boca Raton, pp 63–150
Taub IA, Feeherry FE, Ross EW, Kustin K, Doona CJ (2003) A quasi-chemical kinetics model for
the growth and death of Staphylococcus aureus in intermediate moisture bread. J Food Sci
68:2530–2537
van Boekel MAJS (2009) Kinetic modeling of reactions in foods. CRC, Boca Raton
Importance of Halophilic and Halotolerant
Lactic Acid Bacteria in Cheeses
Abbreviations
aw Water activity
HALAB Halophilic and alkaliphilic lactic acid bacteria
NSLAB Nonstarter lactic acid bacteria
1 Introduction
Water plays a significant role in cheese texture and in bacterial metabolism and
consequently is very important for the events that occur during cheese ripening. The
influence of water content and water activity (aw) on cheese quality is very
complex, due to changes in chemical composition during the continuous ripening
of the product. Salt, along with pH, water activity, and redox potential, helps to
minimize spoilage and prevents the growth of pathogens and spoilage organisms in
cheese. This dairy product contains different low molecular weight compounds
which are partly produced during ripening or, as in the case of salt, are added during
manufacturing. These low molecular soluble compounds are mainly responsible for
the water activity in cheeses.
Milk enzymes and lactic acid bacteria starter cultures progressively hydrolyze
milk compounds and lower water activity. Such changes are relatively small in
fresh cheeses, but very unique in semihard and hard cheeses. The addition of NaCl
and loss of water during storage provide a supplementary effect to lower the water
activity of the product. For most cheese varieties, the salt to moisture ratio is the
most significant factor to manipulate the water activity.
Moisture content in cheese ranges from 32 % in Parmesan cheese to 79 % in
cottage cheese. Salt content is in the range of 1.45 % in Emmental cheese to
11.41 % in some Roquefort cheeses; and aw varies from 0.995 for fresh unripened
cheeses to 0.91 for Parmesan cheese (Marcos et al. 1981). In most cheese varieties
with a moisture content above 40 %, the aw values depend mostly on the moisture
and NaCl contents. Many techniques for calculating the water activity of cheeses
from their chemical composition have been published. A nomograph for calculation
of aw of soft cheeses from percentages of moisture (% H2O) and salt (% NaC1) was
developed by Esteban and Marcos (1982). The additional aw drop in ripened
cheeses is caused to a great extent by the aqueous concentration of other solutes
such as nonprotein nitrogen (NPN) compounds of low molecular weight released
by proteolysis. Linear regression analysis has been used to predict the water
activity values of brie and Camembert cheeses from ash, NaCl, and nonprotein
nitrogen content and of Bleu de Bresse, Cabrales, Danablu, Edelpilzkäse, Gorgon-
zola, and Roquefort from ash and soluble nitrogen data (Fernández-Salguero
et al. 1986).
Work in the field of modeling the effect of salt concentration on the growth of
different microorganisms or on the distribution and final concentration of NaCl in
cheeses with time is abundant. The effect of pH (4–6), NaCl (1.5–3 %)
corresponding to values of aw (0.987–0.910), and temperature (10–30 C) on the
growth of Mucor racemosus isolated from Camembert cheese can be predicted by
means of primary and secondary modeling (Bekada et al. 2008). This indicates that
predictive modeling of the effect of many growth-limiting factors can be performed
by the use of statistical tools such as response surface methodology, Taguchi
orthogonal designs, and others. Other phenomena such as salt diffusion can be
modeled. Multicomponent approaches to salt and water diffusion in cheese have
been used to build theoretically based models. One such model was developed
using the Maxwell–Stefan equation to predict NaCl increase and moisture decrease
in cheese during brine salting. This type of model is used to calculate changes in the
NaCl and moisture profile, as well as dimensions of the cheese. The best solutions
are obtained when the diffusivities are represented as functions of curd porosity and
salt concentration (Payne and Morrison 1999). The shape of the cheese is also very
important for the evaluation of solute diffusion phenomena in cheeses. Salt will
diffuse differently in a brick kind of cheese compared to a cylindrical cheese (like
Gruyere). Simple expressions to determine diffusion coefficients of solutes have
been obtained by use of the generalized Fick law form as a constitutive equation for
Importance of Halophilic and Halotolerant Lactic Acid Bacteria in Cheeses 281
the diffusive molar flux of solutes; the shape of the cheese is very important to solve
this equation. With these expressions, the corresponding effective diffusion coeffi-
cients of NaCl and KCl in cheese were obtained (Zorrilla and Rubiolo 1994). A list
of effective diffusion coefficients for several cheeses has also been published
(Vélez-Ruiz 2011) which is very useful to evaluate salting times.
NaCl has many roles in cheese making. Some of these roles include preservation
against spoilage and pathogenic microorganisms, influence on the final flavor and
texture of the product, and as a complementary source of sodium. During cheese
ripening, the presence of salt involves many other key roles, such as the control of
microbial growth and activity and the control of various enzyme activities in the
cheese. It also has a strong influence on the physical changes in cheese proteins,
which in turn influence cheese texture, protein solubility, and protein conformation.
Salt, along with pH and calcium level, has a great effect on the extent of
para-κ-casein aggregation, which then affects the water-binding capacity of the
protein matrix, its tendency for syneresis, its rheological characteristics, and its
cooking properties (Guinee 2004).
excess of negatively charged ions will destabilize the structure of the molecule, due
to repulsion when the shielding cations are removed (Eisenberg and Wachtel 1987).
Our group has begun some research on the microbiota of certain Mexican regional
cheeses. Many cheeses in Mexico are prepared from raw milk, and the nonstarter
lactic acid bacteria (NSLAB) play a key role in the final characteristics of the
products. The first cheese to be analyzed was doble-crema (double-cream) cheese.
This cheese is traditionally prepared in the state of Chiapas, and the samples we
purchased had 48.2 % moisture, 5.34 % salt, aw of 0.972, and a pH of 4.03. The
halotolerant strains isolated from this cheese were Lactobacillus acidipiscis,
Enterococcus faecium, L. plantarum, L. farciminis, L. rhamnosus, and
L. pentosus. The second cheese, Cotija, is prepared in the state of Michoacán.
This product had 39.5 % moisture, 6.36 % salt, aw of 0.862, and a pH of 6.29. The
halotolerant lactic acid bacteria isolated included Lactobacillus acidipiscis,
Tetragenococcus halophilus, Weissella thailandensis, and L. pentosus (Morales
et al. 2011). These microorganisms will be studied to evaluate their role in cheese
ripening and their possible probiotic potential.
3 Modeling of Halotolerance
Halophiles can be defined as those organisms that can grow in the presence of salt
and which actually require high concentrations of NaCl. Microorganisms that can
withstand, but not require, high concentrations of salt to grow are known as
halotolerants. In the case of LAB, halophiles will only be able to grow in a salt-
added MRS medium. Our group was interested in evaluating these characteristics
for the halotolerant LAB isolated from the Mexican cheeses, so experiments were
performed and the results are presented here. The halotolerance of three strains of
LAB isolated from Chiapas double-cream cheese (Lactobacillus acidipiscis,
L. plantarum, and L. pentosus) was compared with that of two commercial probiotic
LAB strains (L. casei Shirota and L. plantarum 299v) by measuring the absorbance
at 560 nm after 24 h of incubation at 32 C in MRS broth added with salt
concentrations in the range of 0–16 %. Curves of inhibition at each concentration
were fitted by Mystat v12.02 software (Systat Software, Inc., Chicago, IL, USA)
using nonlinear least-squares regression in a sigmoidal dose–response model with
variable slope, also known as the four-parameter logistic equation (1):
DC
y¼Cþ b ð1Þ
1 þ lX50
Table 1 Parameters of the logistic dose–response model for halotolerance and determination
coefficient (R2) for five lactic acid bacteria
LAB strain C DC I50 b R2
L. acidipiscis 0.323 2 7.622 8.377 0.997
L. casei Shirota 0.365 1.825 7.895 7.225 0.995
L. plantarum 299v 0.390 2.021 7.573 6.458 0.989
L. plantarum 0.405 1.854 7.984 8.498 0.996
L. pentosus 0.289 1.989 8.606 4.685 0.975
L. casei Shirota was isolated from Yakul™ (Yakult Mexico, Mexico City, Mexico) and
L. plantarum 299v was isolated from Protransitus LP (Laboratorios Salvat, S.A., Barcelona, Spain)
0.5
0
0 5 10 15 20
[NaCl] %
The results are shown in Table 1. All of the curves were similar, and the curve
obtained for L. acidipiscis is shown as an example in Fig. 1. It can be observed that
all of the data could be fitted with high determination coefficients to a four-
parameter logistic dose–response equation. It can also be observed that the two
strains of L. plantarum have the highest halotolerance (i.e., they had the highest
C values) of all of the LAB tested. This is the first time that halotolerance has been
modeled using a dose–response curve.
4 Conclusions
Halophilic organisms are widespread and found in all three domains of cellular life
(archaea, bacteria, and eukaryote).
The “salt-in” strategy uses less energy, but requires intracellular adaptations.
Only a few prokaryotes use it.
All other halophiles use the “compatible solute” strategy, which is energy
expensive, but does not require special adaptations.
286 G. Melgar-Lalanne et al.
References
Payne MR, Morrison KR (1999) A multi-component approach to salt and water diffusion in
cheese. Int Dairy J 9(12):887–894
Rastelli E, Giraffa G, Carminati D, Parolari G, Barbuti S (2005) Identification and characterization
of halotolerant bacteria in spoiled dry-cured hams. Meat Sci 70:241–246
Vélez-Ruiz JF (2011) Mass transfer in cheese. In: El-Amin M (Ed). Advanced topics in mass
transfer. Intech Open Access ebook
Ventosa A, Nieto JJ, Oren A (1998) Biology of moderately halophilic aerobic bacteria. Microbiol
Mol Biol Rev 62(2):504–544
Wohlfart A, Severin J, Galinski EA (1990) The spectrum of compatible solutes in heterotrophic
halophilic eubacteria of the family Halomonadaceae. J Gen Microbiol 136:705–712
Zorrilla SE, Rubiolo AC (1994) A model for using the diffusion cell in the determination of
multicomponent diffusion coefficients in gels or foods. Chem Eng Sci 49(13):2123–2128
Influence of Water Activity and Molecular
Mobility on Peroxidase Activity in Solution
Abbreviations
aw Water activity
Tg Glass transition temperature
D Translational diffusion
WLF equation Williams–Landel–Ferry equation
HRP Horseradish peroxidase
1 Introduction
Water is one of the most critical components in foods and biological systems
because it affects several physicochemical properties of these systems with effects
on their chemical and enzymatic stability (Bell 2007).
Water availability in foods has historically been described by the concept of
water activity (aw), whereby water freedom in a system is expressed in terms of
relative water vapor pressure (Reid 2007). The relationship between water activity
and chemical and enzymatic reactions has been described by Labuza et al. (1970) in
their food stability map. In this map, the enzymatic activity is retained to be
positively affected by water activity.
Several authors have investigated the dependence of enzyme catalysis on aw
(Acker 1963, 1969; Tome et al. 1978; Schwimmer 1980; Lee and Kim 1995); in dry
solid foods, the aw increase upon water sorption resulted in the enhancement of
enzymatic activity, while in liquid models the aw decrease with solute addition did
not always depleted the enzymatic activity (Acker 1969). This happens because the
lowering of water activity by solute addition affects not only the aw of the system
but also its water content and molecular mobility.
Several authors (Farwell and Ackerman 1963; Silver and Karel 1981; Lim and
Reid 1991; Kerr et al. 1993; Kouassi and Roos 2000) discussed the inhibition of
enzymatic activity that occurs in concentrated food systems in terms of molecular
mobility, since in such systems the rates of chemical reactions may become diffusion
limited; in this case, the ability for reacting molecules to diffuse toward each other,
rather than the activation energy of the activated complex, is the rate limiting factor.
Enzymatic reactions are mainly controlled by diffusion in highly viscous aque-
ous systems, and the concentration of co-solutes was proven to influence the
catalysis via change in the physical properties of the system (Farwell and Ackerman
1963; Cicerone and Soles 2004). On the basis of this result, it is evident that water
activity is not the only factor that influences the enzymatic activity in concentrated
systems, and other physical parameters that describe the molecular mobility within
a system, such as viscosity and glass transition temperature (Tg), should be taken
into account (Kouassi and Roos 2000; Bell 2007).
Viscosity is referred to as an index of the mobility of a system since it is related
to the translational diffusion (D) by the Stokes relationship. Kramers’ theory
(Kramers 1940) predicts that the rate for protein dynamics should have an inverse
relationship with viscosity. Large amplitude (global) protein motions (particularly
those involving exterior portions of the protein) appear to follow Kramers’ relation
at viscosities above 1 cP (Ansari et al. 1992; Hagen et al. 1995), while diffusive
motions involving smaller portions of the protein are less likely to follow Kramers’
relation, especially at high viscosities. Deviations from Kramers’ relation for
exterior protein motions have been attributed to solvent composition and differ-
ences between bulk macroviscosity and local microviscosity in the immediate
vicinity of the protein. Moreover, by increasing the viscosity of the system, the
solvent motions may not be assumed infinitely fast compared to those of the solute
(Cicerone and Soles 2004).
At very high viscosities, protein dynamics are often discussed in terms of Tg
(Lim and Reid 1991; Kerr et al. 1993; Manzocco et al. 1999; Buitnik et al. 2000;
Cicerone and Soles 2004; Bell 2007), which corresponds to the temperature at which
a supercooled liquid loses its translational mobility and its molecules may exhibit
only rotational motions. Below glass transition temperature, chemical and biological
reactions are retained to be impeded, while, above Tg, molecules become mobile
and diffusion-limited reactions could occur (Slade and Levine 1991; Roos 2007).
Influence of Water Activity and Molecular Mobility on Peroxidase Activity in. . . 291
In order to study the influence of aw, bulk viscosity and glass transition temperature
on enzymatic activity, the latter was plotted as a function of the inverse of bulk
viscosity and the individual effect of aw lowering, and Tg shift due to cosolvent
addition was analyzed (Neri 2010). The T’g parameter was taken into consideration
instead of Tg because the latter is correlated to the water content, which in turn
292 G. Sacchetti et al.
affects both the aw and viscosity of the solution, while the T’g value is exclusively
related to the chemical characteristics (mostly the molecular weight) of the solute
(Roos and Karel 1991; Slade and Levine 1991). The T’g parameter could thus enable
the investigation of the specific solute effect in systems with different viscosities.
The lowering of water activity by the addition of ligands (sodium chloride,
glucose, and sorbitol) determined a reduction of enzymatic activity (Table 1).
Sodium chloride, when added in small amounts, depleted the aw of the solutions
without or little affecting their T’g and viscosity, while glucose, being a larger
solute and a weaker osmotic agent, depleted the aw of the solutions only when
present at high concentrations, thus also determining an increase of the system
viscosity and T’g (Table 1). The addition of maltodextrin with DE 8.7 increased the
viscosity and T’g of the solutions by only slightly affecting their water activity
(Table 1). With the addition of ligands to polymer solutions, it was possible to
obtain solutions with the same water activity and different viscosities and T’g
values.
The addition of both ligands and maltodextrin limited HRP activity; salt affected
water activity by minimally affecting bulk viscosity and Tg, while maltodextrin
increased viscosity by minimally affecting aw. The rate of enzymatic reaction was
not linearly correlated to η1, as predicted by Kramers’ theory, but a power-law
relationship k / ην between viscosity and protein dynamics was observed (Fig. 1).
This indicates that in the high viscosity range, the rate of enzymatic reactions is
much more hindered by a viscosity increase than in low viscosity ranges.
At high water activity (aw ¼ 0.98), sodium chloride addition inhibited enzymatic
activity more than the other co-solutes due to both its ionic nature, which could
affect the rate of enzymatic reaction at low viscosities by electrostatic effects
(Farwell and Ackerman 1963), and to a specific interaction with HRP (Laurenti
et al. 2000). On the contrary, at water activity value of 0.87, glucose was more
effective than sodium chloride in limiting enzymatic activity due to its viscosity-
improving effect (Table 1).
Glucose inhibited enzyme activity by affecting the aw, viscosity, and T’g of the
system. When viscosity was equal, maltodextrin limited HRP activity more than
glucose even though negligible water activity changes occurred (Fig. 1). This result,
which is due to the higher T’g of maltodextrin, seems to confirm the hypothesis of
Kouassi and Roos (2001), which suggested that molecular mobility is an important
factor for the regulation of enzyme activity even at high water activities.
The effect of T’g on enzymatic activity was further studied using maltodextrins
of different molecular weights. In diluted solutions with low viscosity values, an
increase of maltodextrin size from DE 33 to DE 8.7 did not limit HRP activity,
likely due to the increase of aw, but a further increase of molecular size to DE 2.5
determined a decrease of enzymatic activity (Neri et al. 2011). In highly
Table 1 Composition and physical properties of binary and ternary solutions and HRP activity in the solutions
Maltodextrin Kinematic
Solute wS/wW wM/wW viscosity
Solutes (S) (g 100 g1) (g 100 g1) Water activity (106 m2s1) T’g ( C) HRP activity (%)
Phosphate buffer 0 0 0.999 0.003 1.09 0.00 n.d. 100.00 0.25
Maltodextrin DE8.7 0 15.00 0.997 0.003 3.49 0.01 10.5 0.07 71.17 5.22
0 30.00 0.995 0.003 20.1 0.03 10.1 0.28 37.28 4.27
Sodium chloride 3.52 0 0.981 0.003 1.09 0.03 n.d. 93.69 1.48
18.00 0 0.866 0.003 1.27 0.00 n.d. 80.39 1.61
3.52 15.00 0.978 0.003 2.96 0.05 10.7 0.3 59.62 1.38
3.52 30.00 0.976 0.003 8.92 0.07 10.7 0.8 41.09 0.91
Sorbitol 18.56 0 0.978 0.003 1.61 0.01 44.8 0.3 99.94 1.51
18.56 15.00 0.975 0.003 4.33 0.02 38.7 0.1 76.02 1.47
18.56 30.00 0.973 0.003 11.3 0.32 35.5 0.1 59.88 2.40
Glucose 16.33 0 0.980 0.003 1.49 0.01 44.2 0.4 97.15 2.93
50.00 0 0.873 0.003 9.28 0.14 43.5 0.4 68.63 5.87
16.33 15.00 0.977 0.003 4.14 0.24 35.7 0.1 66.17 2.20
16.33 30.00 0.975 0.003 11.3 0.14 30.3 0.2 40.99 1.94
Maltodextrin DE 33 0 10.00 0.997 0.003 1.39 0.00 21.8 0.1 87.47 1.94
0 20.00 0.998 0.003 2.14 0.00 21.5 0.3 74.88 1.54
0 30.00 0.993 0.003 3.92 0.00 21.3 0.3 62.62 0.78
Influence of Water Activity and Molecular Mobility on Peroxidase Activity in. . .
Fig. 1 Horseradish peroxidase activity in glucose and maltodextrin systems as a function of the
inverse of kinematic viscosity (modified from Neri et al. 2011)
In order to better understand the role of water activity and viscosity on enzyme
activity, ternary systems characterized by different water activities and viscosities
were obtained by the combination of salt, glucose, sorbitol, and maltodextrin with
DE 8.7 (Table 1).
In ternary solutions with the same viscosity, the aw reduction, upon salt addition,
determined a peroxidase inhibition (Fig. 2). Sodium chloride inhibited HRP activity
exclusively by depleting aw since it did not affect T’g (Neri et al. 2011).
The inhibition of HRP by water activity reduction was more effective at low
viscosity than at high viscosity, thus suggesting that in high viscosity systems
enzyme inhibition is mainly due to the hindrance of molecular mobility. On the
other hand, the hindrance of enzyme activity by a viscosity increase was more
effective at the lowest water activity than at the highest, and this result seems to be
in contrast with that suggested by Kouassi and Roos (2002).
In sorbitol–maltodextrin ternary solutions, peroxidase activity was always
inhibited by a viscosity increase (Fig. 3). On the other hand, water activity reduction
from 0.99 to 0.98 determined an increase of HRP activity. By taking into account
the T’g changes, it is possible to affirm that, viscosity being equal, aw depletion
increased the HRP activity because sorbitol addition to the maltodextrin solution
Influence of Water Activity and Molecular Mobility on Peroxidase Activity in. . . 295
determined a dramatic decrease of the T’g (Neri et al. 2010). On the basis of the
overall data trend (Fig. 2), aw and T’g seemed to play a minor role in HRP inhibition
in highly viscous solutions, where the enzymatic activity is mainly dependent on
viscosity.
In glucose–maltodextrin ternary solutions, peroxidase activity was always
inhibited by a viscosity increase (Fig. 4) as well, and this effect was more evident at
lower aw values than at higher aw values. In maltodextrin systems, lowering of water
296 G. Sacchetti et al.
activity by glucose addition determined an HRP activity increase in the low viscosity
region and an HRP activity decrease in the high viscosity region. In the latter region,
water activity being equal (0.98), glucose and ternary glucose–maltodextrin solu-
tions showed a higher T’g than sorbitol and ternary sorbitol–maltodextrin solutions
(Table 1), permitting direct observation of a negative effect of water activity lowering
on HRP activity.
By taking into account the T’g changes, it is possible to affirm that, viscosity
being equal, an aw depletion from 0.99 to 0.98 could differently affect HRP activity
depending on the effect of the co-solute on the T’g of the system (Neri 2010; Neri
et al. 2010, 2011). On the other hand, aw and T’g seem to play a minor role in HRP
inhibition in highly viscous solutions, where the enzymatic activity is mainly
dependent on the viscosity-dictated molecular mobility.
On the basis of this data, it is possible to affirm that, although viscosity is the
main factor regulating HRP activity in concentrated solutions, T’g and aw changes
should not be overlooked in order to describe or predict the enzymatic activity in a
system.
3 Conclusions
References
Acker L (1963) Enzyme activity at low water contents. In: Leitch JM, Rhodes DN (eds) Recent
advances in food science, 3rd edn. Butterworths, London, pp 239–347
Acker L (1969) Water activity and enzyme activity. Food Technol 23:1257–1270
Ansari AC, Jones CM, Henry ER, Hofrichter J, Eaton WA (1992) The role of protein viscosity in
the dynamics of protein conformational changes. Science 256:1796–1798
Bell LN (2007) Moisture effects on food’s chemical stability. In: Barbosa-Cánovas GV et al (eds)
Water activity in foods. Fundamentals and applications. Blackwell Publishing, Oxford, pp
173–198
Buitnik J, van den Dries IJ, Hoekstra FA, Alberda M, Hemminga M (2000) High critical
temperature above Tg may contribute to the stability of biological systems. Biophys J
79:1119–1128
Burnette FS (1977) Peroxidase and its relationship to food flavor and quality: a review. J Food Sci
42:1–6
Chen YH, Aull JL, Bell LN (1999a) Invertase storage stability and sucrose hydrolysis in solids as
affected by water activity and glass transition. J Agric Food Chem 47:504–509
Chen YH, Aull JL, Bell LN (1999b) Solid state tyrosinase stability as affected by water activity
and glass transition. Food Res Int 32:467–472
Cicerone MT, Soles CL (2004) Fast dynamics and stabilization of proteins: ternary glasses of
trehalose and glycerol. Biophys J 86:3836–3845
Farwell RW, Ackerman E (1963) Effect of physical factors on reactions of horse-radish peroxidase
complexes with reduced cytochrome c. Biophys J 86:479–491
Hagen SJ, Hofrichter J, Eaton WA (1995) Protein reaction kinetics in a room temperature glass.
Science 269:959–962
Kerr WL, Lim MH, Reid DS, Chen H (1993) Chemical reaction kinetics in relation to glass
transition temperatures in frozen food polymer solutions. J Sci Food Agric 61:51–56
Kouassi K, Roos YH (2000) Glass transition and water effects on sucrose inversion by invertase in
a lactose-sucrose system. J Agric Food Chem 48:2461–2466
Kouassi K, Roos YH (2001) Glass transition and water effects on sucrose inversion in
noncrystalline carbohydrate food systems. Food Res Int 34:895–901
Kouassi K, Roos YH (2002) Glass transition, water, and glycerol effects on sucrose inversion in
pullulan-sucrose systems. J Food Sci 67:3402–3407
Kramers HA (1940) Brownian motion in a field of force and the diffusion model of chemical
reaction. Physica 7:284–304
Labuza TP, Tannenbaum SR, Karel M (1970) Water content and stability of low moisture and
intermediate moisture foods. Food Technol 24:543–550
298 G. Sacchetti et al.
Laurenti E, Suriano G, Ghibaudi EM, Ferrari RP (2000) Ionic strength and pH effect on the Fe(III)-
imidazolate bond in the heme pocket of horseradish peroxidase: an EPR and UV-visible
combined approach. J Inorg Biochem 81:259–266
Lee SB, Kim KJ (1995) Effect of water activity on enzyme hydration and enzyme reaction rate in
organic solvent. J Ferment Bioeng 79:473–478
Lim MH, Reid DS (1991) Studies of reaction kinetics in relation to the Tg of polymers in frozen
model systems. In: Slade L, Levine H (eds) Water relationships in food. Plenum, New York, pp
102–122
Manzocco L, Nicoli MC, Anese M, Pitotti A, Maltini E (1999) Polyphenoloxidase and peroxidase
activity in partially frozen systems with different physical properties. Food Res Int 31:363–370
Nelson KA, Labuza TP (1994) Water activity and food polymer science: Implications of state on
Arrhenius and WLF models in predicting shelf life. J Food Eng 22:271–289
Neri L (2010) Enzymatic inactivation and quality of semi-finished plant foods intended for
freezing. PhD dissertation. University of Teramo, Italy
Neri L, Pittia P, Bertolo G, Torreggiani D, Sacchetti G (2011) Influence of water activity and
molecular mobility on peroxidase activity in glucose and maltodextrin solution. Food Biophys
6:281–287
Neri L, Pittia P, Bertolo G, Torreggiani D, Sacchetti G (2010) Influence of water activity and
molecular mobility on peroxidase activity in salt and sorbitol-maltodextrin systems. J Food
Eng 101:289–295
Reid DS (2007) Water activity: fundamentals and relationships. In: Barbosa-Cánovas GV
et al (eds) Water activity in foods. Fundamentals and applications. Blackwell Publishing,
Oxford, pp 15–28
Roos Y (2007) Water activity and glass transition. In: Barbosa-Cánovas GV (ed) Water activity in
foods: fundamentals and applications. Blackwell Publishing, Oxford, pp 29–45
Roos Y, Karel M (1991) Phase transitions of mixtures of amorphous polysaccharides and sugars.
Biotechnol Progr 7:49–53
Schwimmer S (1980) Influence of water activity on enzyme reactivity and stability. Food Technol
34(5):64–74, 82, 83
Silver M, Karel M (1981) The behavior of invertase in model systems at low moisture contents.
J Food Biochem 5:283–311
Slade L, Levine H (1991) Beyond water activity: recent advances based on an alternative approach
to the assessment of food quality and safety. Crit Rev Food Sci Nutr 30:115–360
Terefe NS, Van Loey A, Hendrickx M (2004) Modelling the kinetics of enzyme-catalysed
reactions in frozen systems: the alkaline phosphatase catalysed hydrolysis of di-sodium-p-
nitrophenyl phosphate. Innov Food Sci Emerg 5:335–344
Tijskens LMM, Rodis PS, Hertog MLATM, Waldron KW, Ingham L, Proxenia N, van Dijk C
(1997) Activity of peroxidase during blanching of peaches, carrots and potatoes. J Food Eng
34:355–370
Tome D, Nicolas J, Drapron R (1978) Influence of water activity on the reaction catalyzed by
polyphenoloxidase (E.C.1.14.18.1.) from mushrooms in organic liquid media. Lebensm-Wiss
Technol 11:38–41
Williams ML, Landel RF, Ferry JD (1955) Temperature dependence of relaxation mechanisms in
amorphous polymer and other glass-forming liquids. J Am Chem Soc 77:3701–3707
Phase Transitions in Sugars and Protein
Systems: Study of Stability of Lysozyme
in Amorphous Sugar Matrices
Abbreviations
1 Introduction
decrease in their activity (Passot et al. 2005; Liao et al. 2004; Hinrichs et al. 2001;
Heller et al. 1999).
In the last decade the pharmaceutical and biochemical industry has turned to
vitreous materials of sugars and analogous compounds as preserving agents in the
drying processes and storage of thermolabile molecules (Nicholls 2004; Ohtake and
Wang 2011). The idea of using amorphous sugars for preservation purposes comes
from mimicking the natural phenomenon of anhydrobiosis observed in some
species of spores, fungi, plants, and other organisms, which have the capacity to
survive complete dehydration and extreme conditions of temperature (Buitink and
Leprince 2008; Buitnik and Leprince 2004). This resistance to drought has been
associated to the biosynthesis of nonreducing sugars like trehalose and sucrose.
The mechanism that explains the specific participation of these sugars in the
biopreservation phenomenon is a controversial subject; some scientists (Crowe
et al. 1984; Carpenter and Crowe 1989) suggest that trehalose and other reducing
sugars can form hydrogen bonds that can replace the essential water molecules
responsible for maintenance of the tertiary structure of proteins. In numerous
studies it has been demonstrated that biological membranes conserve their structure
and functionality when they are dehydrated in the presence of trehalose or other
saccharides (Higl et al. 2007; Patist and Zoerb 2005; Oliver et al. 2001); neverthe-
less, it is not clear why not all substances that can form hydrogen bonds confer
preservation with the same efficiency as nonreducing sugars like trehalose or
sucrose. In one study conducted by Crowe et al. (1984), it was found that the degree
of efficiency in the preservation of dehydrated of membranes in the presence of
polyols studied was: glycerol < sucrose < maltose < trehalose. Analyzing these
results, Green and Angell (1989) showed that this order coincides with the increas-
ing order of their glass transition temperatures (Tg). This finding opened the
vitrification theory, which proposes that in order to observe the preservation
phenomenon it is essential that the saccharides are in their glassy state, at which
the viscosity reaches values as great as 1012 Pa s, where the movement of molecules
and diffusion phenomena are extremely slow.
At the glassy state most chemical reactions, such as the decomposition process,
are temporarily suspended. Several studies have shown that when nonreducing
sugars are present, the thermal stability of biomolecules is increased, and their
structure is preserved during the drying processes due to the in situ formation of a
vitreous matrix. Therefore, proteins are immobilized in the glass matrix and thus
protected from the degenerative effects of the processes of crystallization of water
or solutes present in the formulation during water loss or drying (Buitink and
Leprince 2004). This process allows storage at room temperature and eliminates
the need for refrigeration (Nicholls 2004; Worrall et al. 2001; Fox 1995) Although
it is clear that sugar vitrification has much to do with the preservation (Higl
et al. 2007; Duncan et al. 1999) of biomolecules, there is evidence that other
parameters, including the presence of the proteins, play an important role in the
glass formation system and, therefore, on the preservation phenomenon (Buitink
et al. 2000). These theories only partially explain the mechanism of stabilization,
since not all compounds with hydroxyl groups are similar to trehalose, nor are all
Phase Transitions in Sugars and Protein Systems: Study of Stability of. . . 301
A glass can be considered as a liquid that has lost its ability to flow or as an
amorphous solid with structural features typical of a liquid that presents glass
transition temperature, which is the temperature at which a glassy material loses
thermomechanical stability and changes from an amorphous solid highly viscous
(1012 Pa s) into a liquid. The transformation of a liquid into a glass can be observed
by monitoring changes in the material’s properties as the enthalpy is a function of
temperature. Experimental measurement of the glass transition temperature can be
done by measuring the heat capacity of a material as a function of temperature. The
Tg determines the temperature at which thermomechanical stability and viscoelastic
properties of an amorphous solid start to change to be transformed into a
supercooled liquid (Angell 1995). Figure 1 shows the three ways in which a Tg
can be reported based on the position of the “transition.” To achieve an efficient
system for the purpose of preservation, a transition temperature above room
temperature is preferred.
302 L.M. Martı́nez et al.
Heat Flow/mW
24.5
measurement corresponds
to a nonreducing sugar endpoint
24.0
sample, measured with a
differential scanning midpoint
calorimeter (DSC) at 23.5
heating rate of 10 C/min onset
23.0
22.5
60 80 100 120 140 160 180
Temperature/oC
Table 1 Glass transition temperature Tg of three nonreducing sugars reported in the literature
Method of preparation of the Sucrose Trehalose Raffinose
amorphous solid (water content) Tg ( C) Tg ( C) Tg ( C) Reference
Freeze drying (Karl Fischer < 0.2 %) 74 115 102 Zografi and
Saleki-
Gerhardt
(1994)
Freeze drying (Karl Fischer < 0.3 %) 70 (onset) 114 (onset) NR Sheri
78 123 et al. (1999)
(endpoint) (endpoint)
Heating (0.1–0.5 g) above melting 62 100 NR Roos (1993)
temperature (water content not mea-
sured; dehydration assumed due to
melting temperature)
Quick drying of 10 μl of water-solution 57 108 108 Wolkers
[20 mg/ml of sugar] (no water assumed et al. (2004)
due to absence of band at 1650 cm1)
The evaluation of thermal stability (Tm) and changes in the conformation of bio-
molecules such as enzymes and proteins as a function of temperature is also very
important in order to understand its preservation. Thermal analysis can be used to
determine the maximum temperature at which proteins are stable (denaturation
temperature, Tm) and the enthalpy associated with this process. To describe this
process, it can be considered that a protein is a molecule composed of a large
number of amino acids, linked together by peptide bonds (primary structure). In
addition, the side chains of these amino acids, can establish interactions of the type
called “hydrogen bonds,” which causes the protein to acquire a characteristic shape
and suitable for the performance of their intended functions (secondary structure).
The process of distortion, then, according to the Lumry and Eyring model (Lumry
and Eyring 1954), consists of two stages: the first is the cleavage of the protein and
may be reversible, and the second involves a breakdown of the molecule and is an
irreversible process. Either of these steps involves a certain amount of energy being
released or absorbed, so it is expected to be possible to detect the denaturation
process using thermal analysis.
In a study by Nath et al. (1998) thermal stability of the enzyme alcohol dehydro-
genase in aqueous solution was investigated by measuring the biological activity of
the enzyme at different temperatures and measuring the denaturation process using
differential scanning calorimetry (DSC). Although these two measurements are
based on different principles (kinetic in the case of measurement of biological
activity and thermodynamics using DSC), both measurements lead to equivalent
results for the thermal stability of protein, in which denaturation of the enzyme
occurred at 63 and 70 C in the absence and presence of sucrose (44 % w/w),
respectively. Although there are other publications (Cueto et al. 2003; Sun
et al. 1998; Sochava 1997) demonstrating that techniques based on thermal analysis
can be very useful for evaluating protein stability, most of these studies have been
carried out either in solution or at the dry state (freeze-dried or spray-dried proteins–
sugar systems) (Lu et al. 2007; Liao et al. 2002a, b; Bell et al. 1995), but no systematic
study has been carried out using the same model biomolecule to correlate the
preservation effect of the same preservation agents, in solution and at the dry state.
The experimental work of the present study of phase transitions of sugars and
lysozyme was carried out in the following steps: (1) thermal characterization of
pure amorphous sugars; (2) evaluation of the thermal stability of enzymes in
solution, studying the effect of concentration of sugar and concentration of preser-
vation agent; and (3) evaluation of the thermal stability of enzyme–carbohydrate
systems after drying in the presence of different sugars.
304 L.M. Martı́nez et al.
Fig. 2 Thermograms 22
showing glass transition raffinose
20
temperatures of the
disaccharides sucrose and 18 trehalose
trehalose and a
Heat Flow/mW
16
trisaccharide: raffinose. sucrose
Values reported correspond 14
to the onset temperature.
12
Heating rate of
measurement 10 C/min Sugar Tg/◦C
10
Sucrose 67
8 Raffinose 112
6 Trehalose 115
4
40 60 80 100 120 140
Temperature/oC
In order to study thermal stability and phase transition of proteins with and without
preservation agents in solution, denaturation temperature (Tm) of the enzyme was
evaluated in aqueous solution, first at different concentration of the enzyme, then
Phase Transitions in Sugars and Protein Systems: Study of Stability of. . . 305
a 3.0
2.5
Heat Flow/mW
2.0 ΔH
1.5
1.0
0.5
0.0
60 62 64 66 68 70 72 74 76 78 80
Temperature/oC
.
b 8
7
6
.
ΔH/J g−1
5
4
3
.
.
2
1
0
5 10 15 20 25 30 35 40 45
% of lysozyme w/w
with different concentrations of preservation agents, and finally with different types
of preservation agents.
(a) Effect of concentration of enzyme
Aqueous solutions of lysozyme were prepared containing 10, 20, 30, and 40 %
(w/w) of the enzyme, respectively. The temperature and enthalpy of denaturation
were measured by differential scanning calorimeter (DSC) using a heating rate of
10 C/min. Figure 3a shows an example of the thermograms obtained; the denatur-
ation temperature of lysozyme in solution occurred between 71 and 72 C. The
denaturation enthalpy, corresponding to the area beneath the curve, was calculated
for each solution. Good linear correlation of the denaturation enthalpy with the
enzyme concentration is shown in Fig. 3b. It is important to mention that as the
concentration of the enzyme increases, the standard deviation of the measurements
increases. It is probable that the saturation of the solution interferes with this
measurement.
306 L.M. Martı́nez et al.
50%
40%
Diff. Thermovoltage/a.u. w/o trehalose
40 50 60 70 80 90
Temperature/oC
Fig. 4 Thermograms showing Tm of lysozyme in aqueous solution and in the presence of different
amounts of trehalose. Measurements are done at a heating rate of 10 C/min
For the purpose of studying the effect of type of sugar on thermal stability of the
enzyme in solution, three types of sugars were added to aqueous solutions of 20 %
w/w lysozyme. It is important to mention that the effect of type of sugar on solution
can only be analyzed in highly concentrated solutions (above 45 % w/w). To
compare these results in the dried state, aqueous solutions of lysozyme were
freeze-dried in the presence of the same sugars. Figure 5a shows thermograms
obtained on solution and Fig. 5b shows results of freeze-dried samples.
Phase Transitions in Sugars and Protein Systems: Study of Stability of. . . 307
a 0.1
b 17
Thermovoltage/a.u.
Thermovoltage/a.u.
16
0.09
Sucrose
14
0.07
13
sucrose
0.06
12
0.05
11
0.04
10
Raffinose
0.03
0.02
Trehalose
trehalose
8
0.01
0.0 6
Freeze-dried
Thermal denaturation of lysozyme, Tm/oC
150
145
140 Suc
135
Aqueous solutions
95
Suc
90
85
60 70 80 90 100 110 120
Glass transition temperature of excipient, Tg/oC
Fig. 6 Comparison of Tm and Tg of enzyme in solution and freeze-dried. The tendency of thermal
stability observed in aqueous solutions is similar for freeze-dried samples
In order to study the synergic effects of several sugars on the same system, formu-
lations based on mixtures of sugars were also analyzed. From the results obtained, it was
observed that the higher the glass transition temperature of the excipient (pure sugar or
the mixture), the higher the Tm of the enzyme, independent of whether the enzyme is in a
dried state or in an aqueous concentrated solution. See Fig. 6.
308 L.M. Martı́nez et al.
160
Tre
Tre Raf
Tm of lysozyme/oC 155 Raf
150
Sugar
Sugar
145 mixtures
mixtures
140 Suc
Suc
135
40 50 60 70 80 90 100 110 120
Tg lysozyme+excipient Tg of the excipient
Fig. 7 Denaturation temperatures of lysozyme and glass transition temperature of the freeze-dried
enzyme–sugar systems
140
120
100
80
60
Trehalose Raffinose Sucrose
Sugar
3 Conclusions
The degree of thermal stability conferred to the enzyme by sugar glasses depends
on concentration of the excipient, type of excipient, and drying process, among
other parameters. The selection of an efficient preservation agent for freeze-dried
biomolecules can be a time-consuming task if the set of experimental parameters
considered in order to determine its optimal performance is incomplete. Therefore,
it is relevant for development of the appropriate preservation agent to validate
experimental test procedures of thermal stability of biomolecules directly in con-
centrated solution, which allows a much faster and useful tool to make predictions
on efficiency of thermal stability, since these results can be extrapolated to the
performance of the preservation agents in dried systems. The results of this work,
summarized in Fig. 8, show how by using information from concentrated aqueous
solutions we have been able to improve thermal stability of the same enzyme when
compared with other results from the literature.
References
Angell A (1995) Formation of glasses from liquids and biopolymers. Science 267:1924–1935
Bell LN, Hageman MJ, Muraoka LM (1995) Thermally induced denaturation of lyophilized
bovine somatotropin and lysozyme as impacted by moisture and excipients. J Pharm Sci
84:707–712
Buitink J, Van den Dries IJ, Hoekstra FA, Alberda M, Hemminga MA (2000) High critical
temperature above Tg may contribute to the stability of biological systems. Biophys J 79
(2):1119–1128
Buitink J, Leprince O (2004) Glass formation in plant anhydrobiotes: survival in the dry state.
Cryobiology 48:215–228
Buitink J, Leprince O (2008) Intracellular glasses and seed survival in the dry state. C R Biol 331
(10):788–795
Carpenter JF, Crowe JH (1989) An infrared spectroscopic study of the interactions of carbohy-
drates with dried proteins. Biochemistry 28:3916–3922
Crowe LM (1984) Lessons from nature: the role of sugars in anhydrobiosis. Comp Biochem
Physiol A 131:505–513
Crowe JH, Crowe LM, Womersley C, Mouradian R (1984) Preservation of functional integrity
during long term storage of a biological membrane. Biochim Biophys Acta 778(3):615–617
Cueto M, Dorta MJ, Munguı́a O, Llabrés M (2003) New approach to stability assessment of
protein solution formulations by differential scanning calorimetry. Int J Pharm 252:159–166
Duncan QM, Royal PG, Kett VL, Hopton ML (1999) The relevance of the amorphous state to
pharmaceutical dosage forms: glassy drugs and freeze dried systems. Int J Pharm 179:179–207
Fox KC (1995) Putting proteins under glass. Science 267:1922–1923
Green JL, Angell CA (1989) Phase relations and vitrification in saccharide-water solutions and the
trehalose anomaly. J Phys Chem 93:2880–2882
Grigera JR, Bolzicco V (2008) Molecular mobility and ageing of sugar glasses. Food Chem 106
(4):1314–1317
Heller M, Carpenter J, Randolph T (1999) Protein formulation and lyophilization cycle design:
prevention of damage due to freeze-concentration induced phase separation. Biotechnol
Bioeng 63(2):166–174
310 L.M. Martı́nez et al.
Higl B, Kurtmann L, Carlsen CU, Ratjen J, F€orst P, Skibsted LH, Ulrich K, Risbo J (2007) Impact
of water activity, temperature, and physical state on the storage stability of Lactobacillus
paracasei ssp. Paracasei freeze-dried in a lactose matrix. Biotechnol Progr 23:794–800
Hinrichs WLJ, Prinsen MG, Frijlink HW (2001) Inulin glasses for the stabilization of therapeutic
proteins. Int J Pharm 215:163–174
Liao Y-H, Brown MB, Martin GP (2004) Investigation of the stabilisation of freeze-dried
lysozyme and the physical properties of the formulations. Eur J Pharm Biopharm 58:15–24
Liao Y-H, Brown M, Quader A, Martin GP (2002a) Protective mechanism of stabilizing excipients
against dehydration in the freeze-drying of proteins. Pharm Res 19(12):1854–1861
Liao Y-H, Brown MB, Nazir T, Quader A, Martin GP (2002b) Effects of sucrose and trehalose on
the preservation of the native structure of spray-dried lysozyme. Pharm Res 19(12):1847–1853
Lumry R, Eyring H (1954) Conformation changes of proteins. J Phys Chem 58:110–120
Lu J, Wang X-J, Liu Y-X, Ching C (2007) Thermal and FTIR investigation of freeze-dried protein-
excipient mixtures. J Therm Anal Calorim 89(3):913–919
Martı́nez LM, Videa M, Mederos F, Mesquita J (2007) Constructing a high-sensitivity, computer-
interfaced, differential thermal analysis device for teaching and research. J Chem Educ 84
(7):1222–1223
Martı́nez LM, Videa M, Mesquita, J (2013) Design, construction and calibration of a portable
multi sample DTA setup. Thermochim. Acta 560:89–94
Martı́nez LM, Videa M, Mederos F, de Moral Y (2011) Preservation effect of vitreous non
reducing carbohydrates on the enzymatic activity, denaturation temperature and retention of
native structure of Lysozyme. J Mex Chem Soc 55(3):185–189
Nath S, Satpathy GR, Mantri R, Deep S, Ahluwalia J (1998) Thermal stability of alcohol
dehydrogenase enzyme determined by activity assay and calorimetry. Thermochim Acta
309:193–196
Nicholls H (2004) Cash injection for thermostable vaccines. Drug Discov Today 9(22):945
Ohtake S, Wang J (2011) Trehalose: current use and future applications. J Pharm Sci 100
(6):2020–2053
Oliver AE, Leprince O, Wolkers WF, Hincha DK, Heyer AG, Crowe JH (2001) Non-disaccharide-
based mechanisms of protection during drying. Criobiology 43(2):151–167
Passot S, Fonseca F, Alarcon-Lorca M, Rolland D, Marin M (2005) Physical characterisation of
formulations for the development of two stable freeze-dried proteins during both dried and
liquid storage. Eur J Pharm Biopharm 60(3):335–348
Patist A, Zoerb H (2005) Preservation mechanisms of trehalose in food and biosystems. Colloids
Surf 40(2):107–113
Richards AB, Krakowka S, Dexter LB, Schmid H, Wolterbeek APM, Waalkens-Berendsen DH,
Shigoyuki A, Kurimoto M (2002) Trehalose: a review of properties, history of use and human
tolerance, and results of multiple safety studies. Food Chem Toxicol 40(7):871–898
Roos Y (1993) Melting and glass transitions of low molecular weight carbohydrates. Carbohydr
Res 238:39–48
Sheri LS, Tang X, Chang L, Hancock BC, Pikal MJ (1999) Characterization of the time scales of
molecular motion in pharmaceutically important glasses. J Phys Chem B 103(20):4113–4121
Sochava IV (1997) Heat capacity and thermodynamic characteristics of denaturation and glass
transition of hydrated and anhydrous proteins. Biophys Chem 69:31–41
Sun WQ, Davison P, Chan HSO (1998) Protein stability in the amorphous carbohydrate matrix:
relevance to anyhydrobiosis. Biochim Biophys Acta 1425:245–254
Wanh W (2000) Lyophilization and development of solid protein pharmaceuticals. Int J Pharm
203:1–60
Wolkers WF, Oliver A, Tablin F, Crowe JH (2004) Fourier-transform infrared spectroscopy study
of sugar glasses. Carbohydr Res 339(6):1077–1085
Worrall EE, Litamoi JK, Seck BM, Ayelet G (2001) Xerovac: an ultra rapid method for the
dehydration and preservation of live attenuated Rinderpest and Peste des Petits ruminants
vaccines. Vaccine 19:834–839
Zografi G, Saleki-Gerhardt A (1994) Non-isotthermal crystallization of sucrose from amorphous
state. Pharm Res 11(8):1166–1173
An Exhaustive Study on Physicochemical
and Structural Changes During Drying
of Apple Discs
Abbreviations
1
H-NMR Proton nuclear magnetic resonance
A Area under curve
Ai Signal amplitude
ANOVA Analysis of variance
BI Browning index
CPMG Carr–Purcell–Meiboom–Gill pulse sequence
D Deformation at the rupture point
ESEM Environmental scanning electronic microscopy
FID Flame ionization detector
Fmax Maximum rupture force
I Signal intensity of the water strongly associated with the matrix
LM Light microscopy
1 Introduction
Dehydration is one of the most employed techniques in the preservation and devel-
opment of foods. Water is a vital component in the structural preservation and
functional integrity of vegetable tissues. As a consequence of its removal during
convective air drying, great physicochemical changes in biopolymer composition and
spatial conformation are presented, inducing disarrangements and/or destruction of
the natural structure, shrinkage, changes in water and biopolymer mobility and
distribution, changes in thermal transitions, color changes, and modification in
mechanical properties, among others. These characteristics have been identified as
the most important properties associated with modifications in quality properties of
fruits (Chapman 1994; Krokida et al. 1998, 2000; Ruan and Chen 1998).
The objectives of this study were: (1) to analyze, during air drying of apple
tissue, the changes in water mobility, thermal transitions, mechanical behavior,
color changes, and shrinkage, and (2) to relate those changes with the structure
modifications determined by microscopy.
Apple (Malus pumila, Granny Smith cv) discs, 3 cm in diameter and 0.6 cm in
thickness, were air-dried (60 C, 3 m/s) during different times to obtain products
with different moisture contents and relative vapor pressure values (RVP: 0.98, 0.96,
0.94, 0.91, 0.87, 0.82, 0.75, 0.60, 0.47, 0.40, 0.30) (D’Ambrogio de Argüeso 1986).
Water content was determined gravimetrically in a vacuum oven (60 C). RVP was
measured with a hygrometer Aqua Lab (CX-2, Decagon Devices Inc., USA). Water
mobility was studied by measuring the spin–spin relaxation times T2 using a low field
nuclear magnetic resonance spectrometer Bruker Minispec mq20 (Bruker Biospin
Gmbh, Germany).Flame ionization detector (FID) method was applied to determine
An Exhaustive Study on Physicochemical and Structural Changes During Drying. . . 313
the T2 corresponding to protons with fast relaxation as related to the solid matrix and
water interacting strongly with them (Eq. 1).Carr–Purcell–Meiboom–Gill (CPMG)
sequence was applied to study the protons with high mobility as related to free water
and water with low interaction with solids. Experimental curves from CPMG
sequence were fitted by nonlinear regression according to Eq. (2):
2
t
t
I ¼ As e þ AL e ð1Þ
T 2S T 2L
X t
I¼ Ai e ð2Þ
T 2i
where I is the signal intensity at time t, Ai is the signal amplitude, and T2i is the spin–
spin relaxation time, T2S is the relaxation time of the protons of polymeric chains,
and T2L is the proton relaxation time of the water strongly associated with the
matrix. Thermal transitions were studied by differential scanning calorimetry (DSC
822e, Mettler Toledo, Switzerland). Mechanical properties (maximum rupture
force, Fmax; deformation at the rupture point, D; slope in the initial linear portion
of the force–deformation curve, S; and area under curve, A) were determined by
puncture test with an Instron Testing Machine (model 1101, USA). Browning
degree was evaluated with a spectrocolorimeter (Minolta CM-508-d). Shrinkage
was studied by volume displacement method using n-heptane as solvent. Structure
changes were evaluated by light microscopy (LM) (Carl Zeiss Axioskope BI2,
Germany) and environmental scanning electronic microscopy (ESEM) (Phillips,
XL30, Holland) (Snaar and Van As 1992; Rao 2003). The results were statistically
analyzed using Infostat program (v. 2009, Cordoba National University, Argen-
tina). Multivariate analysis of variance (MANOVA) and principal component
analysis (PCA) were used to analyze rheological properties, water mobility, and
color data. Analysis of variance (ANOVA) was applied to analyze shrinkage data.
Maximum rupture force Fmax increased as drying time increased, while the slope of
the initial linear portion S decreased up to RVP 0.96. Then remained approximately
constant until RVP 0.47 and abruptly increased for lower RVP values, showing
that at low RPV values apple tissue turned more resistant to rupture and to
deformation (more rigid) due to water removal and solid concentration (Fig. 1a).
An important change in the pattern of mechanical parameters was observed when
moisture content was nearly the monolayer water content (0.10 g water/g d.b.). An
increment in A was also observed, indicating an increase in the work needed for
rupture. On the other hand, D showed an opposite behavior to S; it increased as the
product turned from high moisture to intermediate moisture contents (i.e., the tissue
became more deformable), and then, at lower moisture contents, the resistance to
deformation increased (Fig. 1b). The volume of apple discs decreased until
314 S.M. Casim et al.
a b
S(N.m) Fmáx(N) D A (J/m)
16.00 9.00 6.00 6.00
2.50 5.00
14.00 8.00
m ( g water / g d.s.)
5.00
m ( g water / g d.s.)
5.00
7.00 2.00 4.00
12.00
6.00 4.00 4.00
10.00 1.50 3.00
5.00
8.00 3.00 3.00
4.00 1.00 2.00
6.00 2.00 2.00
3.00
4.00 2.00 Monolayer water content 0.50 1.00
1.00 1.00
2.00 1.00
0.00 0.00 0.00 0.00
0.00 0.00
0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00
RVP RVP
Fig. 1 Evolution of mechanical properties as a function of RVP during convective air drying of
apple tissue (left axis). Moisture content (dry basis) corresponding to the RVP (right axis). (a) Fmax
and S; (b) D and A. Error bar indicates the standard deviation
Sk (%) BI
100.00
35.00 6.00
80.00 5.00
30.00
m ( g water / g d.s.)
4.00
60.00
25.00
3.00
40.00
20.00 2.00
20.00 1.00
15.00
0.00
0.00 10.00
0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00
RVP
Fig. 2 Evolution of shrinkage (Sk) and browning index (BI) as a function of RVP during
convective air drying of apple tissue (left axis). Moisture content (dry basis) corresponding to
the RVP (right axis). Error bar indicates the standard deviation
reaching RPV 0.47 and then remained approximately constant (Fig. 2). Browning
index (BI) slightly, but significantly, increased along the process (Fig. 2).
Proton nuclear magnetic resonance (1H-NMR) studies indicated different water
populations with different distribution and mobility along the drying process. From
An Exhaustive Study on Physicochemical and Structural Changes During Drying. . . 315
a b
1200.00
6.00
T23 2.50 T2S
m ( g water / g d.s.)
1000.00 0.60
m (g water / g d.s.)
T22 5.00 T2L
T2 - CPMG (ms)
2.00
T2 - FID (ms)
800.00 T21 4.00 m
1.50 0.40
600.00
m
3.00
1.00
400.00 2.00
0.20
0.50
200.00 1.00
RVP RVP
Fig. 3 Evolution of relaxation times T2 as a function of RVP during convective air drying of apple
tissue (left axis). Moisture content (m) (dry basis) corresponding to the RVP (right axis). (a)
CPMG sequence; (b) FID method. Error bar indicates the standard deviation
CMPG sequence, three populations at high moisture contents were detected; mean-
while, at intermediate moisture values, only two populations were noticed, and at
low moisture contents, only one population was observed until moisture contents
closer to the monolayer value (Fig. 3a). Each T2i gradually diminished, indicating a
reduction in water mobility, although this decrease was more pronounced in T23
values, as related to free water (Fig. 3a). The MANOVA of CPMG parameters
indicated significant differences between all samples (Tables 1 and 2). On the other
hand, the FID method showed no significant changes in T2S value, but an important
decrease in T2L was observed as water was removed until the monolayer water
content was reached; then, T2L remained the same with no significant changes
(Fig. 3b). Nevertheless, at this water content, the relative proportion of the protons
that contributed to the solid signal was larger than the amount of protons contrib-
uting to the liquid signal, meaning that around the monolayer water content the
solid components of the tissue were those with dominant influence over the
physicochemical properties of the studied system (results not shown).
316 S.M. Casim et al.
For RVP
0.96, an inverse correlation could be seen in Fmax, S, and T2CPMG
behavior. In low moisture samples (RVP
0.47), because of the disappearance of
CPMG signal and the reduction in mobility of the water that interacts with the cell
wall polymers, the water would be less available to act as a structure plasticizer,
increasing both mechanical parameters.
As expected, thermal transition analysis indicated that glass transition temper-
ature increased as moisture content decreased. The final product at RVP 0.30 still
stayed in rubbery state (results not shown).
The fresh tissue showed turgid cells with parietal cytoplasm and important
intercellular spaces (Fig. 4a, d). At RVP 0.91, the cells seemed shrunken and
deformed, with less contact, some interrupted membranes and undulated cell walls
(Fig. 4b, e). At RVP 0.30, a much collapsed cellular arrangement was presented,
with high shrinkage and notable porosity. Some cell walls appeared disrupted with
low density (Fig. 4c, f).
Among the causes of the changes observed during drying, the following could
can be mentioned: (1) the concentration of solids due to water removal and the
increase on tissue viscosity; (2) the disorganization and loss of membrane integrity
and functionality and the subsequent change in the protein interaction with neigh-
boring water molecules and the diffusive interchange between compartments;
(3) the compactness and alteration of cell wall and, accordingly, the modification
of water interaction with cell wall macromolecules and mechanical properties, and
(4) the loss of turgor that decreased rigidity at high and intermediate moisture
ranges (Slade and Levine 1991; Ruan and Chen 1998; Lewicki and Lukaszuk 2000;
Lewicki and Pawlak 2003).
4 Conclusions
Fig. 4 Microscopic observations of apple tissue. (a–c) LM; (d–f) ESEM. (a) and (d) fresh tissue;
(b) and (e) dried apple at RVP 0.91; (c) and (f) dried apple at RVP 0.30
318 S.M. Casim et al.
References
Chapman D (1994) The role of water in biomembrane structure. J Food Eng 22:376–380
D’Ambrogio de Argüeso A (1986) Manual de Técnicas en Histologı́a Vegetal. Hemisferio Sur
S. A, Buenos Aires, Argentine
Krokida MK, Karathanos V, Maroulis Z (2000) Compression analysis of dehydrated agricultural
product. Drying Technol 18:395–408
Krokida MK, Maroulis Z, Marinos-Kouris D (1998) Viscoelastic behavior of dehydrated carrot
and potato. Drying Technol 16:687–703
Lewicki P, Pawlak G (2003) Effect of drying on microstructure of plant tissue. Drying Technol
21:657–683
Lewicki PP, Lukaszuk A (2000) Changes of rheological properties of apple tissue undergoing
convective drying. Drying Technol 18:707–722
Rao MA (2003) Phase transitions, food texture and structure. In: Brian M. McKenna (Ed). Texture
in food vol. 1: semi-solid foods. Woodhead Publishing Ltd., Cambridge, UK, pp 36–62
Ruan RR, Chen PL (1998) Water in foods and biological material: a nuclear magnetic resonance
approach. Technomic Publishing Company, Inc., Lancaster
Slade L, Levine HA (1991) A food polymers science approach to structure, property relations in
food systems. In: Slade L, Levine H (eds) Water transitions in Foods. Plenum, New York, pp
29–101
Snaar JEM, Van As H (1992) Probing water compartments and membrane permeability in plant
cells by 1H-NMR relaxation measurements. Biophys J 63:1654–1658
Defining High and Low Spray Drying
Temperatures for Aloe vera Gel
Abbreviations
1 Introduction
Spray drying is widely used in the food industry to obtain products in the form of
powder (Masters 1991; Filková and Mujumdar 1999; Nijdam and Langrish 2005;
Roustapour et al. 2006; Leon et al. 2010). Three types of modes of contact between
drying air and the food product can be found: parallel (co-current), counter-current,
or mixed (fountain). The latter is a combination of parallel and counter current flow
patterns, which implies that the already dry material comes into contact with drops
being sprayed, favoring the formation of agglomerates. In turn, the formation of
agglomerates raises the rehydration capacity of the final product (Masters 1991;
Vega et al. 2001; Peighambardoust et al. 2011). Although spray drying is a common
method used for drying thermolabile materials, there are few reports about the
dehydrating conditions used for Aloe vera industrial processing. This plant is widely
used in the pharmaceutical, cosmetological, and food areas because of the beneficial
effects it provides as a functional ingredient in food and cosmetic products (Simal
et al. 2000). However, it has been reported that when gel obtained from the Aloe vera
leaf is subjected to thermal processing, changes in its components and functional
properties, such as wettability, water retention, and oil adsorption capacity, among
others, can occur (Eshun and He 2004; Garcı́a et al. 2010). Aloe gel contains 98.5 %
water, and the remaining solid material consists of polysaccharides, fat, and water-
soluble vitamins, minerals, enzymes, phenolic compounds, and organic acids (Bozzi
et al. 2007; Eshun and He 2004; Hamman 2008). Several studies have focused on
assessing the physical, chemical and functional changes caused by Aloe vera gel
processing conditions, especially during dehydration through convective drying
(Simal et al. 2000; Femenia et al. 2003; Miranda et al. 2009; Garcı́a et al. 2010;
Gulia et al. 2010). However, to date there is limited data regarding the effect that
spray drying temperature has on the physical, calorimetric, and functional charac-
teristics of Aloe vera. Therefore, the objective of this study was to evaluate the effect
of spray drying temperatures on obtaining Aloe vera gel in powder form.
Aloe vera gel was obtained from the “Flor de Aloe” Company (Puebla, México),
which was filtered for processing in a Mobile Minor 2000 spray dryer (Gea-Niro,
Dinamarca) equipped with a dual fluid nozzle in a source-type arrangement and
without any addition of drying aid. The inlet (T1) and outlet (T2) drying-air
Defining High and Low Spray Drying Temperatures for Aloe vera Gel 321
temperatures used were 220, 225, and 250 C and 70, 80, and 90 C, respectively.
The feed stream varied between 1.12 and 1.85 L/h, depending on the desired outlet
air temperature when the pressure of the air spray was 1.53 kg cm2. Powder
samples were gathered at the cyclone base and stored in sealed containers for
further analysis, which was performed three times in every case.
The moisture content of Aloe vera gel without processing was determined using a
thermo gravimetric method, by means of a thermo balance MB 300 (Brainweigh,
USA). Moisture content of the powder (M ) was determined using the vacuum
drying oven method according to 920.151 (AOAC 1997).
Water activity (aw) was determined using Aqualab 4TE equipment (Decagon
Devices, USA) from 1 g (0.035 oz) of the powder while maintaining a temperature
of 24.5 0.1 C (Fang and Bhandari 2011).
aw 1 C 2 K G ð1 CG Þ
¼ þ G aw þ aw2 ð1Þ
M M o CG K G M o CG M o CG
Xs T gs þ kXw T gw
T gm ¼ ð2Þ
Xs þ kXw
where Tgm, Tgs, and Tgw are the mix, solids, and water glass transition temperatures,
respectively, expressed in ( C); Xs and Xm are the solids and water weight fractions,
respectively; and k is the Gordon and Taylor constant, which is related to the level
of interaction between the system components. The model parameters (k and Tgs) of
Eq. (2) were estimated using a nonlinear regression analysis where Tgw was
135 C (Sablani et al. 2007; Welti-Chanes et al. 1999).
The statistical significance of data was evaluated using the analysis of variance
(ANOVA) analysis with α ¼ 0.05. The statistical analysis was performed using the
statistical program MINITAB, version 15 (Minitab, Inc, USA).
To verify the existence of two or more groups and mathematically differentiat-
ing between them, the normal distribution density function was used. Normal
distribution is defined as a continuous variable distribution that is specified by
two parameters: mean and standard deviation (Eq. 3):
1
pffiffiffiffiffie2ð σ Þ
1 xμ 2
f ðx Þ ¼ ð3Þ
/ 2π
where μ is the mean and σ is the standard deviation of the analyzed continuous
variable. Densities were obtained for each of the relationships discussed: moisture
content, water activity, and glass transition-moisture content.
M (kg of water/kg of dry solid) of gel powder showed values between 0.01 and 0.03
(kg of water/kg dry solids), with no significant difference found at a level of
α ¼ 0.05 for any of the obtained products under the different drying-air tempera-
tures. These values are similar to those previously obtained for various powder
products at different drying conditions (mucilage nopal, nopal juice, and acai pulp)
(Tono et al. 2008; Leon et al. 2010). aw ranged from 0.18 to 0.30, without a clear
relationship between inlet/outlet drying-air temperature effect and aw. Similar
values of aw have been reported for other powder products, such as watermelon
(0.20–0.29) (Quek et al. 2007), milk (0.2–0.6) (Laroche et al. 2005), and bayberry
Defining High and Low Spray Drying Temperatures for Aloe vera Gel 323
Fig. 1 Moisture content (M ) and water activity (aw) in relation to samples obtained at different
inlet air temperatures, and moisture adsorption behavior (GAB model) of Aloe powders obtained
by spray drying
(0.1–0.3) (Fang and Bhandari 2011). For all Aloe vera gel powder samples, the
existing relation between M and aw is observed in Fig. 1, and the existence of two
different data groups is also identified in this figure. In the first group, which was
called the low temperature group, samples were assessed using inlet/outlet temper-
atures of 200/70, 200/80, 200/90, 225/70, and 250/70 C. These conditions corre-
spond to all cases in which the lower inlet and three outlet temperatures were used,
and also to the 225 and 250 C temperature conditions, in which the lower outlet
temperature was used. The second group, denoted as the high temperature group,
had the samples with the lowest values of M and aw, and corresponded to the
highest inlet/outlet drying-air temperatures (225/80, 225/90, 250/80, and
250/90 C). When inlet temperatures of 200 C and any outlet temperature were
used, the M and aw values of the products were higher; at 225 and 250 C inlet
temperatures and 70 C outlet temperature, materials were located within the low
temperature drying groups. All conditions were grouped as high temperature.
Applying the GAB model to M and aw, the results showed that they can also be
grouped into the low and the high temperature groups (Fig. 1). In the first case, it is
possible that drying temperatures causes minor heat damage to the material;
324 I.N. Garcı́a-Luna et al.
Fig. 2 Relationship between glass transition temperature (Tg) and moisture content (M ) for Aloe
gel powders obtained from different drying conditions
therefore, they could belong to the same isotherm, which may indicate that under
these drying conditions, the original chemical composition of the material does not
change significantly. The opposite could occur in the high temperature group,
where the material may present different characteristics from those obtained at
low temperatures. Therefore, they would be associated with different hygroscopic
properties.
Fig. 3 Glass transition temperature (Tg) variation of Aloe gel powder obtained by spray drying
according to Gordon and Taylor model
relationship of Tg and M for the different powders obtained; even though a direct
relationship between these properties could not be observed, it was possible to
classify all obtained powders into groups identified. It has been suggested that Aloe
gel powder remains in a vitreous state at room temperature, regardless of the
drying-air temperature. Moreover, at higher temperatures and moisture levels, the
plasticization effect can increases the loss of powder quality. The results experi-
mentally obtained (M, aw, and Tg) were analyzed using the Gordon and Taylor
model (Eq. 2) to determine if exists any correlation between experimental values of
water activity and glass transition. The results for all powders obtained at the
different drying-air temperatures showed relatively low values of aw (0.18–0.33).
The interval of water activity at which the samples are located is a narrow section of
the values of water activity (0.1–0.9) which should cover the samples, as indicated
in the analysis of the model of Gordon and Taylor. In this case, separation of the
samples in the high and low temperature groups was also verified (Fig. 3).
4 Conclusions
In this study, the Aloe vera gel (without drying adjuvant) can be spray dried using
an inlet drying-air temperature greater than or equal to 200 C and the outlet drying-
air temperatures from 70 to 90 C, in a fountain type arrangement. The drying-air
326 I.N. Garcı́a-Luna et al.
Acknowledgements The authors acknowledge the financial support of research projects: SIP
2011745, 20121173, 20111167, 20121754 and CONACyT 84287, as well as the PIFI-program,
COFAA-IPN, Tecnol ogico de Monterrey and CONACyT for study grants.
References
Abbreviations
1 Introduction
Fresh fruits, no matter where they come from, are perishable products due to their
high moisture content. Some fruits are protected with an external thick peel or film
that allows them to remain fresh for a long time if they are appropriately handled.
However, other fruits, even though they have a thick peel, have very short growing
periods. This is the case of prickly pear, mamey, litchi, pitaya or pitahaya, black and
yellow sapote, and Mexican plum, just to mention a few. Pitaya or pitahaya and
Mexican plum possess a delicate peel that may allow the fruit to remain fresh for a
very short time, after which the peel can become dry or be broken due to dryness or
overripening. All of these fruits grow in Mexico because of the microclimates that
exist throughout the country. Therefore, to take advantage of these fruits that last
for a short time or because of overproduction, it is necessary to apply or use
conventional and/or new processing technologies to deliver new or processed
fruit products to domestic and overseas consumers.
One way to increase fruit stability is to remove water by using spray or freeze-
drying processes to obtain fruit powders. Fruit powders can be used to prepare
blends with other food powders or to prepare other high-moisture-content processed
fruit products such as jellies, jams, juices, or nectars. Transportation of fruit
powders may reduce cost since water is no longer transported, while approaches
such as refrigeration or freezing of fruit products can result in greater expense for
transportation. Therefore, obtaining powders may be the best alternative for having
fresh fruit products available anytime and anywhere. However, powders may
undergo physical and chemical changes if they are not handled appropriately.
Fruit powders can show caking due to moisture sorption or crystallize due to
modification of the fruit component. Consequently, powders’ physical properties
may change and affect their functional properties, which in turn may affect the
preparation of other fruit products.
Various mathematical methods describe the relationship of moisture content and
water activity of dehydrated foods. The moisture sorption data of fruit powders can
be analyzed during a number of mathematical models to determine the critical
moisture sorption properties. These moisture sorption properties are useful for
predicting moisture content at different relative humidities. Also, the critical mois-
ture sorption properties can provide an idea regarding storage conditions at differ-
ent temperatures and moisture conditions when storing foods in chambers, large
packages, or small containers, as well as the packaging materials required for each
type of food powder. The Guggenheim-Anderson-De Boer (GAB) (van den Berg
and Bruin 1981), Brunauer-Emmett-Teller (BET) (Brunauer et al. 1938), and
Oswin (Oswin 1946), among other equations, are mathematical models frequently
used for fitting of moisture content vs. water activity experimental data or to obtain
the critical moisture sorption properties of foods. The objective of this research was
to obtain moisture sorption isotherms and model the experimental data of Mexican
plum powder.
Mexican Plum (Spondias purpurea L.) Moisture Sorption Properties 331
Unripe Mexican plums (Spondias purpurea L.) were purchased at the Cholula,
Puebla, marketplace in January (Fig. 1). Plums were stored at room temperature
until achieving edible characteristics; then a purée was obtained using a pulper
machine. Purée was analyzed for pH, total soluble solids (TSS), titratable acidity,
and moisture content.
2.2 Freeze-Drying
Plum purée was poured into Petri dishes and then placed on freeze-dryer
(Labconco) chamber trays to be frozen at 18 C during the night. The frozen
pulp was then freeze-dried at 25 C and 10 μmHg for 24–48 h.
2.4 Modeling
The moisture sorption data were fitted using the GAB, BET, and Oswin models to
determine the critical moisture sorption properties of the fruit powder.
Brunauer-Emmett-Teller (BET) model (Brunauer et al. 1938) equation is useful
within the range 0.11–0.45 of water activity. A straight line is obtained from the
relationship shown below:
aw C1 1
¼ aw þ
ð1 aw ÞW WmC WmC
mo CKaw
m¼
ð1 Kaw Þð1 Kaw þ CKaw Þ
Oswin model (Oswin 1946) can be useful for the entire water activity range:
n
aw
W¼a
ð1 a w Þ
where awpred and awexp are the predicted and experimental water activities,
respectively.
The ripe Mexican plums had an orange-/yellow-colored peel. This type of fruit has
a large stone, weighing as much as 25 2 % of the whole fruit. The amounts of
purée and peel were 60.6 3 and 14.0 1 %, respectively. This type of plum is a
nonacid fruit (pH 4.03; acidity of 0.36 0.01 % citric acid), with very high content
of total soluble solids (21.13 0.12 %) and moisture content of 77.93 0.43 %.
Sorption isotherms of type II (BET classification) were obtained for the Mexican
plum powder at the three selected temperatures (Fig. 2). This type of behavior is
common in fruits with high sugar content. In general, the sugar content in plum
consists of fructose, glucose, and sucrose. These soluble solids are the main
chemical products that contribute to suppress the water activity of the plum powder.
BET Model (0–0.45 aw). Table 1 shows the BET moisture sorption characteristics of
Mexican plum powder. The monolayer moisture content decreases as temperature
increases; this value was not higher than 5.39 g H2O/100 g d.s. presented at 21 C.
Low monolayer moistures are due to the high content of sugars (96 %) in powders.
334 J.A. Guerrero-Beltrán et al.
2.5
21°C
W (g H2O/g d.s.) 32°C
2 40°C
1.5
0.5
0
0 0.2 0.4 0.6 0.8 1
Water activity
Table 1 Moisture sorption characteristics of Spondias purpurea powder using the BETa model
Temp. m (gH2O/g b (g d.s./g Wm (g H2O/100 g
( C) d.s.) H2O) R2 C d.s.) awm
21 18.84 0.301 0.993 61.7 5.39 –
32 24.09 0.024 0.962 1019.7 4.15 –
40 23.72 0.233 0.964 102.8 4.17 0.0897
a
aw 0.11–0.45; m, slope; b, intercept; R2, correlation coefficient; C is a constant related with the net
sorption heat; Wm is the monolayer moisture content; awm is the monolayer water activity
Table 2 Coefficients of the second-order representation of the GAB model for moisture sorption
characteristics of Spondias purpurea powder
Temp. ( C) α β γ R2
21 16.430 16.027 0.0727 0.945
32 21.225 19.643 0.4621 0.866
40 23.968 22.089 0.4630 0.884
α, β, and γ are constants; R2 is the correlation coefficient
GAB Model. Table 2 shows the second-order polynomial coefficients of the GAB
model used to determine moisture sorption properties of Spondias purpurea. GAB
monolayer moisture contents (MMC) were calculated; those values decreased as
temperature increased. MMC are similar to those obtained using the BET model
(Table 1). Average standard errors are presented in Table 3; the lower standard error
(7.87 %) was obtained from the moisture sorption data obtained at 40 C. However,
the GAB model fit experimental data in a better way in the water activity range of
0–0.7.
Oswin Model. Table 4 depicts the Oswin parameters for the moisture sorption
properties of Mexican plum powder (Fig. 2). The correlation coefficients obtained
with this model were higher than those obtained with the GAB model (Table 3).
Consequently, the average standard errors (ASE) were smaller than that of the GAB
Mexican Plum (Spondias purpurea L.) Moisture Sorption Properties 335
Table 3 Moisture sorption characteristics of Spondias purpurea powder using the GABa model
Temp. ( C) K mo (g H2O/100 g d.s.) C R2 ASEb (%)
21 1.02 6.18 218.04 0.945 33.76
32 1.054 4.85 42.31 0.866 38.21
40 1.061 4.33 46.94 0.884 7.87
a
aw up to 0.97; K, factor of correction for multilayer properties; mo, monolayer moisture content;
C, constant of Guggenheim; R2, correlation coefficient; baw up to 0.90
model. These ASE values are still large to adequately fit the experimental sorption
moisture data. However, this fitting is better than that obtained with the GAB
model, since it was obtained from the whole water activity range (0–0.970).
4 Conclusions
Type II isotherms were obtained using the gravimetric static method with discon-
tinuous registration of weight changes for Mexican plum powders. The monolayer
moisture content, obtained using the BET or GAB model, was in the range 6.18–
4.17 g H2O/100 g d.s. The average standard error, obtained using the GAB or Oswin
fitting, was in the range 7.87–33.76. However, the Oswin model better fit the
moisture sorption experimental data of Mexican plum powders.
References
Abbreviations
1 Introduction
2.1 Materials
Dried chile (Capsicum annuum L. grossum Sendt) was used for this work. Edible
vegetable corn oil was purchased from a local market in Mexico City, Mexico. The
wall materials used for microencapsulation were gum arabic, food grade E414
(Distribuidora Quı́mica LEFE S.A. de C.V., Mexico City, Mexico), and maltodex-
trin 20 DE (CPI Ingredientes S.A. de C.V., State of Mexico, Mexico). Potassium
persulfate (K2S2O8) and 2,2-azinobis-(3-ethylbenzothiazoline-6-sulfonic acid)
(ABTS) radical were purchased from Sigma Chemical Company, St. Louis, MO,
USA. The water used in the experiments was doubly distilled.
Antioxidant Activity of Microencapsulated Capsicum annuum Oily. . . 339
Dried chiles were manually cleaned by removing all foreign matter and then cut
into small squares. Next, 50 g of chile was mixed with 250 mL of corn oil heated to
70 C at 5,500 rpm and extracted in a Thermomix (TM31, Vorwerk, Espa~na) until a
homogeneous mixture was obtained. The extract of chile was filtered in order to
separate solids and to obtain an oily extract of chile (OEC).
The antioxidant activity of Capsicum extract was evaluated by studying its ability
to bleach the 2,2-azinobis-(3-ethylbenzothiazoline-6-sulfonic acid) radical
(ABTS+). The ABTS+ was prepared by the reaction of ABTS solution
(3.84 102 g) with 6.6 103 g potassium persulfate (K2S2O8) (both solutions
were prepared in 10 mL of bi-distilled water). The solution was stored for 12–16 h
in the dark and at 10 C (solution A). 0.5 mL of solution A was put in anhydrous
ethanol solution, until it completed 50 mL, and it was adjusted to 0.7 0.02
absorbance in a spectrophotometer (Genesys 10UV Scanning Thermo Spectronic)
at 734 nm to determine antioxidant activity of the extracts. 200 μl oily extract was
diluted with 800 μL acetone (Fermont, México) from this extract; 10 μl was taken to
react with 990 μl of ABTS+ radical. The absorbance at 734 nm was taken at 20 C,
exactly 1 min after initial mixing. The percent of inhibition was calculated as
follows:
ðAst0 Ast f Þ
%Inhibition ¼ h i 100 ð1Þ
0 Adt f
ðAst0 Þ AdtAdt 0
where
Ast0 and Astf: Absorbance of sample at initial and final times
Adt0 and Adtf: Absorbance of dissolvent at initial and final times
A standard curve was constructed using different concentrations of Trolox
(6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid; Sigma Chemical Com-
pany, St. Louis, MO, USA). All measurements were performed in triplicate, and the
average standard deviation (SD) was reported. In all cases, the effect of acetone
and initial values of antioxidant activity of oil were determined.
340 A.Y. Guadarrama-Lezama et al.
Oil-in-water emulsion was prepared (O/W) using, as wall materials, gum arabic
(GA) and maltodextrin (MDX) 20 DE (dextrose equivalent) in a 1:1 ratio. The
biopolymers were dissolved in warm water (35 C) with manual stirring to facilitate
the process of dissolution. The solution was stored for 24 h in order to achieve
complete rehydration of polymers. Oily extract of Capsicum (2 %) was added drop
by drop to the water polymeric solution, while homogenization in a Polytron
(Kinematica PT200D Type Luzern-Switzerland) was carried out by 3 min. In the
preparation of emulsions, an ice bath was used to maintain the temperature below
30 C (Koberstein-Hadja and Dickinson 1996) and avoid instability in the system.
Subsequently, the emulsion was fed by a peristaltic pump (Watson-Marlow 520S)
with a feed flow of 16 mL/min to a spray dryer (Mobile Minor ™ 2000, Niro,
Denmark). A two-fluid pneumatic nozzle atomizer was used in fountain flow. The
inlet and outlet drying air temperatures used for the microencapsulation were
200 2 C and 90 2 C, respectively.
The water activity of the microcapsules was determined using an AquaLab (Deca-
gon Devices, Model 4 TE, USA) at 25 C. Moisture content of the microcapsules
was determined gravimetrically using official AOAC method (AOAC 1995).
Antioxidant activity in oily extract was 32.30 2.0 μmol Trolox/mL. For this work,
nonencapsulated extract is defined as that portion of lipid material that remains on
the surface of microcapsules. The antioxidant activity of nonencapsulated chile oily
extract was less than 10 μmol Trolox/mL, while the antioxidant activity of oily
extract contained in 2 g of microcapsules was 29.45 2.1 μmol Trolox/mL. In all
cases, the content of oily extract encapsulated was higher than the extract
nonencapsulated, giving a percent of encapsulated oil of 60 %, which may be
improved by optimizing the homogenization process as well as modifying the
drying conditions during the microencapsulation process.
The results showed that the microencapsulation process by spray drying does not
degrade the carotenoids present in the oily extract. The antioxidant activity was
preserved in 91.17 % of the oily extract, probably because the evaporation of water
is so fast that the sprayed droplet is kept cold until reaching the state of drying due
to absorption of heat in the vaporization of the liquid (Gharsallaoui et al. 2007). So
the extract encapsulated in biopolymers has no direct contact with the drying air and
the contact is very short; therefore, the temperature that would be reached by the
emulsion droplet corresponds to the wet bulb temperature (~40 C). The presence of
nonencapsulated oil in the microcapsules may be attributed to the use of inlet and
outlet temperatures above 210/90 C during the spray-drying process; in both
342 A.Y. Guadarrama-Lezama et al.
Both water activity and moisture content are considered as quality parameters that
characterize the powdered food. These parameters are important to help prevent
spoilage reactions. The effect of water activity in foods and their interaction with
components have been studied. Some authors have considered the value of the
monolayer (0.2–0.3 water activity) as the range of maximum stability of dehydrated
products, and this effect is related to the reduced mobility of water in the mono-
layer, in which water is tightly bound and cannot act as a reaction medium. It is
considered that at these values of water activity, food powders are stable. The
microcapsules containing oily extract showed a 0.22 water activity and 5 % of
moisture content; therefore, at this value, degradation reactions are not present,
favoring the preservation of carotenoids contained in capsicum extract.
The moisture content on physical changes of the biopolymer matrix used to
encapsulate the oily extract may affect the distribution of oil, allowing the acces-
sibility of oxygen to the oil phase, accelerating the degradation process. It is
therefore important to keep the moisture content at a low level as well as the
water activity of the microencapsulated oil to avoid any degradation process.
By using scanning electron microscopy, it was possible to see that the capsules
showed an irregular geometry, typical of spray-dried materials; large and deep
dents can be seen (Fig. 1a), while other dents are small and relatively shallow
(Fig. 1b), small particles that can occupy spaces between large particles are also
present (Fig. 1c), which may affect rehydration or flow properties. Presence of
fractures in the capsule was not observed, and this fact is important for preserving
the antioxidant activity of the capsicum extract.
Capsules formed have an elastic and enveloping external film, which is due to
the presence of protein contained in the structure of gum arabic (Yadav et al. 2007),
helping to avoid the presence of superficial fractures. The oxidative process also is
avoided using maltodextrin as wall material (Buffo and Reineccius 2000).
Non-homogeneity in the surface of the particles may be undesirable in terms of
retention of microencapsulated extract; however, it is noteworthy that the wetting
and rehydration times are closely related to the external topography of the
Antioxidant Activity of Microencapsulated Capsicum annuum Oily. . . 343
microcapsules. Water penetration is better in areas that have defects, which facil-
itates its adsorption at specific sites (Pérez-Alonso et al. 2009).
Any change in the physical state of a product has the potential to affect the
characteristics of the ingredients of the product. Therefore, morphology is an
important feature to be considered in microencapsulation processes, as it is related
to the moisture content, temperature, and water activity in which they are stored and
with preservation of the antioxidant activity of the Capsicum extract.
The above result indicates that the carotenoids from Capsicum could be an
alternative source for use as natural antioxidants and can be used in the formulation
of functional food products.
4 Conclusions
Chile oily extract with antioxidant activity was encapsulated, maintaining its
antioxidant capacity. The value of water activity (0.20) of the encapsulated extract
is considered to be in a high stability range for dehydrated products. Chile can be
encapsulated as a natural source of antioxidants and may be incorporated in
functional foods.
Acknowledgements The authors thank the financial support for this research from the National
Polytechnic Institute through SIP project 20121026 and 20121754 and the PIFI program as well as
from the Institute of Science and Technology of Mexico City’s Government (ICYTDF), through
the project PICS08-15.
References
Abbreviations
M.F. Fabela-Mor on
Departamento de Graduados e Investigacion en Alimentos, Escuela Nacional de Ciencias
Biol
ogicas, Instituto Politécnico Nacional, Carpio y Plan de Ayala s/n. Col. Santo Tomás.
Deleg Miguel Hidalgo, 11340 México D.F., Mexico
Centro de Desarrollo de Productos Bioticos, Instituto Politécnico Nacional, Carretera
Yautepec-Jojutla, Km. 6, calle CEPROBI No. 8, Col. San Isidro, Apartado Postal 24,
62731 Yautepec, Morelos, Mexico
J. Porras-Saavedra • L. Alamilla-Beltrán (*)
Departamento de Graduados e Investigacion en Alimentos, Escuela Nacional de Ciencias
Biologicas, Instituto Politécnico Nacional, Carpio y Plan de Ayala s/n. Col. Santo Tomás.
Deleg Miguel Hidalgo, 11340 México D.F., Mexico
e-mail: [email protected]
R. Martı́nez-Velarde • A. Jiménez-Aparicio • M.L. Arenas-Ocampo
Centro de Desarrollo de Productos Bioticos, Instituto Politécnico Nacional, Carretera
Yautepec-Jojutla, Km. 6, calle CEPROBI No. 8, Col. San Isidro, Apartado Postal 24,
62731 Yautepec, Morelos, Mexico
1 Introduction
The raw material, agave extract (Agave angustifolia Haw) was obtained from
Morelos State, Mexico. The extract was dehydrated by spray drying to obtain
agave powders. Powder was prepared using a two-fluid nozzle as atomizer, and a
Physical–Chemical Properties and Microstructure of Agave Powders. . . 347
laboratory spray dryer (Mobile Minor, Niro, Denmark). Two inlet/outlet drying air
temperatures of 160/70 C (Sample C1) and 180/80 C (Sample C2) were tested
(Lingyun et al. 2007; Gharsallaoui et al. 2007; Leon-Martı́nez et al. 2010).
To obtain agave powders by spray drying process, two phases were identified. In the
first phase, hot air transfers latent heat to evaporate the surface water of the product,
with an interchange of heat and humidity between the sprayed droplets and drying
air. This phase is a function of extreme conditions of temperature, air humidity,
feed concentration, and pressure. In the second phase, spray drying is denominated
as restrictive drying, because the shift of water depends on the physical and
chemical nature of the solid, temperature, and moisture content (Vehring
et al. 2007; Goula and Adamopoulos 2010).
For processing of agave powders, a relationship was observed between inlet and
outlet temperature (Tinlet/Toutlet) temperatures regarding moisture content of the
powders obtained. In the case of increasing Tinlet/Toutlet temperatures, moisture
content decreases as well as water activity (Table 1). In this sense, is important to
consider the importance of drying conditions for the appropriate dehydration of
agave extract to obtain powder, avoiding the presence of solids encrusted on the
wall drying chamber; therefore, inlet and outlet air temperatures are usually
the main variables controlling water content, bulk properties, and morphology of
the powder (Masters 1985; Thankitsunthorn et al. 2009).
348 M.F. Fabela-Mor
on et al.
Table 1 Moisture content and water activity of Agave angustifolia Haw powders obtained at
different air-drying temperatures
Samples Tin/out ( C) Moisture content (% wb) Water activity (Aw)
C1 160/70 5.5 0.14a 0.32 0.21a
C2 180/80 4.8 0.06a 0.22 0.03a
a
Different lower case letters in the same column for each analysis indicate significant differences
among treatments (P < 0.05).
Fig. 1 Images acquired by SEM of Agave angustifolia Haw powders: (1a) Sample C1, observed
at 1,000; (2a) Sample C1, observed at 3,000; (1b) Sample C2, observed at 1,000; (2b)
Sample C2, observed at 3,000X
both cases, particles were structured by a film or crust on the surface and evident
shrinkage. Theoretically, during crust formation, the temperature of the product
usually remains below the boiling point of water (Masters 1985; Alamilla-Beltrán
et al. 2005; Barbosa-Cánovas et al. 2005; Vehring et al. 2007; Goula and
Adamopoulos 2010).
Bulk density and true density vary in function of moisture content and the Tinlet/
Toutlet. At air-drying temperature of 180/80 C, bulk density is lower than the agave
powder obtained at the condition of 160/70 C; the same relation was observed for
water activity where powder tended to be cohesive at moisture content of 5.5 %
(wb). Density determinations of food powders are fundamental for the establish-
ment of storage, processing, and distribution conditions, so this merits particular
consideration. In this case, the bulk density decreases while increasing air-drying
temperature, maybe because this characteristic is related to the particle structure
developed during the drying process, while true density and water activity increased
when Tinlet/Toutlet increased, due to the space between particles and the moisture
content and water activity associated with increase of cohesiveness by interactions
between links of particle (Gallo et al. 2010; Goula and Adamopoulos 2010). Results
show that all of the operating parameters significantly affect the physical properties
of powder.
With increasing inlet/outlet air temperatures, the wettability of powder is
reduced and its rehydration time is augmented; dispersibility was the same value
at 97 % in both cases, and water activity values were 0.18 and 0.22. Agave powder
presents excellent behavior as an instant and rehydratable powder due to low
rehydration time (less than 60 min) and its dispersibility, comparable with milk
powder, sugars, fruit juice powders, etc. (Barbosa-Cánovas et al. 2005;
Thankitsunthorn et al. 2009).
High values of wettability and rehydration of agave powders (Table 2) are
related to the presence of components such as FOS and sugar in the powder,
which cause shorter wetting times on powders. The rehydration process involves
interactions between the powder and water, which is mainly delineated by the
adsorption mechanisms of the compounds present on the powder and its affinity
with water. In addition, changes in osmotic pressure because of the presence of
solids content due to penetration of dissolvent can cause swelling and relaxation of
material, which affects interaction between particle to particle and particle to
dissolvent; where water provokes the ordering of polymeric chains by steric effect,
water adsorption continues, and at the same time there is a dissolution process
(Barbosa-Cánovas et al. 2005; Gaiani et al. 2009).
Particles of agave powder obtained for both temperatures presented morphology
with rough surfaces and some smooth surfaces with a few dents. This topography is
a common shape developed during the spray drying process, making it possible to
modify the particle morphology depending on the spray drying conditions.
In agave powder, the shape and texture of particles had an effect on bulk density,
with a tendency to compaction. Particles obtained presented shrinkage and a rough
surface, because during the drying process phenomena such as expansion and
shrinking of particles may exist. Fractal dimension determination is a useful tool
350 M.F. Fabela-Mor
on et al.
4 Conclusions
Acknowledgments The authors wish to thank the financial support of Instituto Politécnico
Nacional through SIP project 20121754 as well as program PIFI-IPN and COFAA-I.P.N,
CONACyT, projects 80576, 84287; and scholarships for the PhD studies of Miriam Fabiola Fabela
and Josefina Porras Saavedra. We also acknowledge the experimental support of CNMN-IPN in
conducting this work.
Physical–Chemical Properties and Microstructure of Agave Powders. . . 351
References
Abbreviations
A Alginate
A¥ Maximum invertase activity corresponding to the active invertase
present in the beads
At Invertase activity at time t
A-TCh Alginate-trehalose-chitosan beads
Ch Chitosan
FD Freeze-drying
k Constant related to geometric and structural characteristics to the
macromolecular network
n Release mechanism type
SEM Scanning electron microscopy
T α-α-Trehalose dehydrate
t Time
VD Vacuum drying
1 Introduction
The stability and controlled release of certain active substances (such as flavors,
drugs, enzymes, essential oils, etc.) can be achieved through encapsulation.
Ionically cross-linked hydrogels such as alginate- or alginate-chitosan-CaCl2
beads were proposed as suitable materials for encapsulation systems (Deladino
et al. 2008; Han et al. 2008). Alginate beads coated with chitosan were used for
encapsulation and release of different proteins (Coppi and Iannuccelli 2009; Zhou
et al. 2010). Direct interaction between them forms beads with improved mechan-
ical properties associated with reinforcement of bead structure due to chitosan
binding to free alginate sites by cooperative ionic bounds (Deladino et al. 2008).
Dehydration and freezing were generally performed to enhance stability of labile
proteins. These processes were usually carried out in the presence of additives such
as saccharides; trehalose is one of the most used among those additives. Even
though several enzymes have been stabilized through immobilization in glassy
sugar matrices (Lee et al. 2006; Santagapita et al. 2008), they failed to provide a
controlled release.
Trehalose was fundamental in achieving adequate invertase protection during
freezing, drying (freeze-, vacuum-, or air-drying) and mild thermal treatment
(at 50 C, between 6 and 22 C above their Tg), but was not effective in preventing
enzyme functionality damage during bead generation (Santagapita et al. 2011,
2012). On the other hand, inclusion of chitosan in the bead walls prevented enzyme
activity losses during bead generation, without showing detrimental effects on its
conservation during drying or thermal treatment.
Our goal was to study the effect of trehalose and chitosan addition on the release
mechanism of invertase encapsulated in wet alginate-calcium beads and the effect
of drying method.
2.1 Materials
Bead size and shape evaluation was carried out by analyzing digital images
captured by a digital camera installed on a microscope (magnification at 7).
Pictures were analyzed using Image J software. Feret’s diameter (longest distance
between any two points along bead boundary) and circularity (how closely particle
shape resembles a circle) were analyzed for at least 50 (wet systems) or 40 beads
(dried systems).
Enzymatic activity determination was chosen to follow the delivery of the enzyme
due to low enzyme concentration (0.5–0.4 μg protein/bead) to quantify protein
content. Eight beads were placed in microcentrifuge tubes (in duplicate) with
120 μL of cold (4 C) sodium acetate buffer 50 mM pH 4.5 and were gently shaken.
Then, after appropriate times, enzymatic activity was determined in the beads
356 P.R. Santagapita et al.
(Subheading 2.6). The release curves were modeled by Peppas equation (Peppas
1985) using GraphPad Prism v5 software.
Wet alginate beads were dried by freeze-drying (FD) and by vacuum-drying (VD).
A digital image processing technique was used to obtain bead size (determined as
Feret’s diameter) and circularity after optical microscopy analysis. The relative
changes of these parameters after dehydration, in relation to those of the wet beads,
are shown in Fig. 1. Wet beads had a diameter between 1.3 and 1.6 mm, and
circularity was around 0.84 0.04 for the three employed formulations. After
drying, the bead size was considerably reduced, mainly during VD (Fig. 1). Tre-
halose greatly reduced the loss in size for the FD samples. Upon drying, the beads
lost circularity, particularly during FD. The presence of trehalose seemed to
preserve circularity, especially during VD. There were no further differences in
both size and circularity due to chitosan addition in alginate-trehalose-chitosan
beads (A-TCh) with respect to trehalose inclusion (alginate-trehalose beads (A-T)).
Drying method affected the size and shape of the beads more than the composition
change.
As previously observed (Santagapita et al. 2012), scanning electron microscopy
(SEM) images obtained for dried beads indicated that bead surface characteristics
depended both on the type of drying and on composition (presence of trehalose
and/or chitosan). The surface of FD alginate beads was smoother than that of VD
beads, and average pore sizes were 30 nm and 100 nm, respectively. A smoother
surface was observed in trehalose presence (A-T and A-TCh beads). The alginate
beads with chitosan (A-TCh) showed a rougher surface than alginate (A) or A-T
beads, as a consequence of alginate-chitosan interaction (Deladino et al. 2008).
Stabilization and Controlled Release of Invertase Through the Addition. . . 357
Fig. 1 Percentage reduction of Feret’s diameter (a) and circularity (b) of freeze-dried (FD) and
vacuum-dried (VD) alginate beads in relation to corresponding wet beads. Standard deviation
values are included. A alginate; T trehalose; Ch chitosan
Fig. 2 Enzyme release from wet (a) and dried beads (b) as a function of incubation time in acetate
buffer pH 4.5 at 5 C. Full lines show the fitting obtained by using Peppas equation (Eq. (1));
symbols represent experimental values. A alginate; T trehalose; Ch chitosan
Table 1 Release exponent obtained by Peppas equation (Eq. (1)), transport mechanism, and
relative rates between diffusion and polymer chains relaxation from swellable and non-swellable
spherical systems
Release exponent Diffusion rate vs. polymer chains
(n) Transport mechanism relaxation rate
0.43 Fickian diffusion Diff. rate < relax. rate
0.43 < n < 0.85a Anomalous Diff. rate relax. rate
0.43 < n < 1b Anomalous Diff. rate relax. rate
0.85a Transport case II Diff. rate > relax. rate
1b Transport case II (zero-order Diff. rate > relax. rate
kinetic)
a
Polymers that swell when rehydrated
b
Polymers that do not swell when rehydrated
Shu et al. 2010). Table 1 indicates the limits of n and the transport mechanism
(according to Alfrey et al. 1966) for spherical systems (Pothakamury and Barbosa-
Cánovas 1995). The n higher limit always corresponds to a transport limited by the
rate of the polymer chains relaxation, independent of whether the polymers swell or
do not swell. The n and k parameters obtained by applying (1) to the data shown in
Fig. 2a are shown in Table 2, with R2 higher than 0.96. The enzyme release from A
beads showed an n value of 0.43. Therefore, the main transport mechanism was
Fickian and limited by diffusion. The presence of trehalose and chitosan slightly
influenced the release mechanism by changing the n value, which was still close to
the 0.43 limit for Fickian diffusion, but indicated a certain influence of the polymer
chains relaxation in the transport mechanism. It is not indicated that a comparison
of k values when n values were different (A vs. A-T or A-TCh) should be
performed. However, both bead systems containing trehalose showed similar
n and k values, which reveals that there were no structural differences between
them.
Stabilization and Controlled Release of Invertase Through the Addition. . . 359
Table 2 n and k parameters obtained from Peppas equation and transport type for invertase
release from wet and dried beads in acetate buffer (50 mM) at pH 4.5
pH 4.5
n k 103 (sn) Type of transport
A, wet beads 0.43 0.05 7.6 0.5 Fickian diffusion
A-T, wet beads 0.50 0.05 5.0 0.3 Anomalous
A-TCh, wet beads 0.49 0.05 5.0 0.7 Anomalous
A-T, vacuum-dried 0.31 0.02 34.0 0.8 Fickian diffusion
A-T, freeze-dried 0.30 0.06 9.0 0.5 Fickian diffusion
A-TCh, freeze-dried 0.33 0.01 14.5 0.1 Fickian diffusion
A alginate; T trehalose; Ch chitosan
Invertase release was much faster in VD beads than in those in FD, as shown in
Fig. 2b. This result is in agreement with porous size calculated by analyzing SEM
images. Table 2 shows the parameters obtained by fitting the experimental values
with (1), with R2 higher than 0.997. n values were <0.43 for both types of A-T dried
beads. Even though this situation (n < 0.43) was not contemplated in Table 1,
according to Peppas (1985), the use of (1) to analyze systems that have pores
leads to n values <0.43, but could still be treated as Fickian diffusion. A combined
mechanism of diffusion partially through a swollen matrix and partially through
water-filled pores shifts the release exponent n toward smaller values. Other authors
have also related this behavior to Fickian diffusion of a drug from alginate-chitosan
particles (Coppi and Iannuccelli 2009). A great difference was found in the
k parameter (Table 2) for both dried A-T beads. Since both types of dried beads
were spherical (i.e., having the same geometry), the k parameter shows a great
difference in the structure between them, according to SEM images. The inclusion
of chitosan in the FD A-T beads increased enzyme release rate with respect to A-T
beads, but the type of release mechanism was still the same, thus maintaining the
same profile for the release curve. The k value was higher in relation to A-T beads,
indicating that the difference in release rate was related to structural factors.
4 Conclusions
Acknowledgments The authors acknowledge the financial support of ANPCYT (PICT 0928),
CONICET (PIP 100846), and UBA (Project UBACyT 20020100100397).
References
Alfrey E, Gurnee EF, Lloyd WG (1966) Diffusion in glassy polymers. J Polym Sci C 12:249–261
Bernfeld P (1955) Amylases, α and β. Methods Enzymol 1:149–158
Coppi G, Iannuccelli V (2009) Alginate/chitosan microparticles for tamoxifen delivery to the
lymphatic system. Int J Pharm 367:127–132
Deladino L, Anbinder PS, Navarro AS, Martino MN (2008) Encapsulation of natural antioxidants
extracted from Ilex paraguariensis. Carbohydr Polym 71:126–134
Wee SF, Gombotz WR (1998) Protein release from alginate matrices. Adv Drug Deliv Rev
31:267–285
Han J, Guenier A-S, Salmieri S, Lacroix M (2008) Alginate and chitosan functionalization for
micronutrient encapsulation. J Agric Food Chem 56:2528–2535
Lee SL, Hafeman AE, Debenedetti PG, Pethica BA, Moore DJ (2006) Solid-state stabilization of
α-chymotrypsin and catalase with carbohydrates. Ind Eng Chem Res 45:5134–5147
Martinsen A, Skjak-Braek G, Smidsrød O (1989) Alginate as immobilization material:
I. Correlation between chemical and physical properties of alginate gel beads. Biotechnol
Bioeng 33:79–89
Mi F-L, Her N-L, Kuan C-Y, Wong T-B, Shyu S-S (1997) Chitosan tablets for controlled-release
of theophylline—effect of polymer-drug wet or dry blending and anionic-cationic interpolymer
complex. J Appl Polym Sci 66:2495–2505
Peppas NA (1985) Analysis of Fickian and non-Fickian drug release from polymers. Pharm Acta
Helv 60:110–111
Pothakamury UR, Barbosa-Cánovas GV (1995) Fundamental aspects of controlled release in
foods. Trends Food Sci Technol 6:397–406
Santagapita PR, G omez Brizuela L, Mazzobre MF, Ramirez H, Corti HR, Villalonga Santana R,
Buera MP (2008) Structure/function relationships of several biopolymers as related to inver-
tase stability in dehydrated systems. Biomacromolecules 9:741–747
Santagapita PR, Mazzobre MF, Buera MP (2011) Formulation and drying of alginate beads for
controlled release and stabilization of invertase. Biomacromolecules 12:3147–3155
Santagapita PR, Mazzobre MF, Buera MP (2012) Invertase stability in alginate beads: effect of
trehalose and chitosan inclusion and of drying methods. Food Res Int 47:321–330
Schülke N, Schmid FX (1988) Effect of glycosylation on the mechanism of renaturation of
invertase from yeast. J Biol Chem 263:8832–8837
Shu S, Sun C, Zhang X, Wu Z, Wang Z, Li C (2010) Hallow and degradable polyelectrolyte
nanocapsules for protein drug delivery. Acta Biomater 6:210–217
Smidsrød O, Skjak-Braek G (1990) Alginate as immobilization matrix for cells. Trends Biotechnol
8:71–78
Zhou Z-d, Li G-y, Li Y-j (2010) Immobilization of Saccharomyces cerevisiae alcohol dehydro-
genase on hybrid alginate-chitosan beads. Int J Biol Macromol 47:21–26
Rheology and Stability of Citral/Sugar
Microemulsions
Abbreviations
T2 Relaxation times
1
H NMR Proton nuclear magnetic resonance
T Trehalose
T-MD Trehalose-maltodextrin
S Sucrose
S-MD Sucrose-maltodextrin
Τ Shear stress
Γ Shear rate
D3,2 Volume-surface mean diameter
SDS Sodium dodecyl sulfate
D4,3 Equivalent volume-mean diameter
N. Sosa • C. Schebor
Departamento de Industrias, Universidad de Buenos Aires, Ciudad Universitaria,
1428 Ciudad de Buenos Aires, Argentina
Departamento de Quı́mica Orgánica, Facultad de Ciencias Exactas y Naturales,
Universidad de Buenos Aires, Ciudad Universitaria, 1428 Ciudad de Buenos Aires, Argentina
Consejo Nacional de Investigaciones Cientı́ficas y Técnicas de la República Argentina,
CONICET, Ciudad de Buenos Aires, Argentina
O.E. Pérez (*)
Departamento de Industrias, Universidad de Buenos Aires, Ciudad Universitaria,
1428 Ciudad de Buenos Aires, Argentina
Consejo Nacional de Investigaciones Cientı́ficas y Técnicas de la República Argentina,
CONICET, Ciudad de Buenos Aires, Argentina
e-mail: [email protected]
1 Introduction
Pre-emulsions were prepared by mixing (2 min at 750 rpm) water, starch and either
sucrose or trehalose, maltodextrin, and citral oil. The exact mass of each ingredient
in the systems is detailed in Table 1. Emulsion formation was completed by high-
speed blending the pre-emulsions (10 min at 16,000) into an ice bath to avoid
thermal effects.
Rheology and Stability of Citral/Sugar Microemulsions 363
Table 1 Constituents (expressed as g/100 g) of emulsions formed by citral oil and an aqueous
phase containing trehalose (T), sucrose (S), trehalose-maltodextrin (TM-D), sucrose-maltodextrin
(SM-D)
Emulsions Citral Water Modified starch Trehalose Sucrose Maltodextrin
T 5.0 60.0 3.0 32.0 – –
S 5.0 60.0 3.0 – 32.0 –
T-MD 5.0 60.0 3.0 32.0 – 16.0
S-MD 5.0 60.0 3.0 – 32.0 16.0
2.2 Measurements
Figure 1 shows the flux curves for trehalose (T), trehalose-maltodextrin (T-MD),
sucrose (S), and sucrose-maltodextrin (S-MD) emulsions. For all emulsions, shear
stress (τ) increased linearly as shear rate (γ) increased. Viscosity, ∂τ=∂γ, remained
constant upon shear stress variation, corroborating the Newtonian character of the
emulsions. Emulsions containing maltodextrin resulted in a more viscous product,
as polysaccharides are known to increase the viscosity of the continuous phase. All
emulsions presented similar droplet size distribution, with a multimodal pattern as
in the volume vs. droplets size plots (not shown). The number size distribution
indicated only one population, with droplet sizes lower than 0.3 μm. The volume-
surface mean diameter (D3,2) and the equivalent volume-mean diameter (D4,3), both
derived from droplet size distribution, were used to characterize droplet diameters
(Table 2). D3,2 provides a measure of the mean diameter where most of the particles
fall. Emulsions containing MD presented much smaller D3,2 diameter. D4,3 is a very
sensitive parameter in relation to changes in particle size involving destabilization
processes (flocculation, coalescence). All samples presented high D4,3 values. In
order to identify the type of instability occurring in the emulsions, emulsions were
treated with sodium dodecyl sulfate (SDS), which acts in dissociating any existing
364 N. Sosa et al.
Table 2 Volume-surface mean diameter (D3,2) and equivalent volume-mean diameter (D4,3)
characterizing trehalose (T), sucrose (S), trehalose-maltodextrin (TM-D), and sucrose-
maltodextrin (SM-D) emulsions
Emulsions D3,2 D4,3 D4,3-SDS
T 16.60 4.09a 332.27 4.56a 15.09 1.23a
S 24.66 4.37b 315.99 4.12b 1.240 0.07b
T-MD 3.930 0.56c 281.20 6.65c 0.250 0.01c
S-MD 0.730 0.04c 216.41 9.86d 0.670 0.02d
Average for n ¼ 2 SD independently prepared emulsions. Different superscripts indicate signif-
icant differences ( p < 0.05)
flocs (Anton et al. 2002). A persistent droplet association indicated by the highest
D4,3-SDS values was observed for T emulsions, which indicated a certain degree of
coalescence. This phenomenon was also seen for the S emulsions, but to a lesser
extent. The MD-containing samples showed very low D4,3-SDS values, which means
that no coalescence had developed, but only reversible aggregation was present.
To study the global stability of the emulsions, backscattering (BS) profiles were
analyzed every 30 min for a total storage time of 3 h (Fig. 2). At each analyzed time,
the distribution of particles was homogeneous along the entire tube, since a regular
backscattering profile was obtained for these emulsions. A uniform decrease in the
backscattering profiles all along the entire tube length was observed after 30 min,
which was indicative of an increment of the mean droplet diameter, i.e., floccula-
tion (Álvarez Cerimedo et al. 2010). A minor decrease in BS % was observed after
30 min.
To elucidate the influence of different formulation components on the molecular
mobility of the systems, transverse relaxation times T2 associated with slow
relaxing protons (related to populations of water molecules displaying little
Rheology and Stability of Citral/Sugar Microemulsions 365
Fig. 2 Variation of backscattering (BS) profiles as a function of tube length, also shown with
emulsion storage time (samples were stored for 180 min; arrow denotes time). Measurements were
made at 25 C and natural pH. Each curve represents the mean value of two independently
prepared samples
Fig. 3 1H NMR T2
relaxation times obtained
after the application of
CPMG pulse sequence at
25 C
interaction with solids) were measured using the CPMG sequence (Fig. 3). Relax-
ation times T2-1 and T2-2 correspond to protons in the less and more mobile water
fractions, respectively. The presence of MD caused an important reduction in water
molecular mobility.
366 N. Sosa et al.
4 Conclusions
References
Álvarez Cerimedo MS, Iriart CH, Candal RJ, Herrera ML (2010) Stability of emulsions formulated
with high concentrations of sodium caseinate and trehalose. Food Res Int 43(5):1482–1493
Anton M, Beaumal V, Brossard C, Llamas G, le Denmat M (2002) Droplet flocculation and
physical stability of oil-in-water emulsions prepared with hen egg yolk. In: Anton M (ed) Food
emulsions and dispersions. Research Signpost, Kerala, pp 15–28
Gharsallaoui A, Saurel R, Chambin O, Cases E, Voilley A, Cayot P (2010) Utilisation of pectin
coating to enhance spray-dry stability of pea protein-stabilised oil-in-water emulsions. Food
Chem 122:447–454
McClements DJ (1999) Emulsion formation. In: McClements DJ (ed) Food emulsions: principles,
practice and techniques. CRC Press, Boca Raton, FL, pp 161–183
Thijssen HAC (1971) Flavour retention in drying preconcentrated food liquids. J Appl Chem
Biotechnol 21:372–377
Response Surface Methodology to Assay
the Effect of the Addition of Proteins
and Hydrocolloids on the Water Mobility
of Gluten-Free Pasta Formulations
Abbreviations
1 Introduction
In gluten-free pasta formulation (suitable for those suffering from celiac disease),
the influence of each constituent has a major importance on the final product
quality, especially water and hydrocolloid contents used to replace the gluten
matrix. Gluten-free doughs are mixed dispersed systems; the dispersion medium
contains several types of dispersed particles, with two main construction mate-
rials—polysaccharides and proteins. Four levels of structural hierarchy in dispersed
food systems can be distinguished: submolecular, molecular, supermolecular, and
macroscopic. Structural functions of a biopolymer depend upon its place in the
structural hierarchy of the product (Tolstoguzov 2000).
The presence of hydrocolloids and proteins in dough may modify the mobility of
water to interact with starch in the gelatinization process (Blanshard 1987). The
amount of water content and the presence of different components such as sugars,
gums, and proteins affect the extent to which the amorphous regions are plasticized
during gelatinization, modifying the corresponding enthalpy and the peak temper-
ature in a differential scanning calorimetry (DSC) thermogram.
The extent of hindered mobility of water molecules is related to the amount of
the unfrozen water content of the systems and their glass transition temperature (Tg)
and influences the conditions for starch gelatinization. The aim of this chapter is to
use response surface methodology (RSM) to analyze the effect of the addition of
proteins and hydrocolloids on water–starch interaction during gelatinization in
gluten-free doughs using a triangular mixture design through the analysis of
thermograms obtained by modulated differential scanning calorimetry (MDSC).
The amount of unfrozen water, water melting temperature, and glass transition
temperature for each formulation are also studied, and their relationship with dough
composition is determined.
Mixture designs are a special category of response surface designs that are partic-
ularly useful when the actual amounts of a component (factor) do not matter, but
instead the proportion of the whole made up by each component is taken into
account along with the interdependence of factors, by assuming that the factors
must have a sum equal to a constant value (Cornell 2002). In the present work, a
simplex-centroid augmented design with constraints was chosen to study the effect
of adding gums (0.51–2.52 %), protein (0.68–6.70 %), and water (35.5–39.5 %) to a
dough formulation containing fixed mass fractions of corn starch and corn flour (4:1
ratio, 53.5 %), NaCl (1.1 %), and sunflower oil (2.7 %). The design consisted of
12 runs: four points at the extreme vertices of the feasible quadrangular region
(1,2,3,4), four points at the edge centroids (5,6,7,8), one point at the overall centroid
(9), and three added points (A,B,C) to evenly cover the experimental region
(Fig. 1).
Response Surface Methodology to Assay the Effect of the Addition of Proteins. . . 369
Xa
5
0.0 0.0
nth
5
an
)
+l
g/g
4
0.0
oc
s(
us
0.1 3
in
8
t
ote
0
be
4
Pr
a
C 3
ng
0.0
5
um
9
6
s(
0.1
5
g/g
2
A B 0.0
)
0.2 1 7 2 1
0 0.0
0.80 0.85 0.90 0.95
Water (g/g)
Surface response analysis was used to determine the relationship of the unfrozen
water of this composite system:
X
3 2 X
X 3 Y
3
Y¼ βi Xi þ βij Xi X j þ β123 Xi ð1Þ
i¼1 i¼1 j¼iþ1 i¼1
where Y is the corresponding response variable, Xi are the coded process variables
(G, P, W ), and β are the regression coefficients. The same methodology was
followed with the peak temperature of water melting.
0.0
-0.2
-0.4
Heat Flow (W/g)
-0.6
-0.25
Fig. 2 Complete DSC thermogram corresponding to sample 9. The inset shows the gelatinization
transitions (G) and (M1) and the melting of the amylose–lipid complexes (M2)
a 0.25 b -5
unfrozen Tmw
water (g/g) (°C)
0.22 -6
LB +
XG
ums
(g/g -7
0.19 )
Gums
0.03
(g/g doug
6
h)
0.0
0.01
0.00
0
0.00
gh)
0.0 0.02
/g dou
5
0
0.1 0.4
2
/g)
0 9
0.0
0.04
0.3
ins (g
s (g
8
h)
tein
0.1 0.3
5
0.950 ug
Prote
0.06 o
Pro
7
0.915 0.3
g/ gd
0.2
0 0.880 ) 6 r(
0.845 (g/g 0.3 ate
Water 0.08 W
0.810 0.3
5
Fig. 3 Surface response of (a) unfrozen water content (g/g total water) and (b) peak melting
temperature (Tmw) as a function of dough composition (expressed as coded variables)
-24 78
a b
-25
-27
-28
74
-29
-30
0.20 0.22 0.24 0.26 0.28 0.30 0.32 0.34 72
0.20 0.25 0.30 0.35
Frozen water (g/g dough)
Frozen water (g/g dough)
Fig. 4 Relationship between the amount of frozen water content in the gluten-free doughs
(g water/g dough) on (a) the glass transition temperature, Tg ( C), and (b) temperature of
gelatinization peak G determined by DSC
was external to the granule was exhausted, the cooperative plasticization process
was arrested, and further gelatinization depended upon increased levels of molec-
ular mobility and granular swelling that were initiated and enhanced by heat. These
conditions demanded heating to higher temperatures. The obtained results also
showed that for a given water content, the addition of hydrocolloids caused a
significant increase ( p < 0.05) of peak G; this was probably the result of the
starch–hydrocolloid interaction, which produced a more stable structure, and less
water mobility due to gums–water interactions, requiring a higher temperature for
disorganization.
Moreover, when the frozen water content of the dough was progressively
reduced (0.334–0.219 g water/g dough), endotherms shifted to higher temperatures
(peak G from 74.98 to 77.34 C) as less water was available (Fig. 4b). As the
amount of frozen water decreased, water mobility diminished and the water-
diffusion-mediated step of the transition (peak G) required more energy. With
respect to the gelatinization enthalpies, no significant differences (P > 0.05) were
observed among the samples studied.
These results support the hypothesis that a reduced level of solvent plasticiza-
tion, resulting from the addition of nonaqueous solutes to a pure water system,
produces elevation of the gelatinization temperature. The reduced level of solvent
plasticization of the amorphous growth ring regions requires input of a greater
amount of thermal energy before the starch granule swells and begins to gelatinize.
5 Conclusions
References
Blanshard JMV (1987) Starch granule structure and function: a physicochemical approach. In:
Galliard T (ed) Starch: properties and potential. Wiley, New York, pp 17–18
Cornell JA (2002) Experiments with mixtures: designs, models, and the analysis of mixture.
Wiley, New York, pp. 22–95 and 132–220
Fennema OR (1996) Water and ice. In: Fennema OR (ed) Food chemistry. Marcel Dekker,
New York, pp 17–94
Kalichevsky MT, Blanshard JMV (1992) A study of the effect of water on the glass transition of
1:1 mixtures of amylopectin, casein and gluten using DSC and DMTA. Carbohydr Polym
19:271–278
Laaksonen T, Roos YH (2000) Thermal, dynamic-mechanical, and dielectric analysis of phase and
state transitions of frozen wheat doughs. J Cereal Sci 32:281–292
Räsänen J, Blanshard JMV, Mitchell JR, Derbyshire W, Autio K (1998) Properties of frozen wheat
doughs at subzero temperatures. J Cereal Sci 28:1–14
Roos YH (1986) Phase transitions and unfreezable water content of carrots, reindeer meat and
white bread studied using differential scanning calorimetry. J Food Sci 51:684–686
Roos YH, Karel M (1991) Water and molecular weight effects on glass transition in amorphous
carbohydrates and carbohydrate solutions. J Food Sci 56:1676–1681
Sandhu KS, Singh N, Kaur M (2004) Characteristics of the different types and their grain fractions:
physicochemical, thermal, morphological, and rheological properties of starches. J Food Eng
64:119–127
Stevens DJ, Elton GAH (1971) Thermal properties of the starch–water system. Part
I. Measurement of heat of gelatinisation by differential scanning calorimetry. Starch-Starke
23:8–11
Tester RF, Debon SJJ (2000) Annealing of starch. A review. Int J Biol Macromol 27:1–12
Tolstoguzov VB (2000) Foods as dispersed systems. Thermodynamic aspects of composition-
property relationships in formulated food. J Therm Anal Calorim 61:397–409
White PJ, Abbas IR, Pollak L, Johnson LJ (1990) Intra and interpopulation variability of thermal
properties in maize starch. Cereal Chem 67:70–73
Effects of aw Reduction and Type of Sugar
in Rheological Behavior, Water Mobility,
and Structural Changes in Apples
Abbreviations
1
H NMR Pulsed nuclear magnetic resonance spectrometer
C1 Constant
C2 Constant
C3 Constant
CPMG Carr-Purcell-Meiboom-Gill sequence
Ed Deformability modulus
γ Strain
G0 Storage moduli
G00 Loss moduli
h0 Coefficient of viscosity associated with Newtonian flow
hi Coefficients of viscosity associated with the Kelvin-Voigt elements
J (t t) Creep compliance
J0 Instantaneous compliance
Ji Retarded compliances
λi Retardation times
LM Optical microscopy
LM Light microscopy
MANOVA Multivariate analysis of variance
σR True stress
(σ RR) True rupture stress
τ Constant stress applied
t Time
T2 Transverse relaxation times
TEM Transmission electron microscopy
TEM Transmission electron microscopy
W Work at rupture or toughness
w Frequency
εR Hencky strain
εRR True rupture strain at σ RR
1 Introduction
Fresh apples (Malus pumila, Granny Smith var.; 13 Brix; aw ¼ 0.98) were cut to
obtain outer parenchyma cylinders (15 mm in diameter, 15 mm in thickness) for
compression test, water mobility analysis and structural observations, and discs
(30 mm diameter, 6 mm in length) for dynamic tests.
Effects of aw Reduction and Type of Sugar in Rheological Behavior. . . 377
Osmotic dehydration was performed by immersing apple pieces into 22.0 % w/w and
38.7 % w/w glucose or 34.4 % w/w and 48.0 % w/w trehalose aqueous solutions with
forced convection at atmospheric pressure to reach aw 0.97 or aw 0.94, respectively.
Mechanical behavior at large deformations was determined by uniaxial com-
pression with an Instron 1011 machine (USA) (20 mm/min head rate, 500 N load
range, 30 replicates). True stress (σ R), Hencky strain (εR), true rupture stress (σ RR),
true rupture strain at σ RR (εRR), deformability modulus (Ed) up to 10 % of com-
pression, and work at rupture or toughness (W ) were obtained (Peleg 1978).
Viscoelastic properties were characterized in a Paar Physica MCR 300 rheometer
(Austria) (controlled strain mode 3 cm diameter parallel plate geometry, 12 repli-
cates). Storage (G0 ) and loss (G00 ) moduli were measured in the linear viscoelastic
range (using a strain amplitude value of 0.01 % and frequency (ω) 0.1–100 s1).
Creep/recovery test was conducted by applying a constant shear stress of 35 Pa until
100 s (recovery: 200 s). A mechanical model consisting of a spring in series with
two Voigt elements and a dashpot element was used to fit the compliance response
(Eq. (1)):
X
2
J ðt; τÞ ¼ ðJ 0 Þ þ ðJ i Þ 1 et=λi þ ðt=η0 Þ ð1Þ
i¼1
where J (t, τ) is the creep compliance (¼γ(t)/τ) with γ(t) being the strain at time
t and τ the constant stress applied); J0 is the instantaneous compliance at t ¼ 0; Ji are
the retarded compliances; λi (¼ηi Ji) are the retardation times and ηi are the
coefficients of viscosity associated with the Kelvin-Voigt elements; and η0 is the
coefficient of viscosity associated with Newtonian flow and its inverse, the steady-
state fluidity of the material.
Transverse relaxation times (T2) were determined in a Bruker Minispec mq/20
pulsed nuclear magnetic resonance spectrometer (1H NMR) (0.47 T; 20 MHz) using
the Carr-Purcell-Meiboom-Gill sequence (CPMG) (2 ms pulse spacing; 8 scans;
recycled delay 4 s) (6 replicates). The T2 relaxation experimental curves were fitted
by a three-exponential equation, and Ai values were normalized (Peleg 1978, 1984):
X
n
Y ¼ ð y0 Þ þ ðAi Þ et=T 2i ð2Þ
i¼1
Changes in the microstructure and ultrastructure of apple tissue were assessed using
optical microscopy (LM) (Carl Zeiss Axioskop 2) and transmission electron
microscopy (TEM) (JEOL JEM 1200 EX II).
Fresh apple samples were used in each case as controls (called control 0.94 and
control 0.97 for experiments at aw 0.94 and 0.97, respectively).
Statgraphics Plus (USA) and the InfoStat (Argentina) software were used for
nonlinear regression and for one-factor multivariate analysis of variance
(MANOVA), respectively. In the multivariate analysis, Hotelling test was applied
to determine significance level with Bonferroni-corrected level of α ¼ 0.05.
378 S. Vicente et al.
σR (MPa)
0.40
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
0.00 0.25 0.50 0.75 1.00 1.25 1.50
εR
Fig. 1 Typical compression curves of true stress vs. true strain for fresh apple (control) and apples
osmotically dehydrated in aqueous solutions of glucose or trehalose at aw ¼ 0.97 or aw ¼ 0.94.
Control (filled circle); glucose, aw 0.94 (cross); glucose, aw 0.97 (single solid line); trehalose, aw
0.94 (diamond); trehalose, aw 0.97 (spaced solid line)
Table 1 Mean values of compression parameters for apples osmotically dehydrated in aqueous
solutions of glucose and trehalose with aw ¼ 0.97 or aw ¼ 0.94
σ RR (MPa) Ed (MPa) W (MJ/m3)
Treatment SD εR SD
R
SD SD
Control 0.97 0.33 0.04 0.26 0.02 1.7 0.2 44 8 A
Control 0.94 0.34 0.03 0.19 0.02 2.1 0.3 37 7 B
Immersed in 22.0 % w/w 0.11 0.02 0.79 0.06 0.04 0.01 37 6 C
glucose solution
Immersed in 34.4 % w/w 0.12 0.02 1.02 0.07 0.02 0.01 36 5 D
trehalose solution
Immersed in 38.7 % w/w 0.23 0.04 0.64 0.08 0.05 0.02 41 7 E
glucose solution
Immersed in 48.0 % w/w 0.22 0.07 0.81 0.07 0.03 0.01 42 6 F
trehalose solution
MANOVA of treatments was highly significant for viscoelastic measurements (F20, 652 ¼ 33.16;
P < 0.0001). Post hoc multiple comparisons used Hotelling tests based on Bonferroni correction
α ¼ 0.05. Different letters indicate significant differences at P
0.05. Control 0.97 and Control
0.94: controls (fresh apple) for the fruit batch used for tests at aw 0.97 and 0.94, respectively. SD
standard deviation
a G´ (Pa)
1.0E+06
1.0E+05
1.0E+04
0.1 1 10 100
ω (1/s)
b G´´ (Pa)
1.0E+05
Fig. 2 Mechanical
spectrum for fresh apple
(control) and apples
osmotically dehydrated
in aqueous solutions of
glucose or trehalose at 1.0E+04
aw ¼ 0.97 or aw ¼ 0.94.
(a) Storage modulus G0 ;
(b) loss modulus G00 .
Control (single solid line);
glucose, aw 0.97 (open
circle); glucose, aw 0.94 1.0E+03
(diamond); trehalose, 0.1 1 10 100
aw 0.97 (triangle);
trehalose, aw 0.94 (square) ω (1/s)
380 S. Vicente et al.
γ
0.0030
0.0025
0.0020
0.0015
0.0010
0.0005
0.0000
0 50 100 150 200 250 300
t (s)
Fig. 3 Typical experimental creep/recovery curves (deformation γ vs. time t) for fresh apple
(control) and apples osmotically dehydrated in aqueous solutions of glucose or trehalose at
aw ¼ 0.97 or aw ¼ 0.94. Control (single solid line); glucose, aw 0.97 (open circle); glucose, aw
0.94 (diamond); trehalose, aw 0.97 (triangle); trehalose, aw 0.94 (square)
Table 2 Means of viscoelastic parameters derived by fitting Eq. (2) to compliance curves from
creep phase for apples osmotically dehydrated in aqueous solutions of glucose and trehalose with
aw ¼ 0.97 or aw ¼ 0.94
J0 106 J1 106 J2 106 η0 107
(1/Pa) (1/Pa) (1/Pa) λ1 (s) λ2 (s) (Pa s)
Treatment SD SD SD SD SD SD
Control 0.94 31 21 0.9 0.3 31 15 21 92 A
Control 0.97 41 2.3 0.3 0.9 0.3 30 12 32 52 B
Immersed in 21 2 92 51 27 7 2.8 0.4 1.05 0.06 C
22.0 % w/w glu-
cose solution
Immersed in 27 3 18 3 93 30 7 2.7 0.4 0.6 0.1 D
34.4 % w/w treha-
lose solution
Immersed in 28 2 12 2 7.3 0.8 22 2 2.2 0.2 0.9 0.2 D
38.7 % w/w glu-
cose solution
Immersed in 32 5 15 3 92 21 3 2.3 0.3 0.8 0.2 D
48.0 % w/w treha-
lose solution
MANOVA of treatments was highly significant for viscoelastic measurements (F30, 265 ¼ 5.83;
P < 0.0001). Post hoc multiple comparisons used Hotelling tests based on Bonferroni correction
α ¼ 0.05. Different letters indicate significant differences at P
0.05. SD standard deviation.
Control 0.97 and Control 0.94: controls (fresh apple) for the fruit lot used for tests at aw 0.97
and 0.94, respectively
Effects of aw Reduction and Type of Sugar in Rheological Behavior. . . 381
Table 3 Mean proton transverse relaxation times and signal percentages of the two components
for apples osmotically dehydrated in aqueous solutions of glucose and trehalose with aw ¼ 0.97 or
aw ¼ 0.94
C1 (%) C2 (%) T21 (ms) T22 (ms)
Treatment SD SD SD SD
Immersed in 22.0 % w/w glucose 24 9 76 9 72 25 290 38 A
solution
Immersed in 34.4 % w/w trehalose 40 8 60 9 63 12 201 20 B
solution
Immersed in 38.7 % w/w glucose 27 1 73 1 56 5 337 12 C
solution
Immersed in 48.0 % w/w trehalose 39 3 61 3 31 4 215 20 D
solution
MANOVA of treatments was highly significant for viscoelastic measurements (F9, 126 ¼ 23.32;
P < 0.0001). Post hoc multiple comparisons used Hotelling tests based on Bonferroni correction
α ¼ 0.05. Different letters indicate significant differences at P
0.05. SD standard deviation
4 Conclusions
The loss of turgor could explain the lower elasticity and stiffness (lower Ed, G0 and
higher J0) and lower compression strength (less σ RR) in apples subjected to osmosis.
The plasticization of the wall structure made by sugars as well as the compactness
382 S. Vicente et al.
of the walls due to the removal of water and the lack of turgor could be responsible
for the greater deformation of osmotic dehydrated tissues (higher εRR and higher
capacitances Ji) (Martinez et al. 2007).
These results can improve preservation technology design for minimally
processed apples where aw reduction is one of the hurdles to obtain fruit stability.
Acknowledgements The authors want to thank the financial support from Universidad de Buenos
Aires, CONICET, and ANPCyT of Argentina and from BID. They also thank Cargill Inc.,
Argentina, and EXPOFRUT S.A. for supplying the trehalose and the apples.
References
Alzamora SM, Castro MA, Vidales SL, Nieto AB, Salvatori DM (2000) The role of tissue
microstructure in the textural characteristics of minimally processed fruits. In: Alzamora
SM, Tapia MS, Lopez MA (eds) Minimally processed fruits and vegetables: fundamental
aspects and applications. Aspen Publishers Inc, Gaithersburg, MD
Ferrando M, Spiess WEL (2001) Cellular response of plant tissue during the osmotic treatment
with sucrose, maltose and trehalose solutions. J Food Eng 49:115–127
Martinez VY, Nieto AB, Salvatori DM, Castro MA, Viollaz PE, Alzamora SM (2007) Viscoelastic
characteristics of Granny Smith apple during glucose osmotic dehydration. J Food Eng
83:394–403
Peleg M (1978) An empirical method for estimation of the yield stress from relaxation data. Mater
Sci Eng 33:289–293
Peleg M (1984) A note on the various strain measures at large compressive deformations.
J Texture Stud 15:317–326
Effect of Water Content on Thermo-Physical
Properties and Freezing Times of Foods
Abbreviations
1 Introduction
For the prediction of temperature change in different foodstuffs during freezing and
thawing processes, accurate estimation of the thermo-physical properties of the
product is necessary, such as specific heat, density, freezable water content,
enthalpy, and initial freezing temperature. These data allow the adequate design
and optimization of equipment and processes. Water is a main component in all
foods and greatly influences the behavior of these properties, depending on its
concentration. During the freezing process, which involves the phase change of
water into ice, the specific heat, thermal conductivity, and density undergo abrupt
changes due to the latent heat release. This complex process does not have an
analytical solution and it can be described as a highly nonlinear mathematical
problem. Many difficulties arise when trying to numerically simulate the freezing
process, especially when using the finite element method (FEM), which is espe-
cially useful when dealing with irregular-shaped foodstuffs. Several techniques
have been applied to consider the large latent heat release when using FEM. One
traditional method is the use of the apparent specific heat, where the sensible heat is
merged with the latent heat to produce a specific heat curve with a large peak
around the freezing point, which can be considered a quasi-delta-Dirac function
with temperature (depending on the amount of water in the food product) (Pham
2008). However, this method usually destabilizes the numerical solution. Imple-
mentation of the enthalpy method, which can be obtained through the integration of
the specific heat with temperature (Fikiin 1996; Comini et al. 1990; Pham 2008;
Santos et al. 2010), and the Kirchhoff function, which is the integral of the thermal
conductivity, allows the reformulation of the heat transfer differential equation into
a transformed partial differential system with two mutually related dependent
variables H (enthalpy) and E (Kirchhoff function) (Scheerlinck et al. 2001).
These functions, H and E versus temperature, are smoother mathematical functions
compared to the specific heat, thermal conductivity, and density versus tempera-
ture, avoiding inaccuracies and/or divergence of the numerical method. Even
though it brings great advantage to the resolution of the problem, with the simul-
taneous enhancement of the computational speed of the program, this transforma-
tion of variables is not widely used in the literature. Unleavened dough and cooked
minced meat were selected due to their significant difference in water content in
order to explore the performance of the computational code written using the
enthalpy-Kirchhoff formulation. Another important reason is because cooked
minced meat and dough are both present in several ready-to-eat meals, therefore
contributing valuable information to food processors interested in optimizing
cooling and freezing operating conditions of semi- or fully processed goods.
Effect of Water Content on Thermo-Physical Properties and Freezing Times. . . 385
Two different food materials, unleavened dough and cooked minced meat, which
have significant difference in moisture content, were selected. Thermal properties
of these products were determined using experimental measurements and predic-
tive equations.
Fig. 1 Apparent specific heat versus temperature of dough and meat obtained through DSC
measurements
Table 1 Experimental data obtained through DSC measurements. SEM values are in parentheses
Tp ( C) Tm ( C) ΔHm (J/g) xb (%) xw0 (%)
Cooked 3.17 (0.24) 8.86 (0.79) 172.05 (14.25) 14.66 (2.31) 66.29 (1.96)
minced meat
Unleavened 9.45 (0.74) 18.29 (1.38) 33.27 (5.82) 19.26 (1.155) 29.25 (0.59)
dough
Thermal conductivity of dough and cooked meat was calculated using predictive
equations. In the case of frozen dough, the Choi and Okos (1986) equation was used
for the thermal conductivity and density:
X
k ðT Þ ¼ xi v k i ðT Þ ð1Þ
1
ρð T Þ ¼ X x i ð2Þ
ρi
where ρ(T ) is the global density and ρi is the density of the component i (water,
carbohydrates, ice, ash, etc.). The fractions “xi” correspond to the mass fraction of
each component. Ice content as a function temperature (at T < Tf) was estimated
using the equation proposed by Miles et al. (1983):
Tf
xh ¼ ðxwo xb Þ 1 ð3Þ
T
where xh is the mass fraction of ice, and xw0 is the total mass fraction of water in the
foodstuff. At a temperature range above Tf, the formula proposed by Sorenfors
(1973) for minced meat was applied to calculate the thermal conductivity as a
function of the water content in the meat product:
At temperatures below freezing state the following was used (Miles et al. 1983):
Tf
kðT Þ ¼ 1:745 1 þ 0:233 ð5Þ
T
0.39 1.1
a b
0.38 1
0.9
0.37
0.8
0.36
k [W/m °C]
k [W/m °C]
0.7
0.35
0.6
0.34
0.5
0.33
0.4
0.32
-40 -30 -20 -10 0 10 20 30 40 -40 -30 -20 -10 0 10 20 30 40
Temperature (°C) Temperature (°C)
Fig. 2 Thermal conductivity versus temperature for (a) dough, and (b) cooked minced lean meat
388 M.V. Santos et al.
The equation is valid in the domain Ω, where T is the temperature, and r and z the
radial and axial coordinates, respectively. The boundary and initial conditions are
∂T ∂T
nz þ nr k ¼ hðT ext T Þ t 0 in ∂Ω1 ð7Þ
∂z ∂r
∂T ∂T
nz þ nr k ¼ 0 t 0 in ∂Ω2 ð8Þ
∂z ∂r
T ¼ T0 t ¼ 0 in Ω ð9Þ
where ∂Ω1 is the domain of the convective interface which corresponds to the
surface in contact with the cooling air, nz and nr are the normal outward unit vector
components, ∂Ω2 is the domain with axial symmetry (flux zero condition), Text is
the external air temperature, T0 is the initial temperature, and h is the surface heat
transfer coefficient. By performing the following change of variables:
ZT
H¼ ρðT ÞC pðT ÞdT
T*
ZT
E¼ kðT ÞdT
*
T
∂H
¼ ∇2 E in δΩ t 0 ð10Þ
∂t
ð∇EÞ n ¼ hðT T ext Þ in δΩ t 0 ð11Þ
H ¼ H0 t¼0 ð12Þ
In order to solve the new governing heat transfer equation, a numerical finite
element algorithm was developed. The enthalpy distribution at any point in the
domain was approximated by using interpolating functions and the node enthalpies
Effect of Water Content on Thermo-Physical Properties and Freezing Times. . . 389
E versus Enthalpy
50
a
45
40
35
E (Kirchhoff)
30
25
20
15
10
0
0 0.5 1 1.5 2 2.5
Volumetric Enthalpy x 108
20
Temperature (°C)
10
-10
-20
-30
-40
0 0.5 1 1.5 2 2.5
Volumetric Enthalpy x 108
Fig. 3 (a) Kirchhoff versus enthalpy, and (b) temperature versus enthalpy for cooked
minced meat
in the given element. The resulting equation after applying the weight residual
formulation and Galerkin method in the FEM was (Bathe 1996)
dH
CG þ FG T ðH Þ þ KG EðHÞ ¼ m ð13Þ
dt
The numerical code was used to calculate freezing times of an irregular shape
commonly found in industrial processing of dough. The dimensions of the irregular
domain were fixed according to the industrial process, and it could not be assim-
ilated as a regular object. It has a symmetry axis, and therefore it was considered a
solid of revolution. Figure 4a shows the mesh used for the finite element simulation,
considering that there was a symmetry condition at border 1. The code used as input
two established the thermal properties corresponding to unfermented dough and
minced meat, which had a significant difference in their amount of water, in order to
visualize their effect on the thermo-physical properties and freezing times. More-
over, the effect of the heat transfer coefficients was also analyzed. Freezing time
values were calculated considering an initial uniform temperature of 5 C, an
external fluid temperature of 30 C, and a final temperature of 15 C at the
warmest point inside the foodstuff. The finite element code developed in this work
was very stable and required minimum computational efforts (CPU times <5 min);
this code also solves highly nonlinear problems such as freezing, which has an
apparent specific heat with a very high peak around the freezing region. This is an
important aspect, because most common commercial software are not able to solve
this type of phase change problems. In the case of dough, freezing times of samples
with the same dry matter formulation but different total water contents of 29 and
40 % wb are compared in Table 2; the corresponding values of bound water were
considered in each case. Freezing times for minced meat with different moisture
Effect of Water Content on Thermo-Physical Properties and Freezing Times. . . 391
a
z=3
z (cm)
Simmetry
axis border 1
r (cm) r=4
4
2
h = 10 W/m2 °C
0 0.40 w/w (wet basis)
Temperature (°C)
-4
-6
-8
-10 2
h = 1000 W/m °C
-12
-14
0 10 20 30 40 50 60 70 80 90
Time (min)
Fig. 4 (a) Geometry and mesh of bi-dimensional object used in simulations, and (b) freezing
curves for dough with different water contents considering initial uniform temperature of 5 C, and
external fluid temperature of 30 C
Table 2 Freezing times applied to dough with different moisture contents Text ¼ 30 C
Dough Minced meat
*
xw0 ¼ 0.29 xw0 ¼ 0.40 xw0 ¼ 0.66 xw0 ¼ 0.72
Surface heat transfer coefficient
*
xb ¼ 0.192 xb ¼ 0.166 xb ¼ 0.146 xb ¼ 0.125
h (W/m2 C) Freezing times (min) Text ¼ 30 C
5 108.83 161.32 243.95 261.61
20 40.18 52.31 68.93 72.94
100 18.79 22.19 22.84 23.57
1,000 12.74 13.32 11.00 11.00
*water mass fraction (w/w wet basis)
contents but the same dry matter composition is also shown in Table 2. As an
example, Fig. 4b shows the time–temperature curves for dough formulations with
different free water contents. The extension of the plateau zone depends on the
h value, being shorter as h increases. Similar behavior was observed for minced
meat. When freezing is conducted with low heat transfer coefficients (stagnant air
chambers) h
20 W/m2 C, products with the same geometry but different free
water contents (xfree ¼ xw0xb) such as dough (29 and 40 % water wet basis) and
392 M.V. Santos et al.
meat (66 and 72 % wet basis) show an increase of the freezing times as the amount
of freezable water increases due to a larger amount of latent heat of ice crystalli-
zation that has to be removed from the sample. In contrast, for higher h values,
freezing times become almost independent of the food water content. The Biot
number (Bi ¼ h*L/k) has an important effect on these results since there is an
internal heat transfer control and the thermal conductivity (k) of the solid governs
the freezing rate. Then, as k increases with the ice content, foods containing more
ice will be frozen at higher rates. This analysis explains the shorter freezing times of
meat with respect to dough at high h values. It is commonly assumed that foodstuffs
with large amounts of water will need more time to reach a certain final freezing
temperature. This study shows that this statement is not always true and that
freezing times are greatly dependent on the bound water present and the surface
heat transfer coefficient (h), as well as the total moisture content. This code,
combined with accurate thermo-physical data, allows the adequate design and
optimization of equipment and processes used in the freezing industry in a fast
and accurate manner.
References
Abbreviations
1 Introduction
The physalis (Physalis peruviana l.), member of the family Solanaceae and the
genus Physalis spp., consists of more than 80 varieties; their main characteristic is
that their fruits are locked inside a sepal in bucket form. This fruit, originally from
the South American Andes, is the best known species of this genus. It is also
characterized by its high sugar content and significant amount of vitamins A, B,
and C, as well as iron and phosphorus, which could be considered as medicinal
properties (Puente et al. 2010). Colombia is the main producer of physalis, followed
by South Africa. Colombian physalis is characterized as having a more pronounced
color and higher sugar content, making it more appealing in the market. At present,
various processed products from the physalis are available, including jams, jellies,
raisins, and candies covered in chocolate, juices, nectars, and pulps. According to
the National Research Council, the juice of overripe physalis has a high pectinase
content, making it an excellent raw material to prepare jams and derivative prod-
ucts, with low production costs (Ministerio de Agricultura y Desarrollo Rural
2009).
Most unit operations for fruit juice processing utilize heat transfer. Modelling
and optimization of these operations are difficult, since juice and physical proper-
ties change considerably with concentration and temperature. Physical and thermal
properties of foods determined at temperatures above freezing are of limited use at
freezing conditions. Many food properties show a unique dependence on the state of
the water in a food material. During the freezing process, water changes gradually
from the liquid phase to solid ice. Since the properties of ice are different from those
of liquid water, properties of food determined at temperatures above freezing are
often not valid for subfreezing conditions (Singh and Sarkar 2005).
To determine the thermal properties of juices and derivatives, it is essential to
calculate parameters such as temperature distribution during freezing, frozen stor-
age, and estimation of freezing time. In food engineering processes, the study of
freezing time is important, because it is a tool to estimate refrigeration requirements
for freezing systems, as well as for designing and dimensioning the equipment
necessary for industry (Delgado and Sun 2001).
In this chapter, results of investigations in thermophysical properties of physalis
frozen juice are presented. Freezing-point depression (FPD) and thermal conduc-
tivity at several temperatures and concentrations were calculated (Gabas
et al. 2003; Giraldo et al. 2010). This investigation provides important values for
the freezing storage process, improving preservation and facilitating the marketing
of physalis juice for both export and internal consumption.
Influence of Fluid Concentration on Freezing-Point Depression and Thermal. . . 395
Fig. 1 Schematic diagram of the apparatus used for measuring freezing-point depression
396 J. Telis-Romero et al.
In order to establish the relationship between FPD and soluble solids, the
mathematical model of Miles et al. (1983) was used. The following Eq. (1) was
adjusted with a determination coefficient (R2) of 0.983 for all experimental data:
29:683ð1 Xw Þ
Tf ¼ ð1Þ
0:149ð1 Xw Þ 1
where Tf is the initial freezing temperature ( C), and Xw is the water fraction
(kg kg1).
Although the measurement of freezing point is difficult for many products, we
can easily estimate the freezing point from the soluble solids, according to the
mathematical model showed above.
3 Thermal Conductivity
Many models to predict thermal conductivity of foods have been published, but
they almost always have been adjusted with experimental data, due to the effect of
solid structure and ice formation on thermal conductivity (Delgado et al. 1997).
These models, based on heat conduction through perpendicular layers, provide a
better fit in a defined range, taking into account the rapid increase of ice thermal
conductivity when temperature decreases in values under 0 C. The parameters of
these equations determined the ability to predict thermal conductivity correctly at
different temperatures in the range of application (Singh and Sarkar 2005).
Measurement of thermal conductivity has involved the use of both steady-state
and non-steady-state methods. A steady-state method used a guarded plate to
measure the thermal conductivity of frozen foods, and a non-steady-state method
used a probe to measure the thermal conductivity of food materials (Singh and
Sarkar 2005).
For the steady state, conduction inside the cell is described by the Fourier
equation in cylindrical coordinates, with boundary conditions corresponding to
heat transfer between two concentric cylindrical surfaces at constant temperature,
as given by Eqs. (2)–(4):
∂q_ ∂T
¼ λðT Þ ð2Þ
∂S ∂r
T ð r ¼ R1 Þ ¼ T 1 ð3Þ
T ð r ¼ R2 Þ ¼ T 2 ð4Þ
where q_ is the heat flux in the thermal resistance (W), r the radius (m), R1 and R2 the
external and internal radius of the cylinder (m), respectively, S the surface area of a
cylinder of radius r (m2), T the temperature ( C), T1 the steady-state temperature in
398 J. Telis-Romero et al.
Fig. 2 Schematic diagram of the apparatus used for measuring effective thermal conductivity
the internal cylinder ( C), T2 the steady-state temperature in the thermostatic bath
where the cell was immersed ( C), and λ the thermal conductivity of the sample at
the average temperature (T1+T2)/2 (W m1 C1) (Chen et al. 1996).
Equation (2) can be integrated to find the thermal conductivity in samples, λ,
from experimental measurements of T1 and T2 under controlled conditions
(Delgado et al. 1997; Gabas et al. 2003; Telis et al. 2007):
log RR21
λ ¼ q_ ð5Þ
2π ðT 1 T 2 Þ
where T and Tf are the sample temperature ( C) and the initial freezing temperature
( C), respectively.
The above equation was used to determine the frozen water of food at any
temperature. This method is simple, and the procedure involves the following steps:
assuming that the initial freezing temperature and the water fraction are known; that
their values are substituted in Eq. (6); and that the term representing the frozen
soluble solids is calculated.
For all concentrations of physalis juice, in the frozen state, thermal conductivity
was strongly affected by both juice concentration and temperature. Thermal con-
ductivity increased with decreasing concentration and temperature. This behavior is
a consequence of the higher fraction of ice present in the more diluted samples
exposed to temperatures well below their initial freezing point. Schwartzberg
(1981) presented a mathematical study of the food-freezing process and reported
that the thermal conductivity of ice is roughly 3.7 times greater than that of water,
which explains the marked increase in thermal conductivity of foods during freez-
ing (Schwartzberg 1981).
In order to find real value for thermal conductivity values above the freezing
point and below the freezing point (Fig. 3), Eqs. (7) and (8) are proposed.
Equation (7), which was obtained from water fraction (Xw) and temperature,
was adjusted from experimental data above the freezing point (R2 ¼ 0.988).
This equation describes the concentration effect when the intrinsic thermal con-
ductivity of solute is lower compared to the water:
400
Table 2 Thermal conductivity and frozen water fraction of different physalis juices above (shadow) and below (white) freezing point
Temperature ( C) 24 21 18 15 8 0 5 15 25
1 1
Fraction water (w/w) Brix Thermal conductivity (W m K )
0.50 50.0 0.613 0.544 0.409 0.388 0.398 0.398 0.401 0.389 0.388
0.57 43.0 0.833 0.686 0.590 0.434 0.436 0.417 0.435 0.430 0.429
0.64 36.0 0.921 0.930 0.773 0.647 0.440 0.492 0.482 0.467 0.485
0.71 29.0 1.178 1.019 0.946 0.821 0.433 0.503 0.535 0.524 0.533
0.78 22.0 1.434 1.426 1.309 1.245 0.648 0.562 0.567 0.578 0.566
0.85 15.0 1.526 1.456 1.502 1.276 0.881 0.604 0.598 0.609 0.596
Frozen water fractiona
0.50 50.0 0.160 0.112 0.047 0 0 0 0 0 0
0.57 43.0 0.252 0.206 0.146 0.061 0 0 0 0 0
0.64 36.0 0.344 0.302 0.245 0.166 0 0 0 0 0
0.71 29.0 0.438 0.399 0.347 0.275 0 0 0 0 0
0.78 22.0 0.582 0.553 0.516 0.463 0.185 0 0 0 0
0.85 15.0 0.659 0.631 0.595 0.544 0.276 0 0 0 0
a
Estimated by Eq. (6)
J. Telis-Romero et al.
Influence of Fluid Concentration on Freezing-Point Depression and Thermal. . . 401
Fig. 3 Thermal conductivity for physalis juice at different temperatures and concentrations
Experimental results of thermal conductivity below the freezing point can be described
by the Fikiin equation (Eq. (8)), which was fitted to the experimental data by a multiple
regression procedure, resulting in R2 ¼ 0.989 (Schwartzberg 1981):
Tf
λ ¼ 1:884Xw 1 þ 0:321 ð8Þ
T
Comparing Eqs. (7) and (8), it is revealed, except for the constants that exhibit the
same dependence of temperature and water fraction, we can conclude that the
frozen water fraction is the main factor affecting thermal conductivity in the frozen
domain.
4 Conclusions
The average value of FPD varied between 3.4 and 14.1 C for physalis juice
concentrations in the range of 20–50 Brix. As expected, the initial freezing point
decreased with an increase in physalis juice concentration. The coaxial
402 J. Telis-Romero et al.
dual-cylinder method was applied with good accuracy for measuring the effective
thermal conductivity of frozen physalis juice at different concentrations. In the
frozen state, the thermal conductivity decreased with increasing concentration
and temperature. The mass fraction of ice in physalis juice increased during
freezing, causing an increase in the thermal conductivity of the aqueous mixture
due to the higher thermal conductivity of ice compared to that of pure water.
Predictive methods generally require input values for initial freezing tempera-
ture, product moisture content, and thermal properties of unfrozen product. Any
errors associated with these values can lead to erroneous predictions. Fortunately,
food properties under unfrozen conditions can be measured quite accurately by
using well-established procedures.
Acknowledgements The authors would like to acknowledge and thank CNPq and FAPESP
(process 2009/13033-9, 2009/03840-4) for their financial support.
References
Barnett S (1973) Freezing of coffee extract to produce a dark coloured freeze-dried product.
AICHE Symp Ser 132(69):26–32
Chang HD, Tao LC (1981) Correlations of enthalpies of food systems. J Food Sci 46
(5):1493–1497
Chen P, Chen XD, Free KW (1996) Measurement and data interpretation of the freezing point
depression of milks. J Food Eng 30(1–2):239–253
Chen XD, Chen P (1996) Freezing of aqueous solution in a simple apparatus designed for
measuring freezing point. Food Res Int 29(8):723–729
Delgado AE, Gallo A, de Piante D, Rubiolo A (1997) Thermal conductivity and frozen strawberry
and spinach. J Food Eng 31(2):137–146
Delgado AE, Sun DW (2001) Heat and mass transfer models for predicting freezing processes:
a review. J Food Eng 47(3):157–174
Gabas AL, Telis-Romero J, Telis VRN (2003) Influence of fluid concentration on freezing point
depression and thermal conductivity of frozen orange juice. Int J Food Prop 6:543–556
Giraldo GI, Gabas AL, Telis VRN, Telis-Romero J (2010) Propiedades termofı́sicas del jugo
concentrado de lulo a temperaturas por encima del punto de congelaci on. Cienc Tecnol
Aliment 30(1):90–95
Lerici CR, Piva M, Dallarosa M (1983) Water activity and freezing point depression of aqueous
solutions and liquid foods. J Food Sci 48:1667–1669
Miles CA, van Beek G, Veerkamp CH (1983) Calculation of thermophysical properties of foods
I. In: Jowitt R et al (eds) Physical properties of foods. Applied Science Publishers Ltd, Barking,
Essex, pp 269–312
Ministerio de Agricultura y Desarrollo Rural (2009) La uchuva colombiana lı́der en el
mundo, Boletı́n Informativo del Ministerio de Agricultura y Desarrollo Rural, Bogotá, Colom-
bia, 1–3
Puente LA, Pinto-Mu~noz CA, Castro ES, Cortés M (2010) Physalis peruviana Linnaeus, the
multiple properties of a highly functional fruit: a review. Food Res Int 44(7):1733–1740
Schwartzberg HG (1981) Mathematical analysis of the freezing and thawing of foods. AICHE
Summer National Meeting, Detroit, USA, pp 1–33
Influence of Fluid Concentration on Freezing-Point Depression and Thermal. . . 403
Singh RP (1995) Thermal properties of frozen foods. In: Rao RS (ed) Engineering properties of
foods. Marcel Dekker, New York, pp 139–167
Singh RP, Sarkar A (2005) Thermal properties of frozen foods. In: Rao MA, Rizvi SSH, Datta AK
(eds) Engineering properties of foods. Taylor & Francis, Boca Raton, FL, pp 175–201
Telis VRN, Telis-Romero J, Sobral PJA, Gabas AL (2007) Freezing point and thermal conduc-
tivity of tropical fruit pulps: mango and papaya. Int J Food Prop 10(1):73–84
Freezing Rate Effect on Thermal Transition
of Blueberries
P. Dı́az-Calder
on, O. Henrı́quez, J. Enrione, and S. Matiacevich
Abbreviations
1 Introduction
World blueberry consumption has risen mainly due to of its health benefits
(Nicoletta et al. 2008), such as low calories and the presence of anticancer and
antioxidant compounds that prevent various diseases, which allow to explain
blueberries have become an important component of a healthy diet (Yrjo
P. Dı́az-Calder
on
Biopolymer Research and Engineering Lab, School of Nutrition and Dietetics,
Universidad de los Andes, Monse~nor Alvaro del Portillo 12455, Las Condes, Santiago, Chile
O. Henrı́quez • S. Matiacevich (*)
Departamento de Ciencias y Tecnologı́a de Alimentos, Facultad Tecnol
ogica,
Universidad de Santiago de Chile, Av. Libertador Bernardo O’Higgins No 3363,
Estaci
on Central, Santiago, Chile
e-mail: [email protected]
J. Enrione
Escuela de Nutricion y Dietética, Facultad de Medicina, Universidad de los Andes, Av.
Monse~ nor Alvaro del Portillo 12.455, Las Condes, Santiago, Chile
e-mail: [email protected]
et al. 1996). Chile is the biggest producer of blueberries in South America and the
largest exporter to northern markets (Federacion de Productores de Fruta de Chile
2009; Frederick 2009). Blueberries are small blue fruits from the genus Vaccinium,
with high nutritional value and potential anti-disease effects, but with a short shelf
life of about 20 days under refrigeration conditions at a temperature close to 0 C
(Yommi and Godoy 2010). Due to its short shelf life, conservation methods such as
freezing are required. The effect of freezing rates on fruit quality and their optimal
conditions are strongly dependent on the type of food.
In order to improve the freezing process and to increase the shelf life of this
frozen fruit, the objective of this work was to analyze the effect of freezing rate on
their thermal transition, as part of a comprehensive study of blueberry conservation.
Two cultivars of blueberries (Brigitte and Duke) from the Highbush variety, hand-
harvested in mid-January and mid-February of 2010, respectively, packaged in
plastic boxes and stored under refrigerated conditions were obtained from crop
fields of Hortifrut S.A (Curacavı́, Chile) and generously donated by the Blueberry
Committee of Chile.
Prior to the experimental work, fruits were selected manually, discarding those
that were damaged, those without peduncle and flower residues and any red (unripe)
fruit, in order to determine fruit quality parameters (i.e., moisture and solids soluble
content). Moisture content of fresh blueberries was determined gravimetrically
(AOAC 1995) and was expressed on a % wet basis (g water/100 g wet sample).
Twelve blueberries were dried in an oven (Wiseven, Korea) at 105 C for 24 h until
constant weight was reached. Total soluble solids concentration was determined by
placing 1 mL of blueberry juice obtained from the pulp of five berries, on a portable
refractometer (0–32 Brix, RHB-32ATC). Each analysis was performed in triplicate.
The glass transition temperature (Tg), water crystallization and melting temper-
atures from pulp and cuticle were determined by calorimetry using Differential
Scanning Calorimeter (DSC) (Diamond DSC, Perkin Elmer, USA). All scans were
performed in triplicate following the same protocol. Five blueberries were selected,
the cuticle was removed from each berry using a scalpel, and the pulp was ground
and homogenized. ~20 mg of pulp, cuticle and extracted epicuticular wax were
weighed separately and hermetically sealed in aluminum pans (40 μL). The
performed thermal profile consisted of cooling from 30 C to 70 C at 40 C/min
(intermediate cooling rate) or at 5 C/min (slow cooling rate), isothermal holding at
70 C for 15 min and heating to 95 C at 10 C/min of scanning rate. The glass
transition temperature was considered as the midpoint in the change of heat capacity.
Crystallization and melting water crystals temperature was defined as the onset of
the endothermic peak. The change in the specific heat through the glass transition,
ΔCp (J/g C) and the energy (ΔH (J/g)) associated with crystallization/melting
transitions were also calculated.
Freezing Rate Effect on Thermal Transition of Blueberries 407
In order to achieve the water content of samples after DSC analysis, the pans
were punctured and dried in an oven at 100 C for 7 days. Therefore, enthalpy
values were expressed in dry basis (J/g dry sample).
Statistical significance between data were analyzed by analysis of variance
(ANOVA) and t-test using Prism GraphPad, v4 software. Width of melting water
peak obtained by DSC was obtained by fitting the data with Lorentz equation
(R2 > 0.95) using Origin v8 software.
Fig. 1 Glass transition temperature (Tg0 ) obtained by DSC of thermal transitions for fresh
blueberries (cultivars Duke and Brigitte)
408 P. Dı́az-Calder
on et al.
Fig. 2 Crystal water melting temperatures obtained by DSC in both cultivars analyzed (Duke and
Brigitte). Arrows indicate the onset of temperature transition
accordance with values reported previously for blueberry juice (43.9 C) made
from a pool of cultivars of Vaccinium corymbosum L (Koster 1991) and for wild
blueberry juice (onset at 45.7 C) (Wojtas et al. 1999). The literature has
attributed the Tg value of this fruit to the presence of sugar, specifically glucose
and fructose in a 1:1 ratio (Rizzolo et al. 2003; Roopesh et al. 2010), but this was
not confirmed in the present study.
The onset of exothermic transitions, considered as water crystallization temper-
ature, was 14.0 0.9 C and 16.7 0.7 C for cultivars Duke and Brigitte,
respectively. Meanwhile, the onset of endothermic transitions associated with the
melting temperature of water crystals, were 4.7 1 C and 7.0 1.0 C for
cultivars Duke and Brigitte, respectively (Fig. 2). The significant differences
( p < 0.05) in the onset temperatures between both cultivars (Fig. 2), which was
lower in cv. Brigitte than cv. Duke could be explained by the higher soluble solids
concentration ( Brix) observed in cv. Brigitte. The presence of unfrozen water
indicates that deteriorative reactions are not restricted.
Interestingly, as observed in Fig. 3, the temperatures attributed to crystallization
and melting of water crystals were not significantly different ( p > 0.05) between
freezing rates (slow and intermediate) and between cuticle (78 %, w.b.) and pulp
(68 %, w.b.) for the same cultivar. These results can be explained by the depen-
dency of the temperatures of crystallization and melting of water crystal on the
concentration of soluble solids measured (by colligative properties) rather than their
water content (Nunes et al. 2004).
The width of the peak associated with the melting transition of water crystal was
slightly higher at slow (7.5 1.5 C) than intermediate freezing rate (5.1 1.3 C)
in both cultivars (Fig. 3). This behavior was related with a larger distribution in ice
crystal size, since at slower rate of freezing less nucleation will be induced, which
Freezing Rate Effect on Thermal Transition of Blueberries 409
Fig. 3 Crystal water endotherms obtained by DSC in cv. Duke at different cooling rates. Numbers
indicate the width of endotherm. Similar behavior was observed in cv. Brigitte
Table 1 Normalized enthalpy values (AH) obtained from both cultivars of fresh blueberries
frozen at two different cooling rates and for blueberries stored at 5 C during 22 days
CV. Duke CV. Briggitta
Freezing rate ΔH (J/g dry Lower energy ΔH (J/g dry Lower energy
Fruit ( C/min) sample) released (%) simple) released (%)
Pulp 5 2,722 43 940 48.3
Pulp 40 1,556 486
Cuticle 5 2,118 77 1,604 80.7
Cuticle 40 489 310
Coefficient of Variation (CV) was lower than 10 % in all cases
will produce larger ice crystals and more dislocation (migration) of water molecules
in comparison with faster cooling rates, which would allow producing higher
nucleation and smaller ice crystals.
Table 1 summarizes the values of normalized enthalpy for water crystal melting
obtained at 5 C/min (slow rate) and 40 C/min (intermediate rate or individual
quick freezing (IQF) rate), for pulp and cuticle from both cultivars. By comparing
the ice melting enthalpy, smaller endotherms were observed at higher freezing rate
in both cultivars, showing a significant decrease of 43 % and 48.3 % for pulp and
77 % and 80.7 % for the cuticle for cv. Duke and cv. Brigitte, respectively. It is
important to consider that a higher enthalpy implies higher content of frozen water,
which might affect the internal structure of fruit, especially when the fruit is
thawed. Thus, the cuticle would be less affected by ice formation and by different
freezing rates, which is important when considering the potential protective role of
cuticle on fruit integrity. Wojtas et al. (1999) reported that the formation of large ice
410 P. Dı́az-Calder
on et al.
crystals in the area just below the skin, where the pigment is located, could cause
sufficient structural damage to compromise the compartmentalization of the
anthocyanin in the cell and cause its subsequent “bleeding.” Hence, it is possible
that differences in freezing rate may affect the textural properties of the fruit, where
a fruit slowly frozen may show lower firmness or hardness and higher drip than a
fruit frozen by IQF. Further studies are necessary to confirm this hypothesis.
4 Conclusions
Acknowledgements The authors acknowledge the financial support of Innova Chile-Corfo CT11
PUT-20, CONICYT Project PBCT-PSD-62 and DICYT 339 from VRIDEI-Universidad de San-
tiago de Chile.
References
AOAC (1995) Official methods of analysis of AOAC International. CRC Press, Washington, DC
Federacion de Productores de Fruta de Chile (2009) Breve resumen temporada 2008-2009 en los
principales mercados de destino. Rev Fedefruta 125:30–39
Frederick B (2009) En volumen, Arándanos de Chile más fuerte que nunca. Revista del Campo, El
Mercurio. http://berriesofchilenoticias.blogspot.mx/2009/01/en-volumen-arndanos-de-chile-
ms-fuerte_15.html. Accessed 15 March 2013
Koster KL (1991) Glass formation and desiccation tolerance in seeds. Plant Physiol 96:302–304
Moraga G, Martı́nez-Navarrete N, Chiralt A (2006) Water sorption isotherms and phase transitions
in kiwifruit. J Food Eng 72:147–156
Nicoletta S, Spinardi A, Di Egidio V, Mignani I, Casiraghi E (2008) Evaluation of quality and
nutraceutical ontent of blueberries (Vaccinium corymbosum L.) by near and mid-infrared
spectroscopy. Postharvest Biol Tech 50:31–36
Freezing Rate Effect on Thermal Transition of Blueberries 411
Nunes MCN, Emond JP, Brecht JK (2004) Quality curves for highbush blueberries as a function of
the storage temperature. Small Fruits Rev 3:423–440
Rizzolo A, Nani RC, Viscardi D, Bertolo G, Torreggiani D (2003) Modification of glass transition
temperature through carbohydrates addition and anthocyanin and soluble phenol stability of
frozen blueberry juices. J Food Eng 56:229–231
Roopesh SM, Sablani SS, Tang J, Powers J, Swanson BG (2010) Water sorption and glass
transition temperatures in red raspberry (Rubus idaeus). Thermochim Acta 503–504:90–96
Sá MM, Sereno AM (1994) Glass transitions and state diagrams for typical natural fruits and
vegetables. Thermochim Acta 246:285–297
Wojtas A, Douglas Goff P, Stark R, Carbyn S (1999) The effect of freezing method and frozen
storage conditions on the microstructure of wild blueberries as observed by cold-stage scan-
ning electron microscopy. Scanning 21:334–347
Yommi A, Godoy C (2010) Arándanos, Fisiologı́a y tecnologı́as de postcosecha. http://www.inta.
gov.ar/balcarce/info/documentos/agric/posco/fruyhort/arandano.htm. Accessed 23 March
2010
Yrjo R, Kokini JL, Karel M (1996) Glass transitions in low moisture and frozen foods effects on
shelf life and quality. Food Tech 50:95–108
Glass Transition Temperature of Some Thai
Fruits Using Differential Scanning
Calorimetry: Influence of Annealing
and Sugar Composition
Abbreviations
1 Introduction
2.1 Materials
2.2 Methods
Glass transitions of 20 % sugar (sucrose, trehalose, and maltitol) solutions and the
fruit juices extracted were investigated using a DSC. The DSC used was a
PerkinElmer Pyris 1 DSC equipped with an Intracooler subambient accessory.
Nitrogen gas was used as the purge gas at a flow rate of 20 mL/min during
calibration and measurements. The instrument was calibrated using indium and
cyclohexane. Samples (10–13 mg) were weighed in volatile aluminum pans and
hermetically sealed. An empty aluminum pan was used as a reference sample.
Sugar solutions and fruit samples were determined for Tg0 by the method of
Lowithun and Charoenrein (2009). To determine the Tg0 of the samples in a
non-annealing state, the samples were cooled from 25 to 60 C at a rate of
25 C/min, held for 15 min, and subsequently heated to 10 C at a rate of
5 C/min. For isothermal annealing measurements, the samples were cooled to
60 C and held for 15 min, heated to annealing temperature and held for 30 min;
cooled back to 60 C at a rate of 10 C/min, held for 15 min, and subsequently
reheated to 10 C at a rate of 5 C/min. In this experiment, annealing temperature
was selected as the midpoint of the temperature range between the endpoint of
lower transition and the onset of higher transition at non-annealed conditions. DSC
measurements were carried out in duplicate. PerkinElmer software was used in
obtaining the Tg0 , which was determined as the inflection point of the heat capacity
change at transition temperature.
Extracted fruit juices were determined for sugar composition using an HPLC
(Agilent Model 1100 Series, Agilent Technologies, USA) according to AOAC
Official Method 977.20 (Horwitz 2000). Juice samples were subjected to a Zorbax
Carbohydrate column (5 μm, 4.6 150 mm) (Agilent Technologies, USA) com-
bined with a refractive index detector (RID). The mobile phase consisted of
acetonitrile and water (75:25) with a flow rate of 1 mL/min. The amount of sugar
was calculated from the external standard. The sugar analysis was performed in
duplicate.
416 S. Charoenrein et al.
DSC thermograms of sugar solutions and fruit juices typically show two endother-
mic transitions prior to an ice melting endothermic peak (Fig. 1). This phenomenon
has been reported in carbohydrate solutions and foods that contain freezable water
(Goff and Sahagian 1996; Goff et al. 2003; Rahman 2004; Ohkuma et al. 2008;
Fabra et al. 2009; Sacha and Steven 2009). Goff et al. (2003) proposed that the
lower transition (Tg0 1) represents the glass transition of the bulk phase (Tg0 ),
whereas the higher transition (Tg0 2) is related to the presence of solute inclusions
within ice crystals. Therefore, the Tg0 1 has been assumed to be the Tg0 of the samples
in this study.
The annealing process was achieved at the midpoint of the temperature range
between the endpoint of lower transition and the onset of higher transition at
non-annealed conditions for 30 min. The results of DSC measurement show that
the Tg0 values of sugar solutions in annealed conditions are 40.2 C, 36.7 C,
and 41.3 C for sucrose, trehalose, and maltitol, respectively (Table 1). These
values are close to the ranges previously reported by Roos (1995) who showed that
the Tg0 values of frozen sucrose, trehalose, and maltitol solutions at midpoint are
41, 35, and 42 C, respectively. This demonstrates that the annealing method
in this study can be used to determine the Tg0 values of the samples.
The DSC measurements of both sugar solutions and fruit juices revealed that the
annealing resulted in an increase in the Tg0 1 values, whereas the Tg0 2 was unaffected
(Tables 1 and 2). This is in agreement with Pyne et al. (2003) and Sahagian and
Goff (1994). The annealing process has a potential to allow for the crystallization of
ice in the system; subsequently, the system becomes maximally freeze-
concentrated. The crystallization of the ice decreases the amount of unfrozen
water and also the effect of water plasticization in freeze-concentrated matrix
(Pyne et al. 2003). The results indicate that the annealing process influences the
Tg0 1, which refers to the Tg0 of the bulk system, whereas no effect of annealing was
observed on the Tg0 2 of both sugar and fruit samples. The Tg0 2 overlaps with the
Glass Transition Temperature of Some Thai Fruits Using Differential Scanning. . . 417
Fig. 1 DSC thermograms of rambutan (a), mango (b), and longan (c) juice samples with annealed
and non-annealed condition
418 S. Charoenrein et al.
Table 1 Tg0 1 and Tg0 2 of 20 % sucrose, trehalose, and maltitol solution non annealed and
annealeda condition
Sugar solution (20 % w/w)
Transition temperature ( C) Condition Sucrose Trehalose Maltitol
Tg0 1 Non-anneal 44.6 0.3 40.5 0.2 47.7 0.0
Anneal 40.2 0.5 36.7 0.5 41.3 0.0
Tg0 2 Non-anneal 31.5 0.2 28.3 0.0 32.6 0.8
Anneal 31.7 0.3 28.1 0.5 33.8 0.3
a
Annealed temperatures for sucrose, trehalose, and maltitol were 38, 33, and 39 C,
respectively
Table 2 Tg0 1 and Tg0 2 values of rambutan, mango, and longan in non annealed and annealeda
condition
Transition temperature ( C) Condition Rambutan Mango Longan
Tg0 1 Non-anneal 51.7 0.4 56.1 0.0 56.1 0.0
Anneal 48.6 0.3 51.7 0.0 53.6 0.4
Tg0 2 Non-anneal 36.2 0.0 36.8 0.4 39.4 0.2
Anneal 36.7 0.5 38.1 0.1 39.5 0.1
a
Annealed temperatures for rambutan, mango, and longan were 43, 46, and 47.8 C,
respectively
onset of ice melting and is related to the presence of solute inclusions within ice
crystals (Goff et al. 2003; Pyne et al. 2003). However, in a perspective of freeze
drying, Tg0 2 is probably a more relevant temperature, since structural collapse is
observed a few degrees above Tg0 2 (Pyne et al. 2003; Sacha and Steven 2009).
The HPLC sugar analysis revealed that sucrose, fructose, and glucose are the major
sugar components in longan, mango, and rambutan. Table 3 shows the amount of
sugar found in sample juices. In all fruit samples, sucrose is the main sugar, whereas
the amount of glucose and fructose are similar in each fruit.
Tg0 1 and Tg0 2 values of rambutan, mango, and longan juice in non-annealed and
annealed conditions are shown in Table 2. Although small molecular weight
compounds such as sugar are known as plasticizers that decrease Tg0 of the system
(Roos 1995), the results showed that Tg0 of fruits and total amount of sugars
(Table 3) are not well correlated. This reflects that the Tg0 is independent of the
amount of total sugar.
The Tg0 of fruits is mainly due to its sugar component (Aguilera et al. 1998; Telis
and Sobral 2001). Aguilera et al. (1998) showed that Tg0 of apple tissues coincides
Glass Transition Temperature of Some Thai Fruits Using Differential Scanning. . . 419
with the sugar mixture in juice. Tg0 of neither cell wall nor apple pectin was
observed. Many researchers have revealed that Tg0 of anhydrous dehydrated fruits
correspond to the Tg0 of the main sugar component, such as fructose in grape and
apples (Sa and Sereno 1994; Welti-Chanes et al. 1999). Roos (1995) reported that
the Tg0 values of glucose, fructose, and sucrose are 53, 53, and 41 C,
respectively. The Tg0 of longan (53.6 C), mango (51.7 C), and rambutan
(48.6 C) are in the range of these sugars, which are the fruit compositions.
Although sucrose is the main sugar component in the three fruit samples, their Tg0
values are different. This reflects that Tg0 values of fruits are not affected by only
one predominant sugar composition. The difference among Tg0 values of fruit
samples might be due to the variation of sugar compositions.
The ratio of monosaccharides (glucose and fructose) and sucrose was calculated.
Rambutan had the highest monosaccharides–sucrose ratio, whereas mango and
rambutan have similar ratios. Although longan contains the lowest amount of
total sugars, it has the highest monosaccharides–sucrose ratio, which is coincident
with the lowest Tg0 1 and Tg0 2 values. The high ratio of monosaccharides–sucrose in
the samples resulted in the lowest Tg0 of longan. The results suggest that Tg0 values
of fruits is greater influenced by the monosaccharides–sucrose ratio than by the
amount of total sugars.
4 Conclusion
The annealing process has a potential to allow time for the maximum formation of
ice. The Tg0 values increased in all annealed sugar solutions (sucrose, trehalose, and
maltitol) and fruit samples (longan, mango, and rambutan). The annealing at a
temperature between the endpoint of lower transition and the onset of higher
transition at non-annealed conditions for 30 min resulted in a higher Tg0 . Moreover,
monosaccharides–sucrose ratio played an important role in Tg0 .
References
Aguilera JM, Cuadros TR, Del Valle JM (1998) Differential scanning calorimetry of low-moisture
apple products. Carbohyd Polym 37:79–86
Fabra MJ, Talens P, Moraga G, Martı́nez-Navarrete N (2009) Sorption isotherm and state diagram
of grapefruit as a tool to improve product processing and stability. J Food Eng 93:52–58
Goff HD, Sahagian ME (1996) Glass transitions in aqueous carbohydrate solutions and their
relevance to frozen food stability. Thermochim Acta 280(281):449–464
Goff HD, Verespej E, Jermann D (2003) Glass transitions in frozen sucrose solutions are
influenced by solute inclusions within ice crystals. Thermochim Acta 399:43–55
Horwitz W (ed) (2000) Official methods of analysis of AOAC International, 17th edn. AOAC
International, Gaithersburg
Lowithun N, Charoenrein S (2009) Influence of osmodehydrofreezing with different sugars on the
quality of frozen rambutan. Int J Food Sci Tech 44:2183–2188
Ohkuma C, Kawai K, Viriyarattanasak C, Mahawanich T, Tantratian S, Takai R, Suzuki T (2008)
Glass transition properties of frozen and freeze-dried surimi products: effects of sugar and
moisture on the glass transition temperature. Food Hydrocolloid 22:255–262
Pyne A, Surana R, Suryanarayanan R (2003) Enthalpic relaxation in frozen aqueous trehalose
solutions. Thermochim Acta 405:225–234
Rahman MS (2004) State diagram of date flesh using differential scanning calorimetry (DSC). Int J
Food Prop 7:407–428
Roos YH (1993) Water activity and physical state effects on amorphous food stability. J Food
Process Pres 16:433–447
Roos YH (1995) Phase transition in foods. Academic, New York
Roos Y, Karel M (1991) Amorphous state and delayed Ice formation in sucrose solutions. Int J
Food Sci Tech 26:553–566
Sa MM, Sereno AM (1994) Glass transitions and state diagrams for typical natural fruits and
vegetables. Thermochim Acta 246:285–297
Sacha GA, Steven LN (2009) Thermal analysis of frozen solutions: multiple glass transitions in
amorphous systems. J Pharm Sci-US 98:3397–3405
Sahagian ME, Goff HD (1994) Effect of freezing rate on the thermal, mechanical and physical
aging properties of the glassy state in frozen sucrose solutions. Thermochim Acta 246:271–283
Telis VRN, Sobral PJA (2001) Glass transitions and state diagram for freeze-dried pineapple.
LWT-Food Sci Technol 34:199–205
Welti-Chanes J, Guerrero JA, Barcenas ME, Aguilera JM, Vergara F, Barbosa-Canovas GV
(1999) Glass transition temperature (Tg) and water activity (aw) of dehydrated apple products.
J Food Process Eng 22:91–101
Effect of Emulsifier on Complex Formation
and In Vitro Digestibility of Gelatinized
Potato Starch
Abbreviations
CI Complex index
DG-Pam Decaglycerol palmitic acid ester
GOPOD Glucose oxidase–peroxidase method
OD Optical density
Pam Palmitic acid
Tm Melting temperature
1 Introduction
It is known that amylose and lipid compounds form a helical inclusion complex in
gelatinized starch. The complex formation has attracted much attention because of
impact on retrogradation (Tufvesson et al. 2001), in vivo (Holm et al. 1983) and
in vitro (Guraya et al. 1997; Crowe et al. 2000) digestibility, and rheological
properties (Kaur and Singh 2000; Gelders et al. 2006) of gelatinized starch.
In our previous study (Kawai et al. 2012), effects of the type and amount of fatty
acid (0–2.0 mmol/g-starch of lauric, myristic, palmitic, stearic, oleic, and linoleic
acids) on the complex formation, thermal properties, and in vitro digestibility of
K. Kawai (*)
Department of Biofunctional Science and Technology, Graduate School of Biosphere Science,
Hiroshima University, 1-4-4 Kagamiyama, Higashi-Hiroshima, Hiroshima 739-8528, Japan
e-mail: [email protected]
S. Takato • K. Kajiwara
Department of Bioscience and biotechnology, Tokyo University of Technology,
1404-1 Katakura-cho, Hachioji, Tokyo 192-0982, Japan
gelatinized potato starch–fatty acid mixtures were investigated. From the results, it
was found that the extent of complex formation tended to increase with decrease in the
number of carbons and increase in the number of double bonds in the fatty acid used.
In addition, it was demonstrated that a certain amount of fatty acid reduced the starch
content hydrolyzed at a given condition. This result was related to the extent of
complex formation and its helical order characterized by melting temperature (Tm)
of the complex. On the other hand, it was demonstrated that many types of emulsifier
showed a higher degree of complex formation than that of the fatty acids (Guraya
et al. 1997). In addition, a kind of emulsifiers (e.g., decaglycerol monostearate and
decaglycerol monooleate) reduced greatly the hydrolysis of starch. This will be
because the hydrophilic emulsifiers have greater dispersivity in the gelatinized starch
than the fatty acids. However, the data are not yet sufficient to permit a complete
understanding of the characteristics of complex formation, because the complex
formation and its chemical and physical properties are affected by many factors.
In order to obtain some insight into the subject, decaglycerol palmitic acid ester
(DG-Pam) was employed as a hydrophilic emulsifier, and effect of DG-Pam on
complex formation and in vitro digestibility of gelatinized potato starch was
investigated. In addition, the results were compared with those of palmitic acid
(Pam) reported in our previous study (Kawai et al. 2012).
Potato starch was purchased from Kanto Chemical Co., Inc., Japan. DG-Pam
(HLB ¼ 15) was provided by Mitsubishi-Kagaku Foods Co., Tokyo, Japan.
A potato starch suspension (2 %, w/w) was prepared in a glass vial, and the
starch suspension was stirred for 20 min at approximately 95 C using a magnetic
stirrer. The gelatinized starch (5 g) and DG-Pam (0–30 mg) were hermetically put
into a glass vial, and the mixture was mixed intermittently using a vortex.
The sample (5 g) was mixed with 25 ml of deionized water for 2 min by vortex, and
centrifuged at approximately 1,500 g. The supernatant (0.4 ml) was mixed with
8.6 ml of deionized water and 1 ml of iodine solution (2.0 % (w/w) KI and 1.3 %
(w/w) I2 aqueous mixture). The optical density (OD) values of the sample and a
reference (gelatinized potato starch alone) were measured at 690 nm, and complex
index (CI) was evaluated as follow:
CI (%) ¼ 100 (OD of reference – OD of sample)/OD of reference
All testes were performed in triplicate and the results averaged.
Effect of Emulsifier on Complex Formation and In Vitro Digestibility. . . 423
Effects of DG-Pam and Pam on CI of the samples are shown in Fig. 1. The CI values
increased with the amount of the additives. The CI value of the DG-Pam samples
was much higher than that of the Pam samples. Since DG-Pam is more hydrophilic
than Pam, there will be many opportunities for contact and complex formation with
amylose in the gelatinized potato starch.
DSC thermogram for the DG-Pam sample showed an endothermic peak due to
the melting of complex (data not shown). From the onset point of the peak, Tm of
the DG-Pam complex was evaluated to be 84.5 1.5 C. This value was lower than
that of Pam complex (91.3 1.5 C) (Kawai et al. 2012). The fact that the complex
has a higher Tm reflects that helical length of the complex is longer. This means that
the physical strength of the DG-Pam complex is lower than that of Pam.
424 K. Kawai et al.
CI (%)
40
20
0
0 50 100 150 200 250 300 350
Lipid compound content (mg/g-starch)
40
Effects of PG-Pam and Pam on starch contents hydrolyzed within 20 min and
120 min for the samples are shown in Fig. 2. The starch contents hydrolyzed within
20 min decreased with increase in the amount of the additives. The hydrolyzed
starch content for the DG-Pam sample was always lower than that of the Pam
sample. There was a minor effect of Pam on the hydrolyzed starch content within
120 min. The hydrolyzed starch content for DG-Pam sample, on the other hand,
slightly decreased with increase in the amount of DG-Pam.
Effect of Emulsifier on Complex Formation and In Vitro Digestibility. . . 425
In our previous study (Kawai et al. 2012), it was suggested that the decrease in the
hydrolyzed starch content was affected by the quantity and quality of the complex;
they were characterized by CI and by Tm, respectively. The DG-Pam sample showed
much higher CI but lower Tm of complex than those of the Pam sample. Thus, the
low hydrolyzed starch content for DG-Pam sample will be attributed to the large
amount of complex. In addition, DG-Pam may form complex during the hydrolysis
treatment (Tufvesson et al. 2003). Non-complexed Pam will crystallize in the
system, because the Tm of Pam (63 C) is much higher than the hydrolysis treatment
temperature (37 C). On the other hand, DG-Pam exists as liquid molecules in the
gelatinized potato starch, and thus non-complexed DG-Pam has potentially complex
forming ability during hydrolysis treatment. The hydrolysis of amylopectin and
amylose cause a decrease in the viscosity of system, and then the complex formation
between partially hydrolyzed amylose and DG-Pam will be promoted temporarily as
suggested in a previous study (Heinemann et al. 2005). In addition, partial hydro-
lyzed amylopectin may form a complex with DG-Pam. The short chain branches of
amylopectin are not long enough to form a complex with lipids. Furthermore, steric
hindrance between the branches of amylopectin also prevents the complex forma-
tion. However, the chain length of amylopectin of potato starch is much longer than
other botanical types of starch (Hizukuri 1986; Hoover 1995). In addition, the steric
hindrance between the branches of amylopectin will loosen during the hydrolysis
treatment. In order to understand the relationship between the molecular structure of
complex and the in vitro digestibility of starch, further studies are required.
References
Crowe TC, Seligman SA, Copeland L (2000) Inhibition of enzymic digestion of amylose by free
fatty acids in vitro contributes to resistant starch formation. J Nutr 130:2006
Gelders GG, Goesaert H, Delcour JA (2006) Amylose-lipid complexes as controlled lipid release
agents during starch gelatinization and pasting. J Agric Food Chem 54:1493
Guraya HS, Kadan RS, Champagne ET (1997) Effect of rice starch–lipid complexes on in vitro
digestibility, complexing index, and viscosity. Cereal Chem 74:561
Heinemann C, Zinsli M, Renggli A, Escher F, Conde-Petit B (2005) Influence of amylose-flavor
complexation on build-up and breakdown of starch structures in aqueous food model systems.
LWT-Food Sci Technol 38:885
Hizukuri S (1986) Polymodal distribution of the chain lengths of amylopectins, and its signifi-
cance. Carbohydr Polym 147:342
Holm J, Bj€orck I, Ostrowska S, Eliasson AC, Asp NG, Larsson K, Lundquist I (1983) Digestibility
of amylose–lipid complexes in-vitro and in-vivo. Starch-Starke 35:294
Hoover R (1995) Starch retrogradation. Food Rev Int 11:331
Kaur K, Singh N (2000) Amylose-lipid complex formation during cooking of rice flour. Food
Chem 71:511
Kawai K, Takato S, Sasaki T, Kajiwara K (2012) Complex formation, thermal properties, and
in vitro digestibility of gelatinized starch-fatty acid mixtures. Food Hydrocoll 27:228
Tufvesson F, Skrabanja V, Bj€orck I, Elmståhl HL, Eliasson AC (2001) Digestibility of starch
systems containing amylose-glycerol monopalmitin complexes. LWT-Food Sci Technol
34:131
Tufvesson F, Wahlgren M, Eliasson AC (2003) Formation of amylase-lipid complexes and effects
of temperature treatment. Part 2. Fatty acids. Starch/Stärke 55:138
Influence of Alcohols and Polyols
on the Behavior of Aqueous Solutions
of β-Lactoglobulin at pH 5.5
Abbreviations
1 Introduction
β-lactoglobulin is a small globular protein that contains 162 amino acid residues
and has a molecular weight of 18,362 g mol1. The secondary structure of
β-lactoglobulin comprises a β barrel with eight antiparallel β strands (βA–βH), an
additional β strand (βI) involved in dimer formation, a short α-helix segment, and
three helicoidal turns (Qin et al 1998; Oliveira et al. 2001). Structure patterns for
β-lactoglobulin solution in acetate buffer pH 5.5 obtained by circular dichroism
analysis give 36 % of β-strand profile, 19 % of α-helix elements, 19 % of helicoidal
turns, and 23 % of non-structured regions (Giraldo et al. 2010).
Influence of Alcohols and Polyols on the Behavior of Aqueous Solutions. . . 429
3 Thermal Denaturation
In the two step models, the protein moves from the bulk phase to the fluid interface
in a rapid way, controlled by diffusion, and in the next step the protein is adsorbed
and reorientation and conformational changes of the protein take place (Baeza
et al. 2005; Alahverdjieva et al. 2008). For small globular proteins at low concen-
tration, it is usually considered that adsorption at the air–water interface occurs in
two main steps: diffusion and rearrangement.
Some authors have proposed that the stabilizing effect of sugars and polyols is a
consequence of the increase in the surface tension of water caused by the addition of
these compounds (Lin and Timasheff 1996; Chanasattru et al. 2008). However,
glycerol and other polyols decrease the surface tension of water while increasing
Influence of Alcohols and Polyols on the Behavior of Aqueous Solutions. . . 431
protein stability. Other results with lactalbumin in the presence of polyols and
glucose show that even though the denaturation temperature increases as the
solvent’s surface tension becomes higher, there is no general correlation between
the surface tension of the solvent and the change in the denaturation temperature
(Romero and Albis 2010).
Table 2 shows the results for the surface tension of solutions of 1-butanol,
1,2-butanediol, and 1,2,3,4-butanetetrol in water and in the presence of
β-lactoglobulin at 298.15 K. 1-butanol and 1,2-butanediol induce an important
decrease in the surface tension of water, while 1,2,3,4-butanetetrol, which is the
most hydrophilic of the cosolutes employed, increases the surface tension of water.
For aqueous solutions of β-lactoglobulin, the addition of polyols decreases the
surface tension, with the change being higher as the number of hydroxyl groups
increase.
The evolution of the adsorption process with time was followed at 298.15 K in
terms of the surface pressure (π) defined as the difference between the surface
tension of the protein solution (γ) and the surface tension of the aqueous solvent
(γ 0). Figure 1 shows the behavior of surface pressure of β-lactoglobulin in aqueous
solutions of 1-butanol, 1,2-butanediol, and 1,2,3,4-butanetetrol at 298.15 K. The
addition of polyols increases the surface pressure, with the change being higher as
the number of hydroxyl groups increase (Mendieta 2010).
Experimental data, adjusted by nonlinear regression, show that the adsorption
behavior of lactoglobulin at the air–water interface can be represented as a three-
step process: a very rapid step due to diffusion and two additional steps that follow a
logarithmic behavior. Time lag is not observed.
The most relevant result is that in general a correlation between the change in
surface tension of the aqueous solvent and the denaturation temperature of
β-lactoglobulin is not observed, as some authors have suggested for other proteins.
However, for each of the cosolutes considered, the change in denaturation temper-
ature increases as the solvent surface tension becomes higher.
432 G.I. Giraldo and C.M. Romero
Table 3 summarizes the values of activation energy obtained for the solutions of
β-lactoglobulin in water and in the aqueous alcohol solutions.
The results show an increase in the activation energy with respect to its value in
water suggesting a smaller mobility of the protein in the presence of the selected
cosolutes. The activation energy increases with concentration being the largest incre-
ment due to butanol. The increase of activation energy follows the order:
1-butanol > 1,2-butanediol > 1,2,3-butanetriol ffi 1,2,3,4-butanetetrol. The observed
change in the case of butanetetrol is almost negligible and the energy value remains
nearly constant as concentration increases (Romero et al. 2008).
434 G.I. Giraldo and C.M. Romero
6 Conclusions
The above is a very brief survey of what we have found studying the thermody-
namic behavior of β-lactoglobulin in the attempt to comprehend protein stability.
The denaturing effect induced by alcohols and polyols depends clearly on the
number of OH groups, as well on concentration. The behavior of thermal denatur-
ation of aqueous solutions of β-lactoglobulin in alcohols and polyols of four carbon
atoms shows that 1-butanol produces the largest decrease in Tm as alcohol concen-
tration increases. The effect diminishes as the number of hydroxyl groups increase
and for 1,2,3,4-butanetetrol, which is a typical hydrophilic solute, an increase in
temperature is observed as concentration increases. In the case of butanediols, the
decrease in Tm is greater for 1,2-butanediol, which has a larger hydrophobic domain
due to the vicinal position of the OH groups, and with 1,2,4-butanetriol a negligible
effect is observed.
The number and position of OH groups is also clearly reflected on the surface
and viscometric behavior of β-lactoglobulin in aqueous solution. The decrease of
surface tension with alcohol mole fraction becomes less pronounced as the number
of OH groups increases, while for 1,2,3,4-butanetetrol the behavior changes and the
solution surface tension becomes larger than the corresponding value for water. The
largest increase in activation energy is due to butanol, and diminishes as the number
of OH groups increase.
Experimental results reveal that thermal stability, surface tension and viscomet-
ric properties of aqueous solutions of β-lactoglobulin are clearly influenced by the
presence of alcohols and polyols. The observed behavior follows a similar trend to
that observed for the properties of these alcohols in water, suggesting that their
effect on β-lactoglobulin stability is a consequence of the alcohol–water interaction
and the induced modification in solvent structure, rather than alcohol–protein
interactions.
Acknowledgments The authors would like to acknowledge the support given by Universidad
Nacional de Colombia, campus Manizales and Bogotá and to ISOPOW which accepted this work
to be published as part of the book. We also want to express our gratitude to those colleagues and
students who, directly or indirectly, contributed to this work.
References
Alahverdjieva VS, Grigoriev DO, Ferri JK, Fainerman VB, Aksenenko EV (2008) Adsorption
behaviour of hen egg-white lysozyme at the air/water interface. Colloids Surf A: Physicochem
Eng Aspects 323:167–174
Apenten RKO, Khokhar S, Galani D (2002) Stability parameters for β-lactoglobulin thermal
dissociation and unfolding in phosphate buffer at pH 7.0. Food Hydrocoll 16:95–103
Baeza R, Sanchez CC, Pilosof AM, Patino JM (2005) Interactions of polysaccharides with
β-lactoglobulin adsorbed films at the air–water interface. Food Hydrocoll 19:239–248
Influence of Alcohols and Polyols on the Behavior of Aqueous Solutions. . . 435
Boye JI, Ismail AA, Alli I (1996) Effects of physicochemical factors on the secondary structure of
beta-lactoglobulin. J Dairy Res 63:97–109
Chanasattru W, Decker EA, McClements J (2008) Impact of cosolvents (polyols) on globular
protein functionality: Ultrasonic velocity, density, surface tension and solubility study. Food
Hydrocoll 22:1475–1484
D’Alfonso L, Collini M, Baldini G (2002) Does β-lactoglobulin denaturation occur via an
intermediate state? Biochemistry 41:326–333
Giraldo GI, Lozano JM, Romero CM (2010) Efecto de alcoholes y polioles sobre la estabilidad
estructural de la B-lactoglobulina en solucion acuosa. Rev Colomb Quim 39:7–18
Gomez JM, Henestrosa VP, Sánchez C (2008) Patino J M. Food Hydrocoll 22:1105–1116
Griko YV, Freire E, Privalov PL (1994) Energetics of the alpha-lactalbumin states: a calorimetric
and statistical thermodynamic study. Biochemistry 33:1889–1899
Joyce IB, Ashraf AI, Inteaz A (1996) Effect of physicochemical factors on the secondary structure
of β-Lactoglobulin. J Dairy Res 63:97–109
Lin TY, Timasheff S (1996) On the role of surface tension in the stabilization of globular proteins.
Protein Sci 5:372–381
Madgassi S (1996) Surface activity of proteins: chemical and physicochemical modifications.
Marcel Dekker, Inc., New York
Mendieta N (2010) Influencia del 1-butanol, 1,2-butanodiol y eritritol sobre la estabilidad térmica
de la β-lactoglobulina. Chemistry Final Work, Universidad Nacional de Colombia, Bogota
Oliveira KM, Valente-Mesquita VL, Botelho MM, Sawyer L, Ferreira ST, Polikarpov I (2001)
Crystal structures of bovine beta-lactoglobulin in the orthorhombic space group C222(1).
Structural differences between genetic variants A and B and features of the Tanford transition.
Eur J Biochem 268:477–483
Qin BY, Bewley MC, Creamer K, Baker HM, Baker EN, Jameson GB (1998) Structural basis of
the Tanford transition of bovine beta-lactoglobulin. Biochemistry 37:14014–14023
Rao CS, Damodaran S (2000) Is surface pressure a measure of interfacial water activity? Evidence
from protein adsorption behavior at interfaces. Langmuir 16:9468–77
Romero CM, Albis A (2010) Influence of polyols and glucose on the surface tension of bovine
α-lactalbumin in aqueous solution. J Solution Chem 39:1865–1876
Romero CM, Lozano JM, Giraldo GI, Sancho J (2007) Thermal stability of β-lactoglobulin in the
presence of aqueous solution of alcohols and polyols. Int J Biol Macromol 40:423–428
Romero CM, Lozano JM, Giraldo GI (2008) Effect of temperature on partial molar volumes and
viscosities of dilute aqueous solutions of 1-butanol, 1,2-butanediol, 1,4-butanediol, 1,2,4-
butanetriol, and butanetetrol. Phys Chem Liq 46:78–85
Ruı́z-Henestrosa VP, Sánchez CC, Escobar MM, Jiménez JP, Rodrı́guez FM, Patino MR (2007)
Interfacial and foaming characteristics of soy globulins as a function of pH and ionic strength.
Colloids Surf A: Physicochem Eng Aspects 309:202–215
Effects of Protein Conformational
Modifications, Enthalpy Relaxation,
and Interaction with Water on the Solubility
of Milk Protein Concentrate Powder
Abbreviations
aw Water activity
CP Cross polarization
CPMAS Cross polarization magic angle spinning
CSA Chemical shift anisotropy
DP Direct polarization
MAS Magic angle spinning
MPC Milk protein concentrate
NMR Nuclear magnetic resonance
1 Introduction
Dairy powder is one of the most widely used dry food ingredients because of its high
nutritional value and taste. Being in the dry state it is easy to handle and requires less
transportation and storage cost. Despite many years of development and research, a
gradual loss of milk proteins solubility and functional properties on storage is still
Part of this content comes from the thesis (E. Haque 2011).
E. Haque (*)
School of Applied Sciences, RMIT University, City Campus, 459-469 Swanston
St. Building 39, Level 4, Room 8, Melbourne, VIC 3000, Australia
e-mail: [email protected]
B.R. Bhandari
School of Agriculture and Food Sciences, The University of Queensland,
Brisbane, QLD 4072, Australia
e-mail: [email protected]
remains a big issue for the dairy powder producing industries. There is a lack of
fundamental understanding of changes in protein structure at a molecular level
during the processing of milk to produce milk powder and also during storage due
to moisture uptake and equilibration. Protein stability is usually associated with an
ability to maintain primary and secondary structures. Transformation of proteins
native (folded) structure to denatured (unfolded) structure alters its functionality
(Fasman 1989). Besides protein denaturation there are many other factors that may
cause physicochemical changes in high protein powder which lead to decrease in
solubility and other functional properties (Thomas et al. 2004). This review will
highlight the underlying mechanisms of the gradual loss of functional properties,
especially solubility, of high protein dairy powders and to discover a chemical
and/or technological means to improve this loss of solubility upon storage.
Milk proteins are one of the most widely consumed food proteins. Normal bovine
milk contains roughly 30–35 g protein/L consisting of two main categories of
protein, i.e., caseins and whey proteins (Fox 1982). The caseins, which account
for 76–86 % of the total protein, can be represented by four gene products: αs1-
caseins, αs2-caseins, β-caseins, and κ-caseins (Risso et al. 2007). The major whey
proteins representative gene products are β-lactoglobulins, α-lactalbumins, immu-
noglobulins, and serum albumin, in order of decreasing amounts.
2.2 Casein
Casein is the major protein fraction in bovine milk, occurring in the aqueous phase
of milk as stable colloidal aggregates known as casein micelles. All the caseins are
relatively small molecules ~20–24,000 Da, and are insoluble at their isoelectric
points, pH ¼ 4.5–4.9 which has a major influence on their functional properties
(Fox 1988). All caseins are phosphorylated to a variable extent and usually the
phosphate is esterified to the polypeptides as monoesters of serine. αs1-caseins
usually contain 8 mol P/mol protein, whereas αs2-caseins consist of four proteins
with a common polypeptide but 10–13 mol P/mol protein. β-casein usually contains
5 mol P/mol protein though some genetic variants contain only four phosphate
residues. Most molecules of κ-casein contain only one mole P per mole protein but
studies suggest that some minor κ-casein may contain 2–3 phosphate groups
(Vreeman et al. 1986).
Effects of Protein Conformational Modifications, Enthalpy Relaxation. . . 439
The caseins are strongly hydrophobic in order β > κ > αs1 > αs2. However, the
hydrophobic and polar or charged residues are not uniformly distributed throughout
the sequences.
Studies using circular dichroism and optical rotary dispersion suggest that the
caseins have little secondary structure. However, theoretical considerations suggest
that αs1-casein probably possesses some β-sheet structure and β-turns. Up to 10 %
of the β-casein molecule may exist as α-helices (sequence 97–103 and 138–146),
13 % as β-sheet (sequence 52–60, 77–87, 187–195), and 77 % as unordered
structures (Creamer et al. 1981). κ appears to be the most highly structured of the
caseins with 23 % of the sequence as α-helices, 31 % as β-sheet, and 24 % as β-turns
(Fox 1988).
Due to the high content of phosphoseryl residues, αs1, αs2, and β-caseins
strongly bind polyvalent cations mainly Ca2+ but also Zn2+ which are nutritionally
important, leading to charge neutralization, aggregation, and eventually precipita-
tion. Many of the technologically important properties of milk, such as heat
stability, rennet coagulation, and the syneresis properties of rennet gels, are strongly
influenced by Ca2+ in the manufacture of certain dairy products (Fox 1988).
All the caseins contain high numbers of proline residues which are fairly
uniformly distributed throughout the sequences. Prolyl residues interrupt α-helical
and β-sheet structures but give the casein a type of stable poly-L-proline helical
structure. The caseins are therefore, relatively small, amphipathic, randomly coiled,
unstructured, open molecules. These features of caseins markedly influence their
functional properties.
In normal milk most (~95 %) of the casein exists as coarse colloidal particles called
micelles with a molecular weight of ~108. Casein micelles are roughly spherical,
fairly swollen particles with a diameter ranging from 80 to 680 nm (Risso
et al. 2007). The micelles consist of (dry basis) ~6 % of small ions, principally
calcium, phosphate, magnesium, and citrate, referred to collectively as colloidal
calcium phosphate (CCP). In milk, the micelles are highly hydrated, typically ~2 g
H2O/g protein (Fox 1988). The technological properties of milk are very strongly
influenced by the stability of casein micelles, the structure of which has been the
subject of considerable research. According to classical theory (Horne 2006;
Slattery and Evard 1973) it is believed that the micelles are composed of spherical
submicelles, 10–15 nm in diameter, and have a porous structure. κ which is the
principal micelle-stabilizing factor is located predominantly at the surface although
other caseins can also be present at the surface. An assumption is that the presence
of κ-casein in casein micelle outer layer limits their growth and determines the size.
In the last 20 years or more it has been suggested by different authors based on
different technical approaches that higher proportion of κ-casein corresponds to the
formation of smaller and more solvated micelles (Anema and Creamer 1993;
GutierrezAdan et al. 1996). The submicelles are linked together by colloidal
440 E. Haque and B.R. Bhandari
Fig. 1 Field emission scanning electron micrographs of aged milk protein concentrate powder
particles after 2-month storage at 20 C and water activity of 0.23. (a) spray-dried powder particle;
(b, c) hydrated powder particle after rehydration for 10 min; (d) details of the surface of the
rehydrated powder particle (50,000) observed in panel (c). Adapted from Mimouni et al. 2010a
Whey Proteins represent ~20 % of nitrogen in bovine milk. The whey protein can be
fractionated into two principal groups, an insoluble lactoglobulin fraction and a
soluble lactalbumin fraction. The lactalbumin fraction contains β-lactoglobulin,
α-lactalbumin, bovine serum albumin (BSA), and several minor proteins, the
most significant of which are lactotransferrin, serotransferrin, and β2-microglobulin
(Fox 1988).
Many milk powder including MPC produced via spray drying remain in amorphous
glassy and thermodynamically nonequilibrium state (Bhandari and Hartel 2005)
Below glass transition temperature (Tg) amorphous glassy system like MPC would
approach a thermodynamically more stable state upon ageing with a reduction in
molecular mobility. This kind of change is known as enthalpy relaxation or
444 E. Haque and B.R. Bhandari
50
40
% Relative proportion
30
20
10
0
α-helix β-sheet β-turn
2 weeks 6 weeks 7 weeks MPC raw
structural relaxation or β-relaxation (Liu et al. 2006; Correia et al. 2009; Zhou and
Labuza 2007). In a glass, translational and rotational mobility of large molecules
are restricted and only rotation and vibration of small groups of atoms still exists.
These motion do not involve surrounding atoms/molecules and mainly local (Liu
et al. 2006). Thus, β-relaxation is caused by the specific chemical groups such as
side groups branched on a polymer chain. It could affect the physical properties of
amorphous food materials such as MPC powder during storage. A number of
factors can affect the enthalpy relaxation such as previous thermal history, ageing
period and water activity etc (Haque et al. 2006; Descamps et al. 2009).
In our recent study, kinetics of enthalpy relaxation of milk protein concentrate
(MPC) powder upon short- and long-term (up to 48 days) storage at 25 C and a
range of aw values (0–0.85) were studied by differential scanning calorimetry
(DSC). Long-term storage showed that enthalpy relaxation depends on both storage
period and water activity. The enthalpy value was much less for lower moisture
content (mc) (aw
0.23, mc
5.5 %) than for higher mc (aw 0.45, mc 8 %)
samples for a particular storage period (Fig. 3). Also for storage under defined
relative humidity conditions, the enthalpic region moved to higher temperature and
the overshoot value increased slowly with increasing storage period at least up to
48 days as shown in Fig. 3. The results suggest that the presence of more water
molecules in close proximity to the protein surface facilitates kinetic unfreezing and
subsequent motion of molecular segments of protein molecules towards thermody-
namic equilibrium (Haque et al. 2012).
Effects of Protein Conformational Modifications, Enthalpy Relaxation. . . 445
b
4.0
3.6
Heat flow (mW)
3.2
2.8
2.4
2.0
1.6
50 55 60 65 70 75 80 85
Temperature C
aw 0 aw 0.23 aw 0.45
aw 0.75 aw 0.85
c 4.0
3.6
Heat flow (mW)
3.2
2.8
2.4
2.0
1.6
50 55 60 65 70 75 80 85
Temperature C
aw 0 aw 0.23 aw 0.45
aw 0.75 aw 0.85
446 E. Haque and B.R. Bhandari
Water plays an important role with respect to the properties of food systems. The
water in food influences the physical or textural characteristics of the product as
well as its chemical stability (Bell and Labuza 2000). The quality of preserved
foods can depend on the moisture content, moisture migration, or moisture uptake
during storage. The extent of sorption or desorption from food products depends
on the vapor pressure of water present in the food system and in the surroundings
(Basu et al. 2006). The water in food has been classified as “bound” water
(or non-freezing water) and “retention” water (free water and capillary water).
Bound water is claimed to correspond to structural water while water retain in a
protein matrix like casein micelles or casein gels can be designated as “retention”
water (Hardy et al. 2002). However, these static descriptions as “bound” or
“retained” can be misleading, as water molecules can be exchange between
different sites rapidly even in a matrix that is below its glass transition temper-
ature. Measurement of water properties using advanced instrumental techniques
such as NMR relaxometry can reflect these exchange processes. Several types of
interaction between water and food solids may occur at the molecular level. These
include Columbic interactions between charged groups, London-van der Waals
forces, steric effects, solution effects, capillary condensation, plasticizing of
macromolecular structure, and multiple effect between various components
(Van der Berg and Bruin 1981). Liquid water is composed of transient
H-bonded clusters of disturbed ice-like structures at 0 C where there is 10 %
free (non-bonded) hydroxyl (OH) groups, and this increases to 35–40 % at 100 C
(Kinsella and Fox 1986). There is strong cooperative H bonding among water
molecules in bulk water with assemblies of for example six-membered hexagonal
rings in rapid exchange with free water molecules (Luck 1981). The charged
groups on proteins alter the normal structure of the water in their vicinity, and
water become oriented in layers around these groups (Kinsella and Fox 1986).
Polar groups (hydroxyl, carboxyl, and amide) attract water via H bonding. Apolar
groups and hydrophobic regions induce local clathrate or crystalline
hydrate structures in their immediate vicinity. The innermost molecules are
polarized and immobilized with an intermediate region of less structured mole-
cules and an outer shell of normally tetrahedrally oriented water molecules
(Kinsella and Fox 1986).
A generalized sequence is found for the progressing interaction of water with
protein with increasing equilibrium relative humidity (ERH) or water activity
(aw). As the water activity increases additional water molecules progressively
associated with protein in less structured layers and as capillary-held water. These
water molecules show a gradual continuum in physical properties from highly
structured initial layers to normal bulk liquid water (Kinsella and Fox 1986). At
low aw (0.2), water molecules bind to specific sites on the protein, and at the BET
monolayer, where there is an average of one molecule of water per polar group.
As vapor pressure is increased (aw 0.2–0.7), the nonionic groups and the
Effects of Protein Conformational Modifications, Enthalpy Relaxation. . . 447
Fig. 4 T2 distribution 5
profile of MPC samples aw 0.0
stored at room temperature aw 0.23
(24 1 C) for 3 weeks 4 aw 0.45
period. X-axis is in
logarithmic scale. Adapted aw 0.69
aw 0.85
Intensity (a.u.)
from Haque et al. 2010
3
0
0 20 40 60 80 100
T2, normalized
accessible peptide bonds are progressively saturated. These peptide bonds may
account over 50 % of the water bound at aw 0.6–0.8 (Kinsella and Fox 1986).
Above aw 0.7, multilayer water with negligible restricted mobility is H bonded to
the previous layer, and eventually bulk water becomes entrapped in voids and
crevices.
In our recent study water status (interaction of water with the protein molecule/
surface) of MPC powder stored at varying aw for up to 12 weeks was determined by
measuring the transverse relaxation time (T2) with proton nuclear magnetic reso-
nance (1H NMR). The NMR-T2 results indicated the presence of three distinct
populations of water molecules (differing in their interaction with and distance
from protein surfaces) and that the proton signal intensity and T2 values of proton
fractions varied with storage condition (humidity, temperature) and ageing (Fig. 4)
(Haque et al. 2010). As shown in Fig. 4, with increasing aw, T2 value for water
population close to the protein surface decrease up to aw 0.45 (mc 8.2 %) and then
increase for higher aw samples, indicating higher interaction with protein surface
for lower aw samples. This study also show that with longer storage period water
population close to the protein surface (short T2 fraction) increased with time,
indicating enhanced water–protein interactions (Haque et al. 2010).
In another study using solid state 13C NMR, we probed the changes in molecular
structure and dynamics of proteins in MPC with varying moisture content
(5.5–16.5 % w/w) over a storage period. The results indicate a slight higher rigidity
of molecular domains of protein molecules of non-aged MPC compared to that of
the long aged (14 weeks at 25 C) MPC (Fig. 5). It could be suggested from this
observation that long-term storage at high relative humidity (RH) may reduce
rigidity of the molecular domains due to interaction with water rather than
short-term storage at high RH (Haque et al. 2015).
448 E. Haque and B.R. Bhandari
Fig. 5 CPMAS spectra of MPC (non-aged and stored at aw 0.85 for 14 weeks at 25 C temper-
ature) for a 500 μs contact time. *spinning side band. Adapted from Haque et al. 2015
4 Conclusion
In this review the underlying molecular mechanism associated with protein dena-
turation and loss of solubility in high protein dairy powder system is presented.
Protein conformational modifications, kinetics of water–protein interactions,
enthalpy or structural relaxation, change in structural and dynamic feature of
protein molecules are the major areas of investigations reviewed. This review
provides a collective knowledge in this area of research and help to discover a
chemical and/or technological means to improve the loss of solubility of high
protein dairy powder upon storage.
Acknowledgement This research was supported by Dairy Innovation Australia Limited and the
Australian Research Council (ARC) through ARC Linkage Grant No LP0669191 and ARC
Linkage Infrastructure Grants No LE0775684 and LE0668517.
References
Anema SG, Creamer LK (1993) Effect of the a and B variants of both alpha(Si)- and kappa-casein
on bovine casein micelle solvation and kappa-casein content. J Dairy Res 60(4):505–516
Anema SG, Pinder DN, Hunter RJ, Hemar Y (2006) Effects of storage temperature on the
solubility of milk protein concentrate (MPC85). Food Hydrocoll 20(2–3):386–393
Arakawa T, Prestrelski SJ, Kenney WC, Carpenter JF (2001) Factors affecting short-term and
long-term stabilities of proteins. Adv Drug Deliv Rev 46(1–3):307–326
Arrondo JLR, Goni FM (1999) Structure and dynamics of membrane proteins as studied by
infrared spectroscopy. Progress Biophys Mol Biol 72(4):367–405
Effects of Protein Conformational Modifications, Enthalpy Relaxation. . . 449
Basu S, Shivhare US, Mujumdar AS (2006) Models for sorption isotherms for foods: a review (vol
24, pg 917, 2006). Dry Technol 24(12):1705–1705
Bell LN, Labuza TP (2000) Moisture sorption: practical aspects of isotherm measurement and use,
2nd edn. American Association of Cereal Chemists, St. Paul, MN
Bhandari BR, Hartel RW (2005) Phase transitions during food powder production and powder
stability. In: Encapsulated and powdered foods, vol 146. Crc Press-Taylor & Francis Group,
Boca Raton, pp 261–292
Bhandari BR, Howes T (1999) Implication of glass transition for the drying and stability of dried
foods. J Food Eng 40(1–2):71–79
Correia NT, Diogo HP, Ramos JJM (2009) Slow molecular mobility in the amorphous solid state
of fructose: fragility and aging. J Food Sci 74(9):E526–E533
Creamer LK, Richardson T, Parry DAD (1981) Secondary structure of bovine alpha s1- and beta-
casein in solution. Arch Biochem Biophys 211(2):689–696
De Block J, Merchiers M, Van Renterghem R (1998) Capillary electrophoresis of the whey protein
fraction of milk powders: a possible method for monitoring storage conditions. Int Dairy J 8
(9):787–792
Descamps N, Palzer S, Zuercher U (2009) The amorphous state of spray-dried maltodextrin: sub-
sub-Tg enthalpy relaxation and impact of temperature and water annealing. Carbohydr Res 344
(1):85–90
Fasman GD (1989) Prediction of protein structure and the principles of protein conformation.
Plenum Press, New York
Fox PF (1982) Developments in dairy chemistry. Elsevier Applied Science Publishers, London
Fox PF (1988) The milk protein system. In: Fox PF (ed) Developments in dairy chemistry-4-
functional milk proteins. Elsevier Applied Science, London, pp 1–53
GutierrezAdan A, Maga EA, Meade H, Shoemaker CF, Medrano JF, Anderson GB, Murray JD
(1996) Alterations of the physical characteristics of milk from transgenic mice producing
bovine kappa-casein. J Dairy Sci 79(5):791–799
Haque E (2011) Ageing induced solubility loss of milk protein concentrate powder: effect of
protein conformational modifications, enthalpy relaxation and interaction with water (doctoral
dissertation). The University of Queensland, Brisbane
Haque MK, Kawai K, Suzuki T (2006) Glass transition and enthalpy relaxation of amorphous
lactose glass. Carbohydr Res 341(11):1884–1889
Haque E, Bhandari BR, Gidley MJ, Deeth HC, Moller SM, Whittaker AK (2010) Protein
conformational modifications and kinetics of water-protein interactions in milk protein con-
centrate powder upon aging: effect on solubility. J Agric Food Chem 58(13):7748–7755
Haque E, Bhandari BR, Gidley MJ, Deeth HC, Whittaker AK (2011) Ageing-induced solubility
loss in milk protein concentrate powder: effect of protein conformational modifications and
interactions with water. J Sci Food Agric 91(14):2576–2581
Haque E, Whittaker AK, Gidley MJ, Deeth HC, Fibrianto K, Bhandari BR (2012) Kinetics of
enthalpy relaxation of milk protein concentrate powder upon ageing and its effect on solubility.
Food Chem 134(3):1368–1373. doi:10.1016/j.foodchem.2012.03.034
Haque E, Bhandari BR, Gidley MJ, Deeth HC, Whittaker AK (2015) Change in molecular
structure and dynamics of protein in milk protein concentrate powder upon ageing by solid-
state carbon NMR. Food Hydrocoll 44:66–70
Hardy J, Scher J, Banon S (2002) Water activity and hydration of dairy powders. Lait 82
(4):441–452
Havea P (2006) Protein interactions in milk protein concentrate powders. Int Dairy J 16
(5):415–422
Horne D (2006) Casein micelle structure: models and muddles. Curr Opin Collied Interface Sci
11:148–153
Jouppila K, Kansikas J, Roos YH (1997) Glass transition, water plasticization, and lactose
crystallization in skim milk powder. J Dairy Sci 80(12):3152–3160
450 E. Haque and B.R. Bhandari
Abbreviations
1 Introduction
The lifestyle of consumers along with the desire to consume naturally healthy
beneficial products has increased the demand for production of fresh produce.
Minimally processed fruits and vegetables deteriorate over time, either by means
of microbial activity or by physicochemical reactions with the food environment.
Therefore, the food industry is investigating ways to protect the integrity and
quality of their products by developing better packaging technologies that can
maintain or improve the quality attributes of fresh produce, as well as extend
their shelf life and ensure the microbiological stability (Rojas-Graü et al. 2007;
Lin and Zhao 2007).
The development and characterization of edible films (EFs) is becoming an
innovative field in the food preservation area due to the wide variety of delivery
applications in the food and packaging industry, particularly the selective function-
ality given to fresh fruits and vegetables (Janjarasskul and Krochta 2010).
The main function of EFs focuses on their ability to act as a complement to improve
food quality and shelf life extension. Functional properties of EFs include the
regulation of moisture transfer, reduction of gas transport (O2, CO2, C2H4), reten-
tion of aroma and flavor components, improvement of mechanical properties, and
support of foods’ structural integrity. EFs may also serve as a vehicle for additives
(flavors, colorants, antioxidants, antimicrobials in a food system), where in addition
to being sensory acceptable, they can improve food quality, safety, and shelf life.
However, each functional property of EFs depends on the type of material used,
method of production and application. These attributes enable the success of
storage and preservation techniques that will determine consumer preference
(Tharanathan 2003).
sorbitol, glycerol, and polyethylene glycol, followed by oils and waxes (McHugh
and Krochta 1994). Beeswax (BW) used as a nanoemulsion may enhance hydro-
phobic properties of EFs, while some nanoparticles may provide useful properties
to active EFs, such as oxygen scavenging, enzymes immobilization, antimicrobial
effect, or indication of degradation of certain quality factors (de Azeredo 2009).
The addition of surfactants and emulsifiers reduces the rate of moisture loss,
stabilizing the dispersed polymer phase in the suspension before casting to produce
a dry film, or before coating food surfaces. Most emulsifiers are derivatives of
glycerol and fatty acids such as polysorbate (Tween), which have the ability to
improve adhesion at the interface between the food and the edible coating
(Janjarasskul and Krochta 2010). Nanocomposites are promising materials due to
their contribution to active EFs reinforcement at low concentrations, which pro-
duces enhanced mechanical, barrier, and thermal properties (de Azeredo 2009).
One of the most important functions arising from the application of EFs is their
use as an antimicrobial support in order to ensure microbiological stability and
longer shelf life of foods (Quezada-Gallo 2009).
A variety of polysaccharides and their derivatives have been tested for potential use in
EFs production due to their abundance, low cost, and ease of use. Commercially, the
most important source is corn starch. Amylopectin is the major component of corn
starch granules, typically accounting for up to 70–80 % (w/w), although waxy corn
starch (WCS) can reach levels of 98–99 % (w/w). Some properties of WCS include
high water solubility, ease of gelatinization, being less prone to retrogradation, as well
as being inexpensive and generally available (Romero-Bastida et al. 2005). However,
the main limitation on the use of native corn starch in EFs is its hydrophilic nature.
Starch chemical modifications are a viable alternative to improve their properties,
which are intended to change the characteristics of the starch granule to meet a desired
function. Chemical modification is used to introduce functional groups to the starch
molecule via esterification, cross-linking reactions, treatment with strong acids, enzy-
matic hydrolysis, or oxidation (Liu and Han 2005; Singh et al. 2007).
EFs based on modified corn starch are clear, flexible, and transparent and
provide a good barrier against gases such as CO2 and O2, despite exhibiting poor
mechanical properties. Due to their hydrophilic nature, they offer little resistance to
migration of water, leading to high water vapor permeability (WVP); values were
compared with commercial packaging films (Kim et al. 2002; Garcı́a et al. 2009).
The main cause of fresh produce deterioration is microbial growth on the surface of
the food. Growth of fungi and bacteria during storage and distribution can drastically
reduce the quality of food and compromise its safety. Reduction in water activity
454 T. Arredondo-Ochoa et al.
(aw) levels and protection with moisture-proof packaging are methods commonly
used to prevent deterioration. However, reorganization of water inside the package
due to temperature changes can lead to condensation of moisture on the food’s
surface, increasing the possibility of microbial growth (Quezada-Gallo 2009).
The incorporation of antimicrobials has been the main purpose for the new gener-
ation of active EFs due to their biodegradability, reduced moisture loss, and especially
the prevention of microbial growth (Suppakul et al. 2003; Ozdemir and Floros 2003).
Antimicrobial compounds isolated from natural plant sources incorporated into
EFs include spice extracts such as cinnamon, pepper, thyme, clove, rosemary, and
oregano. In addition, other types of natural substances used against the action of
fungi and bacteria are nisin, lauric arginate ester (LAE), and natamycin (Brody
et al. 2001). Antimicrobial EFs could extend the shelf life of foods (fresh produce,
meat, cheese slices, etc.) and improve food safety, where slow migration of
antimicrobial agents to the product surface improves efficiency and helps maintain
high concentration of these agents (Zhou et al. 2010).
The objective of this study was to design and characterize active edible films
based on modified waxy corn starch, incorporating nano-emulsified beeswax as a
hydrophobic agent.
Modified corn starch and non-crystallizable sorbitol (70 % w/v) was provided by
CPIngredienes (San Juan del Rı́o, México). Tween 80 and BW were purchased
from Sigma (St. Louis, MO, USA). Potato dextrose broth (PDB) and agar (PDA)
were acquired from Bioxon (Cuautitlan, Mexico); rose bengal chloramphenicol
(RBC) agar was provided by Difco; LAE was provided by Vedeqsa (Lamirsa
Group, Valencia, Spain); natamycin was purchased from EcoBio (Columbus, OH,
USA). All other chemicals used were of analytical grade or better.
The MLC of two antimicrobials against B. cinerea ATCC 1481 were evaluated
based on preliminary studies. LAE was tested at 100, 150, 200 and 250 ppm, while
natamycin concentrations were 5, 10, 15, and 20 ppm. Antimicrobial effect was
carried in test tubes containing PDB plus the antimicrobial concentration, and
1 106 spores/mL. One hundred μl of each test tube contents were smeared on
RBC agar, to selectively monitor spore germination. Fungal growth was monitored
for 5 days, and the MLC found for each antimicrobial was used as a base for the
antimicrobial incorporation into the EF formulation that presented the best charac-
teristics. Antimicrobial activity may decrease due to interactions with FS, and thus
concentrations tested when incorporated into EFs were MLC. Active EFs were
produced following the same procedure, incorporating the antimicrobials before
high speed mixing. Dried EFs were cut into 4 4 cm squares and placed in contact
with PDA plates inoculated with 1 106 spores/mL followed by incubation for
5–10 days at 28 C.
456 T. Arredondo-Ochoa et al.
The best EF was used to coat fresh strawberries purchased from a local market.
Following Garcı́a et al. (1998), five strawberries showing uniform size, without
mechanical damage or visible fungal infections were selected. The fruits were
sanitized by immersion in a 250 ppm sodium hypochlorite solution for 1 min, and
dried in a laminar flow cabinet for 30 min, then immersed for 1 min in the FS and
dried for 2 h. A control was submerged with sterile distilled water. The samples
were stored at 4 C and 98 % RH and weight loss was recorded every day up to
9 days of storage. The experiment was performed in duplicate.
Table 1 Effect of nano-emulsified beeswax on physical properties and WVP of active EFs
Thickness Opacity Solubility % WVP [(g mm)/
Formulation μm SD AU nm SD SM SD (m2 h kPa)] SD
ACLS-0.2 % BW 52.67 0.50d 23.48 0.90c 97.38 0.15a 0.77 0.01a
ACLS-0.6 % BW 55.20 0.72bc 29.98 0.84c 96.65 0.07bc 0.69 0.06bc
ACLS-1 % BW 56.87 0.70b 36.75 2.27ab 96.03 0.12d 0.57 0.04d
OS-0.2 % BW 54.13 0.31cd 26.56 1.29c 97.01 0.13b 0.74 0.02b
OS-0.6 % BW 57.33 1.21b 35.00 3.02b 96.82 0.07bc 0.63 0.04bc
OS-1 % BW 64.47 0.95a 41.15 2.35a 96.51 0.20c 0.56 0.05c
ACLS acetylated cross-linked starch; OS oxidized starch; BW beeswax. Values are the mean of
three replicates SD. Means in columns not sharing the same letter are significantly different
( p < 0.05)
Active Edible Films Based on Modified Corn Starch for Food Applications 457
Fig. 1 In vitro minimum lethal concentration (MLC) of natural antimicrobials in RBC agar. (a)
LAE; (b) natamycin. Results are the average of three replicates
solubility and type of modified starch was observed at 0.6 % BW, where lower
solubility was found at the highest BW concentration for both starches
96.03 0.12 % SM and 96.51 0.20 % SM.
Complete inhibition of fungus was found when LAE was used at a concentration of
250 ppm, for initial fungal population of 5.5 log cfu/mL (Fig. 1a). A similar
behavior for natamycin antimicrobial effect was found, where complete fungal
inhibition was achieved at 20 ppm (Fig. 1b). The MLC is defined as the lowest
concentration of antimicrobial that produces a 99.9 % reduction of tested microor-
ganism (L opez-Malo et al. 2005). When the antimicrobials were incorporated into
the active EFs, inhibition effects were observed at 428 μg cm2 for LAE and
48 μg cm2 for natamycin (Fig. 2).
Strawberries were coated with the active EFs which gave a brilliant appearance.
After 9 days of storage, the control treatment showed fungal growth and decay
(Fig. 3). Weight loss of strawberries was reduced when using the active EF
incorporated with 428 μg cm2 of LAE and 1 % BW (Fig. 3). Weight loss is a
458 T. Arredondo-Ochoa et al.
a b
3 3
weight loss (%)
2 2
1 1
0 0
0 1 2 3 4 5 9 0 1 2 3 4 5 9
Days Days
Experiment 1 Experiment 2 Experiment 1 Experiment 2
c
3
weight loss (%)
0
0 1 2 3 4 5 9
Days
Experiment 1 Experiment 2
Fig. 3 Percentage of strawberry weight loss. (a) ACLS and 1 % BW; (b) OS and 1 % BW; (c)
Control
Active Edible Films Based on Modified Corn Starch for Food Applications 459
major cause of spoilage during storage of fresh fruits. The active EF provided a
good moisture barrier and surface antimicrobial properties, which extended the
shelf life of strawberries; therefore, the usefulness of this technology as applied to
preservation of fresh produce is promising.
4 Conclusions
Modified corn starches ACLS and OS as major components of the structural matrix
of FS are able to form EFs with good visual and mechanical properties, representing
an excellent alternative for developing tailor-made films or coatings for specific
food products. The use of BW nano-emulsion as a hydrophobic agent at 1 % (w/v)
concentration showed lower WVP and provided good solubility properties to EFs,
which is important to increase the shelf life of minimally processed fruits. Use of
the natural antimicrobials LAE and natamycin incorporated into active EFs
achieved complete inhibition of B. cinerea, the most common deteriorative fungus
of strawberries. Strawberries coated with active EFs showed brighter appearance
and lower weight loss than the control treatment.
References
ASTM (2000) Standard test method for water vapor transmission of materials. Standard Desig-
nation E96-00
Brody AL, Strupinsky ER, Kline LR (2001) Active packaging for food applications. CRC Press,
Boca Raton, pp 131–194
De Azeredo HMC (2009) Nanocomposites for food packaging applications. Food Res Int
42:1240–1253
Garcı́a MA, Martino MN, Zaritzky NE (1998) Plasticized starch-based coatings to improve
strawberry (Fragaria anannasa) quality and stability. J Agric Food Chem 46:3758–3764
Garcı́a MA, Pinotti A, Martino MN, Zaritzky NE (2009) Characterization of starch and composite
edible films and coating. In: Embuscado ME, Huber KC (eds) Edible films and coatings for
food applications. Springer, New York, pp 169–210
Hagenmaier RD, Baker RA (1994) Wax microemulsions and emulsions as citrus coatings. J Agric
Food Chem 42:899–902
Hu G, Chen J, Gao J (2009) Preparation and characteristics of oxidized potato starch films.
Carbohydr Polym 76:291–298
Janjarasskul T, Krochta MJ (2010) Edible packaging materials. Annu Rev Food Sci Technol
1:415–448
Kim KW, Ko CJ, Park HJ (2002) Mechanical properties, water vapor permeabilities and solubil-
ities of highly carboxymethylated starch-based edible films. J Food Sci 67:218–222
Lin D, Zhao Y (2007) Innovations in the development and application of edible coatings for fresh
and minimally processed fruits and vegetables. Compr Rev Food Sci F 6:60–75
Liu Z, Han JH (2005) Film-forming characteristics of starches. J Food Sci 70:31–34
Lopez OV, Garcı́a MA, Zaritzky NE (2008) Film forming capacity of chemically modified corn
starches. Carbohydr Polym 73:573–581
460 T. Arredondo-Ochoa et al.
L
opez-Malo A, Palou E, Parish ME, Davidson M (2005) Methods for activity assay and evaluation
of results. In: Davidson MP, Sofos JN, Branen AL (eds) Antimicrobials in food, 3rd edn. CRC
Taylor & Francis, Boca Raton, pp 659–680
McHugh TH, Krochta JM (1994) Edible coatings and films to improve food quality. CRC Press,
Florida, pp 139–141
Olivas GI, Barbosa-Cánovas G (2009) Edible films and coatings for fruit and vegetables. In:
Embuscado ME, Huber KC (eds) Edible films and coatings for food applications. Springer,
New York, pp 211–244
Ozdemir M, Floros J (2003) Film composition effects on diffusion of potassium sorbate through
whey protein films. J Food Sci 68:511–516
Pérez-Gallardo A, Bello-Pérez LA, Garcı́a-Almendárez B, Montejano-Gaitán G, Barbosa-
Cánovas G, Regalado-González C (2012) Effect of structural characteristics of modified
waxy corn starches on rheological properties, film-forming solutions, and on water vapor
permeability, solubility, and opacity of films. Starch-Starke 64:27–36
Quezada-Gallo JA (2009) Delivery of food additives and antimicrobials using edible films and
coatings. In: Embuscado ME, Huber KC (eds) Edible films and coatings for food applications.
Springer, New York, pp 315–334
Rojas-Graü M, Raybaudi-Massilia RM, Soliva-Fortuny RC, Avena-Bustillos RJ, McHugh TH,
Martı́n-Belloso O (2007) Apple puree-alginate edible coating as carrier of antimicrobial agents
to prolong shelf-life of fresh-cut apples. Postharvest Biol Technol 45:254–264
Romero-Bastida CA, Bello-Pérez LA, Garcı́a MA, Martino MN, Solorza-Feria J, Zaritzky NE
(2005) Physicochemical and microstructural characterization of films prepared by thermal and
cold gelatinization from non-conventional sources of starches. Carbohydr Polym 60:235–244
Singh J, Kaur L, McCarthy OJ (2007) Factors influencing the physicochemical, morphological,
thermal and rheological properties of some chemically modified starches for food applications:
a review. Food Hydrocoll 21:1–22
Suppakul P, Miltz J, Sonneveld K, Bigger SW (2003) Active packaging technologies with an
emphasis on antimicrobial packaging and its applications. J Food Sci 68:408–420
Tharanathan RN (2003) Biodegradable films and composite coatings: past, present and future: a
review. Trends Food Sci Technol 14:71–78
Zhou GH, Xu XL, Liu Y (2010) Preservation technologies for fresh meat: a review. Meat Sci
86:119–128
Influence of Moisture Content
and Temperature on the Stability
of a Dehydrated Probiotic Dairy Product
Containing Bifidobacterium infantis
or Lactobacillus acidophilus
Abbreviations
aw Water activity
awMS Water activity of maximum stability
BET Brunauer-Emmett-Teller monolayer
C Constant
DLPB Bifidobacterium infantis
DLPL Lactobacillus acidophilus
GAB Guggenheim-Anderson-de Boer monolayer
GSW Goat’s sweet whey
IDF International Dairy Federation
K Constant
mo Monolayer moisture
Mw Equilibrium moisture content
RH Relative humidity
1 Introduction
Since the beginning of the twentieth century, there has been growing interest in the
health benefits provided by probiotic bacteria (Trebichavský and Šplı́chal 2006).
A variety of subjects of interest revolve around this theme, such as the mechanisms
A process to obtain DLPB and DLPL was previously developed by our research
group. The process consisted of five stages: (1) fermenting goat’s sweet whey
(GSW) with B. infantis or L. acidophilus until mid-log phase using 10 % inoculum
concentrations and incubation conditions of 38 2 C and vacuum equivalent of
22 mmHg; (2) formulating 58.33 % fermented GSW with 8.33 % of a 6 % p/v
suspension of resistant starch (RS), 16.66 % of 15 % p/v solution of inulin, and
16.66 % of 10 % p/v solution of gelatin; (3) pouring the formulation onto a
non-stick pan and letting it stand at 4 C for 1 h; (4) drying in a convection oven
at 55 5 C for 2.66 0.22 h with an airflow of 0.212 0.03 m/min; and
(5) dehydrating in a hermetic chamber at 22 C and ~0 % relative humidity (RH),
provided by CaCl2, for 22 h. At the final point of the process, the probiotic products
of both strains were obtained with concentrations over 109 cfu/g of viable bacteria.
m o C aW
MW ¼ ð1Þ
ð1 aW Þ½1 þ ðC 1ÞaW
m o C K aW
MW ¼ ð2Þ
ð1 K aW Þ½1 þ ðC 1ÞaW
Two conditions of RH and temperature were chosen to study the stability of the
DLPB and DLPL. A condition, designated “adequate,” consisted of the same
temperature as the sorption isotherm (22 1 C) and an RH close to its analogue
awMS. A second condition, designated “inadequate,” consisted of an RH that
provided an aw higher than awMS and a temperature above that of the sorption
isotherm (37 2 C). This inadequate condition accelerated deteriorative reactions
according to the Arrhenius Law (Taub and Singh 1998). DLPB and DLPL samples,
without packaging, were exposed to each condition and triplicate specimens were
analyzed weekly. Two parameters were evaluated: the product moisture content, by
the gravimetric method, and the viable probiotic bacteria concentration, by the plate
counting method. Plate counting was conducted in accordance with the Official
Mexican Norm NOM-092-SSA1-1994, using conventional synthetic MRS agar
(DIFCO, Laboratories). Agar plates were incubated in a vacuum incubator (Shel-
don Manufacturing Inc., Cornelius, OR) at 38 2 C and a vacuum of 20 mmHg.
Figure 1 shows the experimental data fit to the BET and GAB models. A sigmoid
type III isotherm was generated (Labuza and Bell 2000). An abrupt increase
occurred after aw ¼ 0.5, which could reflect the dissolution of sugars (lactose)
once equilibrium was achieved. The BET equation was only applied to aw values
below 0.5, whereas the whole aw range was used for the GAB equation (Rangel-
Marron et al. 2010). Parameters are shown in Table 1. Considering the R2, the
experimental data fit better to GAB than to the BET model. The awMS was
determined from the GAB monolayer; the interception used to obtain it is shown
in Fig. 1. Since the awMS was very close to 0.33, this condition was employed as the
“adequate” condition for the subsequent DLPs stability study.
Influence of Moisture Content and Temperature on the Stability. . . 465
Fig. 1 Sorption isotherm at 22 1 C of a dehydrated and laminated dairy product using BET and
GAB models. Determination of maximum stability aw by interception of the y- and x-axis (the
GAB monolayer in the y-axis is used as the reference data)
Based on the information derived from the sorption isotherm, the “adequate” and
“inadequate” storage conditions were 33.9 % RH at 22 1 C and 53 % RH at
37 2 C, respectively. Figure 2 shows the evolution of moisture content during a
month of storage in both storage conditions; moisture content did not exceed
monolayer moisture in any case. DLPB and DLPL samples remained dry and crispy
throughout the entire study. Figure 3 shows the probiotic bacteria concentrations for
Fig. 2 Kinetics of moisture content of dehydrated and laminated probiotic dairy product
containing Bifidobacterium infantis (DLPB) and Lactobacillus acidophilus (DLPL), stored
under two conditions for 4 weeks: “adequate” (33.9 % RH and 22 1 C) and “inadequate”
(53 % RH at 37 2 C)
Fig. 3 Kinetics of concentration of viable probiotic bacteria (log10 cfu/g) of dehydrated and
laminated probiotic dairy product with Bifidobacterium infantis (DLPB) and Lactobacillus aci-
dophilus (DLPL), stored under two conditions for 4 weeks: “adequate” (33.9 % RH and 22 1 C)
and “inadequate” (53 % RH at 37 2 C)
Influence of Moisture Content and Temperature on the Stability. . . 467
DLPB and DLPL at both conditions over the 4-week monitoring period. Both
strains were stable under the adequate condition, maintaining concentrations
above 9 log10 cfu/g over the 4 weeks. In contrast, bacterial viability decayed
exponentially at the inadequate condition. DLPB and DLPL lost the minimal
effective concentration at the fourth week and the third week, respectively.
Consistent with these results, the storage conditions studied did not significantly
impact the DLPB and DLPL moisture content; in terms of moisture, both conditions
were satisfactory. Nevertheless, the rate of cellular death was significantly accel-
erated at storage conditions of 53 % RH and 37 2 C. Cellular death rate could be
considered as a deteriorative reaction that is potentiated at higher RH and temper-
ature. Viability of B. infantis and L. acidophilus was a critical factor for DLPB and
DLPL stabilities. When DLPB and DLPL were stored under proper conditions, they
were as stable as the refrigerated set yogurt with microencapsulated bifidobacteria
studied by Adhikari et al. (2000) and were more stable than yogurts with
nonencapsulated cells, which usually do not maintain acceptable viable probiotic
concentrations beyond 1 month of refrigerated storage.
4 Conclusions
DLPB and DLPL lasted more than a month when they were stored at 33.9 % RH
and 22 1 C, and they retained their viable probiotic bacteria counts at over
9 log10 cfu/g. In contrast, cellular death rate was exponentially accelerated under
conditions of 53 % RH and 37 2 C, and the probiotic concentration decreased to
below the minimum required for a probiotic product (7 log10 cfu/g). As expected,
increases in storage RH and temperature diminished the stability of the probiotic
products studied, compromising their shelf life and their concentrations of viable
bacteria that are highly associated with therapeutic effectiveness.
References
Greenspan L (1977) Humidity fixed points of binary saturated aqueous solutions. J Res Nat Bur St
Phys Chem 81a(1):89–96
Kailasapathy K (2002) Microencapsulation of probiotic bacteria: technology and potential appli-
cations. Curr Issues Intest Microbiol 3(2):39–48
Kailasapathy K, Rybka S (1997) L. acidophilus and Bifidobacterium spp.: their therapeutic
potential and survival in yogurt. Aust J Dairy Technol 52(1):28–35
Labuza TP, Bell LN (2000) Moisture sorption: practical aspects of isotherm measurement and use,
2nd edn. American Association of Cereal Chemists, St. Paul, MN, USA, p 122
Norma Oficial Mexicana NOM-092-SSA1-1994 (1994) Bienes y servicios. Método para la cuenta
de bacterias aerobias en placa. http://www.salud.gob.mx/unidades/cdi/nom/092ssa14.html
Rangel-Marr on M, Welti-Chanes J, Cordova-Quiroz AV, Cer on-Breton JV, Anguebes-Franseschi-
F, Moreno-Martı́nez V (2010) Sorption isotherms of mango (Mangifera indica L.) pulp freeze-
dried. In: Mladenov V et al (eds) Advances in control, chemical engineering, civil engineering
and mechanical engineering. WSEAS Press, USA, pp 114–118
Taub IA, Singh RP (1998) Food storage stability. CRC Press, Boca Rat on, USA, p 539
Timmermann EO, Chirife J, Iglesias HA (2001) Water sorption isotherms of foods and foodstuffs:
BET or GAB parameters. J Food Eng 48(1):19–31
Trebichavský I, Šplı́chal I (2006) Probiotics manipulate host cytokine response and induce
antimicrobial peptides. Folia Microbiol 51(5):507–510
Effect of Relative Humidity on Shrinkage
and Color Changes in Dehydrated
Strawberry
Abbreviations
1 Introduction
Dehydrated fruits are prone to discoloration during storage. Many natural pigments
are unstable in dried fruits, and brown pigments can be formed. The rate of physical
and chemical changes in dried vegetables and food models is slow in the glassy
state. At temperatures above the glass transition, in addition to decreasing viscosity
and increasing rate, other changes such as crystallization and structural collapse
affect the rate of discoloration (Karmas et al. 1992). Diffusion-controlled chemical
reactions are particularly dependent on translational diffusivity of the reactants
(or on the viscosity of the matrix material) and are thus susceptible to the physical
state of the system (Slade et al. 1995). It has been shown that matrix collapse caused
by storage above the glass transition temperature (Tg) or by mechanical compres-
sion and porosity (Burin et al. 2004; White and Bell 1999) affected browning rates,
indicating that besides water content, system structure plays a relevant role.
Color and size changes affect fruit appearance and organoleptic quality and may
be an indication of decreased nutritional and functional properties of foods. Food
discoloration can occur homogeneously, but most often, heterogeneous distribution
of color is observed. Color meters typically claim to measure the average color of
the food material’s surface exposed to the measuring window of the instrument
(Balaban 2008). However, many food products, such as strawberry slices, show
heterogeneous color distribution. While averaging is acceptable for uniformly
colored surfaces (Briones and Aguilera 2005; Mendoza et al. 2006; Venir
et al. 2007; Acevedo et al. 2008a, b), when the colors change with location or are
non-homogeneous, averaging may result in inaccurate color determination
(Balaban 2008). The objective of this work was to assess the effect of relative
humidity (RH) on color alterations and shrinkage in freeze-dried strawberry slices.
2.1 Fruits
Fully ripe fresh strawberries were obtained from a local market and stored at 4 C
until the initiation of the experiment. The fruits were washed and then cut
transversally into slices (2.0 cm diameter and 0.5 cm thickness). The cut material
was immediately frozen with liquid nitrogen and stored at 20 C.
Fruit slices were covered with liquid nitrogen before freeze-drying. A freeze-drier
(ALPHA 1-4 LD2 Martin Christ Gefriertrocknungsanlagen GmbH, Germany) was
used. The freeze-drier was operated at 55 C, at a chamber pressure of 4 Pa, and
Effect of Relative Humidity on Shrinkage and Color Changes in Dehydrated. . . 471
the process lasts 48 h. After freeze-drying, several pieces were distributed into vials
for the determination of anthocyanin content. Samples used for all determinations
were humidified in a range of 11–84 % RH for 14 days at 20 C (Greenspan 1977).
After humidification, the strawberry slices used to determine color changes were
placed inside rubber O-rings and sandwiched between two glass plates and then
hermetically sealed to avoid water loss, according to the method reported by
Acevedo et al. (2008a, b). They were then placed in a forced air oven operated at
45 1 C during 120 h. Two glass plates containing six strawberry slices for each
relative humidity were removed from the oven after adequate time to acquire
images and measured in their glass-covered containers.
Strawberry color changes were determined by image analysis. The CVS consisted
of three elements: a lighting system, a digital camera, and a personal computer. The
lighting system included a D65 lamp (corresponding to solar irradiation with a color
temperature of 6,500 K, Lozano 1978) inside a gray chamber (N7 in the Munsell
color space). Warm up of the lamps was 15 min before taking pictures. The angle
between the camera axis and the sample plane was 90 , and the angle between the
light source and the sample plane was 45 , in order to capture the diffuse reflection
responsible for color produced at this angle (Yam and Papadakis 2004). A high-
resolution (10.1 megapixel) digital camera EOS 40D (Canon Inc., Japan) was used,
with an EF-S 60 mm f2.8 macro lens (Canon Inc., Japan). Samples were transferred
from the forced air oven to the gray box, and images were acquired at different
times during the entire storage period. The digital camera was operated in manual
mode, with the lens aperture at f ¼ 6.3 and speed 1/8 s (no zoom, no flash) to
achieve high uniformity and repeatability. The calibration of the camera and the
parameters used for image capture are described in Briones and Aguilera (2005).
Images have a resolution of 3,888 2,592 pixels and were stored in JPEG format
using Canon’s Remote Capture program (EOS Utility, Canon Inc., USA). The
images were taken using a white background.
CVS permitted acquisition of information for the whole pieces directly inside the
glass plates. Color images were obtained in lab values using Adobe Photoshop CS4
software (Adobe Systems Inc., San Jose, CA) and were then converted to standard
CIELAB space using mathematical formulas described by Papadakis et al. (2000)
from CIELAB coordinates (L*, a*, and b*).
It is possible for the software to identify every pixel in an image that has color
attributes (L, a, b) lower than or higher than a given threshold or attributes between
472 L.M. Agudelo-Laverde et al.
2 threshold values. Once these pixels are identified and counted, then their
percentage based on the total view area of an object can be calculated. In this
way, images of the strawberry slices were segmented in two different zones,
according to their a* values. The selected intervals of a* values were from 0.5 to
17 (representing the lighter sections) and from 17 to 28 (representing the reddish
sections). The number of pixels of each zone of the strawberry slices was quantified,
and their corresponding percentage was calculated. The color functions selected to
follow browning changes in fruits were the L*, a*, and b*.
Collapse of the systems was determined by measuring the volume of the fruit discs
from the pictures taken with the CVS (as previously described). Modeling program
SolidWorks 2011 3D mechanical was used for analysis of collapse. Volume
reduction was expressed as the diameter percentage with respect to a control disc
without shrinkage. Six replicates of each sample were measured and the average
and standard deviation determined.
Figure 1 shows the global color changes observed for freeze-dried strawberries
humidified at several RHs when stored at 45 C as a function of time. At 11 % RH,
the samples presented a lighter color, manifested by the higher L* and lower b*
values, while at 43 % RH L* and b* values were relatively constant during storage
time. Slight changes were observed in the L* and b* values at 52 % RH. In the rest
of the samples, due to browning reactions, L* and b* values decreased, causing a
change from the original light color to different brown degrees. However, the a*
component decreased drastically at all RH values, and the rate of these changes was
more important with the increase in RH.
The rate of redness loss was approximated by the slope of the initial linear part of
the curves shown in Fig. 1 (as pseudo-zero-order reaction rate), and shrinkage was
determined as the % of volume reduction relative to the control sample. Both
parameters were plotted in Fig. 2 as a function of (T-Tg), the difference between
storage temperature, and Tg, for the samples stored at 45 C for 22.5 h.
As observed in Fig. 2, the volume reduction and redness loss increased at
increasing (T-Tg) due to RH increase. The observed shrinkage was caused by
storage at conditions well above the Tg (T-Tg > 40 C). The rate of initial redness
loss showed a drastic increase at (T-Tg) values higher than 70 C, when the sample
shrinkage approximated 50 % (corresponding to 84 % RH). The 1H NMR signal
decay of the transversal relaxation times, T2, was obtained by the CPMG pulse
sequence in order to relate the observed behavior of the change of the chromatic
attributes with the molecular mobility in the samples. It is important to note that at
these conditions, populations of water molecules with high mobility are
present, as determined by pulsed 1H NMR, and frozen water is detected by DSC
(Agudelo-Laverde et al. 2011).
Because strawberry slices presented heterogeneous color distribution, in addi-
tion to analyzing the global color changes, the image analysis of the slices was done
by dividing the area into two different zones. The color changes were better
represented by the a* variable, showing a decrease along storage time because
the heat treatment caused anthocyanin pigment destruction. Before storage at 45 C,
the a* values could be distributed in different sections having different red degree,
as expected for strawberry slices (Fig. 3). The pixel proportion of red areas in the
pictures was greater at low RHs and short storage times. As RH and storage time
increased, the areas with high a* values decreased and the homogeneity of color
distribution also increased.
474 L.M. Agudelo-Laverde et al.
L*
52% RH
20 75% RH
84% RH
0
0 5 10 15 20 25
Time, hour
40
30
20
a*
10 11% RH
43% RH
52% RH
0 84% RH
75% RH
-10
0 5 10 15 20 25
Time, hour
50
40
11% RH
30 43% RH
b*
52% RH
20
75% RH
10 84% RH
0
0 5 10 15 20 25
Time, hour
Effect of Relative Humidity on Shrinkage and Color Changes in Dehydrated. . . 475
Volume reduction, %
changes (asterisk) after 0.30
22.5 h of storage at 45 C, 40
as a function of (T-Tg)
K, a*/h.
30 0.20
20
0.10
10
0 0.00
40 50 60 70 80
T-Tg, °C
t = 0h t = 0h t = 0h
Fig. 3 Proportion (%) of pixels corresponding to different sections of a* values for 0 and 22.5 h of
storage at 45 C at 11, 43, and 84 % RH
4 Conclusions
The final color of freeze-dried strawberries during storage was the result of loss of
redness and browning development. Redness loss and browning could be related to
the decrease of CIELAB coordinates a* and L*, respectively. During storage at
45 C for 22.5 h, a* constantly decreased with increasing RH, but L* decrease was
not significant below 43 % RH.
476 L.M. Agudelo-Laverde et al.
Sample shrinkage and redness loss were significantly affected by RH. Both
shrinkage and color changes seem to be affected by RH and (T-Tg) values at
which water mobility is characterized by T2 relaxation times in the order of ms,
and water is able to crystallize.
The segmented image analysis revealed that humidification level and storage
time, besides influencing the kinetics of changes, were determinant factors in
defining the degree of color heterogeneity in dehydrated strawberries.
References
Acevedo N, Briones V, Buera M, Aguilera J (2008a) Microstructure affects the rates of chemical,
physical and color changes during storage of dried apple discs. J Food Eng 85(2):222–231
Acevedo NC, Schebor C, Buera P (2008b) Non-enzymatic browning kinetics analyzed through
water-solid interactions and water mobility in dehydrated potato. Food Chem 108(3):900–906
Agudelo-Laverde LM, Acevedo N, Schebor C, Buera MP (2011) Integrated approach for
interpreting browning rate dependence with relative humidity in dehydrated fruits.
LWT-Food Sci Technol 44:963–968
Balaban MO (2008) Quantifying nonhomogeneous colors in agricultural materials. Part I: method
development. J Food Sci 73(9):431–437
Briones V, Aguilera JM (2005) Image analysis of changes in surface color of chocolate. Food Res
Int 38(1):87–94
Burin L, Jouppila K, Roos Y, Kansikas J, Buera MP (2004) Retention of β-galactosidase activity as
related to Maillard reaction, lactose crystallization, collapse and glass transition in low
moisture whey systems. Int Dairy J 14(6):517–525
Greenspan L (1977) Humidity fixed points of binary saturated aqueous solutions. J Res Natl Bur
Stand 8:89–96
Karmas R, Buera M, Karel M (1992) Effect of glass transition on rates of non-enzymatic browning
in food systems. J Agric Food Chem 40(4):873–879
Lozano R (1978) El color y su medicion. Editorial América Lee, Buenos Aires
Mendoza F, Dejmek P, Aguilera JM (2006) Calibrated color measurement of agricultural food
using image analysis. Postharvest Biol Technol 41(3):285–295
Papadakis S, Abdul-Malek S, Kandem R, Yam K (2000) A versatile and inexpensive technique for
measuring colour of foods. Food Technol 54(12):48–51
Slade L, Levine H, John EK, Taylor SL (1995) Glass transitions and water food structure
interactions. Adv Food Nutr Res 38:103–269
Venir E, Munari M, Tonizzo A, Maltini E (2007) Structure related changes during moistening of
freeze dried apple tissue. J Food Eng 81(1):27–32
White K, Bell L (1999) Glucose loss and Maillard reaction in solids as affected by porosity and
collapse. J Food Sci 64:1010–1014
Yam KL, Papadakis SE (2004) A simple digital imaging method for measuring and analysing
color of food surfaces. J Food Eng 61(1):137
Effect of Blanching and/or Osmotic
Dehydration on Texture and Rheological
Properties of Apple Tissue
Abbreviations
A1 Area 1
Adh Adhesiveness
ANOVA Analysis of variance
B Blanched apples
BOD Blanched and osmotically dehydrated apples
C Raw apples
Chew Chewiness
Coh Cohesiveness
Ed Modulus of deformability
F Fracturability
G00 Loss modulus
Gum Gumminess
G0 Storage modulus
H Hardness
H2 Hardness 2
J (t, τ) Creep compliance
MANOVA Multivariate analysis of variance
OD Osmotically dehydrated apples
PCA Principal component analysis
A.B. Garcı́a Loredo, S. Guerrero and S.M. Alzamora are members of Consejo Nacional de
Investigaciones Cientı́ficas y Técnicas de la República Argentina.
A.B. Garcı́a Loredo (*) • S. Guerrero • S.M. Alzamora
Departamento de Industrias, Facultad de Ciencias Exactas y Naturales,
Universidad de Buenos Aires, Ciudad Universitaria, Buenos Aires, Argentina
e-mail: [email protected]
Sp Springiness
T Time
TPA Texture profile analyses
γ Strain
ηi Coefficients of viscosity associated with the Kelvin-Voigt elements
ηN Coefficient of viscosity associated with Newtonian flow
λ Retardation times
τ Constant stress applied
1 Introduction
Apple slices (Malus pumila, Granny Smith cv) were cut parallel to the main axis
(3 cm in diameter, 1 cm in thickness) and subjected to (1) osmotic dehydration at
atmospheric pressure by immersion into a 22 % w/w aqueous glucose (®Cerelose,
Refinerı́as de Maı́z, Argentina) solution at 25 C to reach aw 0.97 (ffi6½ h);
(2) steam blanching by immersion in saturated steam for 90 s and cooling in
water for 20 s at 5 C; and (3) steam blanching followed by osmotic dehydration,
as previously described. Rheological and texture characteristics and microstructure
were analyzed in raw (C), blanched (B), osmotically dehydrated (OD), and
blanched and osmotically dehydrated (BOD) apples.
Linear viscoelastic properties were characterized at 25 C in a Paar Physica
MCR 300 rheometer. Dynamic oscillatory tests were performed in a controlled
strain mode (frequency range 0.1–100/s; strain amplitude 0.005 %; 10 replicates).
Creep/recovery tests were conducted by applying a constant shear stress of 35 Pa
until 100 s (recovery phase: 200 s; 15 replicates). Compliance data from creep
Effect of Blanching and/or Osmotic Dehydration on Texture and Rheological. . . 479
2
X
J ðt; τÞ ¼ ðJ 0 Þ þ J i 1 et=λi þ t=ηN ð1Þ
i¼1
where J(t, τ) is the creep compliance (¼ γ(t)/τ), with γ(t) being the strain at time
t and τ the constant stress applied. J0 is the instantaneous compliance at t ¼ 0; Ji are
the retarded compliances; λi (¼ ηi Ji) are the retardation times and ηi are the
coefficients of viscosity associated with the Kelvin-Voigt elements; and ηN is the
coefficient of viscosity associated with Newtonian flow and its inverse, the steady-
state fluidity of the material.
Instrumental texture profile analyses (TPA) were characterized at 25 C (20 rep-
licates, cross-head speed ¼ 60 mm/min, two cycle compression, 70 % deformation,
load range ¼ 5,000 kgf) using an Instron Testing Machine model 3345. The test was
set for calculating fracturability, hardness, hardness2, area1, adhesiveness to palate,
cohesiveness, springiness, chewiness, gumminess, and modulus of deformability
(Bourne 1978).
Sensory analysis was performed by nine assessors trained in the texture profile
method following the procedure described by Civille and Szczesniak (1973) to
recognize texture attributes of hardness, fracturability, juiciness, crispness, cohe-
siveness, and adhesiveness to palate.
Light microscopy observations of raw and treated fruit tissues were performed in
a Carl Zeiss Axioskop 2 microscope according to conventional techniques
(D’Ambrogio 1986).
Multivariate analysis of variance (MANOVA) and principal component analysis
(PCA) were used to analyze rheological data. Analysis of variance (ANOVA) was
used to analyze sensory data. Statistical analysis was conducted using Infostat
v. 2009 software (Cordoba National University, Argentina).
All apple samples showed solid behavior with the storage modulus (G0 ) exceeding
the loss modulus (G00 ) over the entire frequency range, but both dynamic moduli
were reduced due to processing. G0 and G00 values for fresh and processed tissues
showed a weak dependence on frequency (Fig. 1a, b, respectively). The storage
modulus G0 slightly increased with increasing angular frequency. The frequency
dependence of G00 was more complex. In general, tan δ was not sensitive for
distinguishing physical differences between raw and treated apple tissues or
between treatments. The pattern found for dynamic spectra was in agreement
480 A.B. Garcı́a Loredo et al.
a
1.0E+06
G' (Pa)
1.0E+05
1.0E+04
0.1 1 10 w (1/s) 100
b
1.0E+05
G'' (Pa)
1.0E+04
1.0E+03
w (1/s)
0.1 1 10 100
Fig. 1 Variation of (a) storage modulus (G0 ) and (b) loss modulus (G00 ) with frequency for raw
and treated apple tissue at a strain of 0.005 %. (Filled diamond) fresh, (open square) blanched,
( filled triangle) osmotically dehydrated, ( filled circle) blanched and osmotically dehydrated
with those previously reported for apple (Martı́nez et al. 2007), melon (Martı́nez
et al. 2005), and Korla pear (Wu and Guo 2010).
Creep/recovery curves for fresh and treated samples are presented in Fig. 2.
Creep curve parameters showed significant differences ( p < 0.0001) between fresh
and treated samples and also between apples subjected to OD and heated samples
(with or without OD) (Table 1). Instantaneous compliance (J0), retarded compli-
ances (J1, J2), and steady-state viscous compliance (1/ηN) increased in different
degrees for treated samples. Principal component analysis (PCA) showed the
spatial relationships of creep parameters for each sample. The first two principal
components described 86.8 % and 10.7 % of the variance, respectively. The first
Effect of Blanching and/or Osmotic Dehydration on Texture and Rheological. . . 481
0.0014
γ (%)
0.0012
0.001
0.0008
0.0006
0.0004
0.0002
0
0 50 100 150 200 250 t (s) 300
Table 1 Viscoelastic parameters derived by fitting Eq. (1) to average compliance curves from the
creep phase for fresh and treated apple tissue
J0 (1/Pa) J1(1/Pa) J2(1/Pa) ηN (Pa.s)
Treatment (105) (105) (105) λ1(s) λ2 (s) (107)
C 0.34 0.11 0.083 25.45 2.16 11.2a
B 2.1 0.73 0.42 23.81 2.7 1.38b
OD 1.5 0.3 0.26 23.09 2.46 2.5c
ODB 2.1 0.66 0.39 21.96 2.38 1.8b
Treatments with same letter are not significantly different ( p < 0.05)
axis was positively defined by J0, J1 and J2 and negatively by ηN, and the second
axis was positively defined by retardation times (λ1 and λ2).
Figure 3 shows typical compression-time curves for fresh and treated apples.
Significant differences ( p < 0.0001) were observed among all samples. The param-
eters for hardness (H), hardness 2 (H2), fracturability (F), area 1 (A1), cohesiveness
(Coh), springiness (Sp), adhesiveness (Adh), gumminess (gum), chewiness (chew),
and modulus of deformability (Ed) decreased for heated samples (with or without
OD) with respect to control. Hardness and fracturability slightly increased for
osmotic dehydrated samples (Table 2). PCA indicated that the first two principal
components explained 86.6 % and 9.8 % of the variance, respectively. The first axis
was positively defined by hardness 2, area 1, springiness, cohesiveness, chewiness,
and gumminess, and the second axis was positively presented by hardness and
fracturability.
Sensory texture analysis (data not shown) showed significant differences in
hardness (F3,12 ¼ 5.52, p ¼ 0.0063) between fresh and B tissues and between
482 A.B. Garcı́a Loredo et al.
400
Force (N)
300
200
100
0
0 5 10 15 20 t (s) 25
Fig. 3 Force/time curves of raw and treated apple tissues. (Filled diamond) fresh, (open square)
blanched, ( filled triangle) osmotically dehydrated, ( filled circle) blanched and osmotically
dehydrated
Table 2 TPA curve parameters for fresh and treated apple tissue
H1 A1 Sp Coh Gum Chew Ed (N/m2)
Treatment F (N ) H (N ) (N ) (J ) () () (N ) (N ) (106)
C 287.9 293.2 225 1.39 0.55 0.081 23.9 13.2 1.92a
B 196.2 196.2 80.4 0.7 0.33 0.04 8 2.6 0.11b
OD 337 337 187.4 1.31 0.53 0.064 21.3 11.4 0.17c
ODB 0 288 92.6 0.93 0.41 0.045 12.8 5.3 0.09d
Treatments with same letter are not significantly different ( p < 0.05)
Fig. 4 Light microscopy images from fresh and treated apple tissue. C: raw (control);
B: blanched; OD: osmotically dehydrated; BOD: blanched and osmotically dehydrated
OD cell walls exhibited a slightly lower density of staining than did the control.
BOD tissues showed disruption of membranes and severe folding of cell walls,
probably due to the important loss of turgor. But walls appeared more stained than
those of the only heated sample. Shrinkage of BOD tissue and reduction of
intercellular spaces were notable (Fig. 4).
Considering structural changes, rheological properties behavior of treated tis-
sues could be attributed to turgor loss (membrane breakage and/or plasmolysis),
shrinkage of tissue or folding, and degradation of cell walls. The decrease in G0
values, crispness, and instrumental and sensory fracturability of heated and/or
osmosed apples could be attributed to the decrease in turgor pressure and the
reduction of intercellular air spaces. In contrast, instantaneous compliance (J0)
and retarded compliances (J1 and J2), instrumental and sensory hardness, G0 and
juiciness were most sensitive parameters to changes in cell wall structure. On the
other hand, increase of G00 and 1/ηN and decrease of juiciness could be attributed in
part to an increase of cell wall fluidity arising from solubilization and depolymer-
ization of pectin and hemicelluloses as well as disruption of membranes and
exosmosis.
484 A.B. Garcı́a Loredo et al.
4 Conclusions
Apple tissues (raw and treated) behaved predominantly as solids, but G0 and G00 moduli
decreased because of processing. Texture parameters (hardness, fracturability, hard-
ness 2, area 1, cohesiveness, springiness, gumminess, chewiness, and deformability
modulus) decreased for heated tissues (with and without osmosis) and hardness and
fracturability slightly increased in OD samples. Clear correlations could be suggested
between some structural elements and sensory and rheological parameters, for
instance: loss of turgor-crispness-fracturability (sensory and instrumental)-G0 ; cell
wall degradation-hardness (sensory and instrumental)-Ji; and disruption of mem-
branes-G00 -juiciness-ηN. Blanching provoked a greater modification of structure and
rheological properties than osmotic dehydration.
References
Abbreviations
aw Water activity
IS Isotonic solution
OS Sucrose solution concentration
pv Vacuum pulse
t Processing time
tv Vacuum pulse duration
Vo Initial volume of the sample
Wf Tejocote weight after impregnation
Wo Tejocote weight before impregnation
X Impregnated volume fraction
εr Real porosity
ρa Apparent density of tejocote
ρIS Density of the impregnation IS
1 Introduction
Tejocote (Crataegus pubescens) is a fruit that has been used in México since
pre-Hispanic times (Nú~nez-Colı́n et al. 2008; Franco-Mora et al. 2010). It is
commonly preserved by cooking in hypertonic sucrose solutions and then packing
in hot sterilized jars. Thermal treatment also involves osmotic dehydration and
solutes impregnation phenomena of the fruit. The overall effect of this mass transfer
leads to aw reduction of the product, which may contribute to its stability. In
addition to inactivation of enzymes and deteriorative microorganisms, thermal
treatment accelerates dehydration-impregnation of the product.
The mass transfer phenomena can also be accelerated by using a pressure
gradient at ambient temperature, in addition to the concentration gradient (Tortoe
2010). Under these conditions, a minimally processed product can be obtained,
while at the same time reducing undesirable quality changes of the product caused
by high temperatures. The feasibility of pressure gradient application depends
partially on the natural porous properties of the material and impregnation condi-
tions (Zhao and Xie 2004). Two important porous properties are volume fraction of
the sample occupied by the impregnating solution as a result of pressure changes (X,
cm3 isotonic solution/cm3 sample) and real porosity (εr), defined as the ratio of
volume of voids in the sample to bulk volume of the material (cm3 void/cm3
sample) (Gras et al. 2002; Zhao and Xie 2004).
Stabilization of tejocote by minimal processing has not been studied; therefore,
the objectives of this work were to determine the best impregnation conditions of
tejocote fruit using pressure gradients and to determine the effect of the osmotic
solution and processing time on water activity (aw) of tejocote at ambient temper-
ature, assisted by an initial pressure gradient.
2 Methodology
W f Wo
X¼ ð1Þ
ρIS V o
Water Activity Depression of Tejocote Fruit (Crataegus pubescens). . . 487
Table 2 Variables and levels of the experimental design for dehydration-impregnation of tejocote
Variables Levels
[Sucrose], (Brix) 30 40 50
tp (min) 0.5 2.75 5
ρr ρa
εr ¼ ð2Þ
ρa
where Wo and Wf are tejocote weight before and after impregnation, respectively,
Vo is the initial volume of the sample, ρIS is the density of the impregnation IS, εr is
the real density of the tejocote puree, and ρa is the apparent density of tejocote
(Mújica-Paz et al. 2003a,b; Zhao and Xie 2004).
Table 3 Characteristics
Moisture (%) 78.26 0.12
of fresh tejocote
Titratable acidity (%) 2.30 0.07
Soluble solids ( Brix) 9.20 0.01
pH 3.63 0.01
aw 0.982 0.002
60.0 60.0
0.450
45.7 45.7
0.400
0.250
tv (min)
tv (min)
3.0 3.0
15.0 23.7 32.5 41.2 50.0 15.0 23.7 32.5 41.2 50.0
pv (cmHg) pv (cmHg)
Fig. 1 Effect of vacuum pressure and application time on the impregnated volume fraction of
tejocote, at 3 and 120 min of relaxation
Real porosity of tejocote was 0.53; according to this fraction of total voids in tejocote,
it is more porous than other whole products such as prickly pear (0.013), quince
(0.142), red jalape~no pepper (0.120), and apple (0.273) (Valdez-Fragoso et al. 2007).
Significant polynomial ( p < 0.05) with R2 ¼ 0.805 and no significant lack of fit
was found to predict X in the experimental studied domain. The factors pv and the
interaction tv-tr showed significant effect on X levels. Nearly 45 % of the porous
structure of tejocote can be occupied by external sucrose IS (Fig. 1), and due to the
tv-tr interaction effect, this impregnation level can be attained at two combinations
of independent factors: pv ¼ 50 cmHg, tv ¼ 60 min y tr ¼3 min or pv ¼ 50 cmHg,
tv ¼ 3 min y tr ¼ 120 min.
Operating conditions pv ¼ 50 cmHg and tv ¼ 3 min were chosen to create a pressure
gradient at the beginning of the impregnation-dehydration treatment of tejocote.
4 Conclusions
0.965
0.956
0.948
tejocote aw
0.939
0.930
30
5.0
35 3.9
40 2.8
[sucrose] (°Brix) 45 1.6 time (days)
50 0.5
Acknowledgments The authors gratefully acknowledge support from Tecnol ogico de Monterrey
(Research Chair Funds CAT-200) and tejocote producers of Sn Juan Tetla, Puebla, México.
References
AOAC (1990) Official methods of analysis of the Association of Official Analytical Chemists,
15th edn. AOAC International, Washington, DC
Franco-Mora O, Aguirre-Ortega S, Morales-Rosales EJ, González-Huerta A, Gutiérrez-Rodrı́guez
F (2010) Caracterizacion morfologica y bioquı́mica de frutos de tejocote (crataegus mexicana
DC), de Lerma y Ocoyoacac, Méx. Ciencia ergo Sum 17-1, 61-66. UAEM
Gras M, Vidal-Brotons D, Betoret N, Chiralt A, Fito P (2002) The response of some vegetables to
vacuum impregnation. Innov Food Sci Emerg Technol 3:263–269
Leistner L (1995) Use of hurdle technology in food processing: recent advances. In: Barbosa-
Cánovas GV, Welti-Chanes J (eds) Food preservation by moisture control fundamentals and
applications. Technomic Co., Lancaster, PA, USA, pp 377–392
Montgomery DC (2001) Design and analysis of experiments, 5th edn. Wiley, New York, pp
170–210
Mújica-Paz H, Valdez-Fragoso A, Lopez-Malo A, Palou E, Welti-Chanes J (2003a) Impregnation
and osmotic dehydration of some fruits: effect of the vacuum pressure and syrup concentration.
J Food Eng 57:305–314
Mújica-Paz H, Valdez-Fragoso A, Lopez-Malo A, Palou E, Welti-Chanes J (2003b) Impregnation
properties of some fruits at vacuum pressure. J Food Eng 56:307–314
Nú~
nez-Colı́n CA, Nieto-Ángel R, Barrientos-Priego AF, Sahagún-Castellanos J, Segura S,
González-Andrés F (2008) Variability of three regional sources of germplasm of Tejocote
(Crataegus spp.) from central and southern Mexico. Genet Resour Crop Evol 55:1159–1165
Tortoe C (2010) A review of osmodehydration for food industry. Afr J Food Sci 4(6):303–324
Water Activity Depression of Tejocote Fruit (Crataegus pubescens). . . 491
Abbreviations
A Constant (g/mol)
DP Degree of polymerization
DSC Differential scanning calorimetry
k Constant reflecting the sensitivity to the water plasticizing effect
MW Molecular weight
RH Relative humidity
Tg Glass transition temperature
Tg1 Glass transition of the anhydrous inulin
Tg2 Glass transition of water
W1 Mass fraction of inulin
W2 Mass fraction of water
K. Kawai (*)
Department of Biofunctional Science and Technology, Graduate School of Biosphere Science,
Hiroshima University, 1-4-4 Kagamiyama, Higashi-Hiroshima, Hiroshima 739-8528, Japan
e-mail: [email protected]
K. Fukami
San-ei Sucrochemical Co., Ltd., 24-5 Kitahama-cho, Chita, Aichi 478-8503, Japan
P. Thanatuksorn • C. Viriyarattanasak • K. Kajiwara
School of Bioscience and Biotechnology, Tokyo University of Technology, 1404-1 Katakura,
Hachioji, Tokyo 192-0982, Japan
1 Introduction
Three types of inulin reagent, low molecular weight (low-MW), native, and high
molecular weight (high-MW), were provided from San-ei Sucrochemical, Co., Ltd.,
Aichi, Japan. The average molecular weight of low-MW, native, and high-MW
inulin samples was 1,197, 2,175, and 4,395 g/mol (DP ¼ 7, 13, and 27), respec-
tively. Since the high-MW reagent was preliminarily confirmed to be semi-crystal
state by X-ray diffraction, amorphous reagent was prepared by thermal hydrolysis
treatment (Zimeri and Kokini 2002; Ronkart et al. 2009). The fully dried reagents
were held under various relative humidity (RH) conditions at 25 C, and samples
with varying of water content were prepared.
2.2 DSC
The Tg of the inulin samples was investigated using a DSC (DSC8230; Rigaku Co.,
Tokyo, Japan). The sample (2–10 mg) was placed in an aluminum pan and
hermetically sealed. In order to evaluate anhydrous inulin Tg, inulin that had been
vacuum-dried at 60 C for 12 h was put into the pan and held at 105 C for 2 h prior
to sealing. DSC measurement was performed at 5 C/min in the temperature range
Prediction Approach to the Glass Transition Temperature of Amorphous. . . 495
of 0 and 180 C. In order to cancel the effect of the thermal history on Tg, the
samples were re-scanned after cooling, and Tg was evaluated from the onset point of
an endothermic shift observed in the second scanning. All the tests were performed
in duplicate and the results were averaged.
Typical DSC thermograms for the native inulin samples are shown in Fig. 1. The
DSC thermograms showed an endothermic shift due to glass transition, and Tg
value was evaluated from the onset point of the shift. It was noted that the Tg
decreased with an increase in the RH of sample. Water content of the samples was
evaluated gravimetrically, and the Tg was plotted against water content in Fig. 2. It
was confirmed that the Tg decreased with an increase in water content because of
water plasticizing. In order to understand the plasticizing effect, the data was fitted
by the Gordon-Taylor equation (Eq. 1):
300
RH = 0%
250 RH = 7.0 %
DSC out put (W/sample-g)
RH = 11.0 %
200
RH = 22.5 %
150
RH = 32.8 %
100
RH = 43.2 %
50 RH = 52.9 %
Endo.
↓
RH = 57.6 %
Fig. 1 Typical DSC 0
thermograms for the native 270 290 310 330 350 370 390 410 430
inulin samples Temperature (°C)
496 K. Kawai et al.
k
the data. The inset figure
400 k = -0.0412Tg
shows a relationship 5.4
between k and Tg + 22.291
380 5.0
380 390 400 410 420
360 Tg (K)
Tg (K)
340
320
300 low-MW
Native
280
high-MW
260
0 2 4 6 8 10 12 14
Moisture content (%, w/w)
W 1 T g1 þ kW 2 T g2
Tg ¼ ð1Þ
W 1 þ kW 2
where Tg1 and Tg2 are Tg of the anhydrous inulin and water, W1 and W2 are the mass
fraction of inulin and water, and k is a constant reflecting the sensitivity to the water
plasticizing effect. The Tg2 was obtained as 136 K from a previous publication
(Johari et al. 1987), and k was obtained as a fitted parameter. A relationship between
the obtained k value and Tg of anhydrous inulin is shown in the inserted figure. In
previous studies, it was reported that the k value of mono- and disaccharides
increases linearly with an increase in Tg (Roos 1995; Crowe et al. 1996). However,
k of three amorphous inulin samples decreased linearly with increases in
Tg (R2 ¼ 0.9976). Polysaccharides have generally a higher Tg and lower k than
those of disaccharides, and thus the relationship between k and Tg will differ
between mono- and disaccharides and oligo- and polysaccharides (Roos 1995).
The empirical relationship shown in the inserted figure remains an outstanding
theoretical problem but will be of practical use in the prediction of the effect of
water content on amorphous inulin Tg.
The Tg increased with an increase in the MW of inulin. In order to understand the
effect of MW on the Tg, the Tg of anhydrous amorphous inulin including previous
results (Hinrichs et al. 2001; Ronkart et al. 2009) was potted against MW (Fig. 3).
For comparison, the Tg values of anhydrous fructose and sucrose (Levine and Slade
1988; Orford et al. 1989; Kawai et al. 2005) were also shown there. It is well known
that the effect of MW on Tg of amorphous polymer can be described by Fox-Flory
equation (Eq. 2):
Prediction Approach to the Glass Transition Temperature of Amorphous. . . 497
Tg (K)
350
250
0 1000 2000 3000 4000 5000
MW (g/mol)
B
T g ¼ T g1 ð2Þ
MW
where Tg1 and B are maximum anhydrous Tg and constant, respectively. From the
fitting result, Tg1 and B were determined to be 421.54 and 31,016, respectively.
The obtained curve (R2 ¼ 0.93413), however, did not appear to be reasonably
consistent with the data. As the other approach, a stretched exponential function
(Eq. 3) could be employed:
Tg A n
1 ¼ exp ð3Þ
T g1 MW
300
250
0
2
4
6
8
10 200
12
14 5000
Water content (%, w/w) 16 4000
18 3000
2000 MW (g/mol)
1000
Acknowledgments This work was supported by Japan Society for the Promotion of Science
(grant-in-aid for young scientists (B); 21780126).
References
Crowe LM, Reid DS, Crowe JH (1996) Is trehalose special for preserving dry biomaterials?
Biophys J 71:2087
Hinrichs WLJ, Prinsen MG, Frijlink HW (2001) Inulin glasses for the stabilization of therapeutic
proteins. Int J Pharm 215:163
Johari GP, Hallbrucker A, Mayer E (1987) The glass–liquid transition of hyperquenched water.
Nature 330:552
Kawai K, Hagiwara T, Takai R, Suzuki T (2005) Comparative investigation by two analytical
approaches of enthalpy relaxation for glassy glucose, sucrose, maltose, and trehalose. Pharm
Res 22:490
Le Meste M, Champion D, Roudaut G, Blond G, Simatos D (2002) Glass transition and food
technology: a critical appraisal. J Food Sci 67:2444
Levine H, Slade L (1988) Principles of “cryostabilization” technology from structure/property
relationships of carbohydrate/water systems—a review. Cryo-Letters 9:21
Orford PD, Parker R, Ring SG, Smith AC (1989) Effect of water as a diluent on the glass transition
behaviour of malto-oligosaccharides, amylase and amylopectin. Int J Biol Macromol 11:91
Ronkart SN, Paquot M, Fougnies C, Deroanne C, Blecker CS (2009) Effect of water uptake on
amorphous inulin properties. Food Hydrocoll 23:922
Roos YH (1995) Phase transitions in foods. Academic, London, pp 109–156
Zimeri JE, Kokini JL (2002) The effect of moisture contention the crystallinity and glass transition
temperature of inulin. Carbohydr Polym 48:299
Effect of Sugars on the Release of Aroma
Compounds in Model Systems
Abbreviations
1 Introduction
Kinetic and thermodynamic mechanisms control the rate of release and the con-
centration of volatiles in the vapor phase in equilibrium with a food. Chemical
nature of the aroma compounds, composition, and structure of foods are main
characteristics that influence the transfer of aroma compounds within the foods
and their release (Seuvre et al. 2006; Cayot et al. 2008). Besides the intrinsic
characteristics of aroma and food matrix and the environmental factors, interactions
occurring between volatile and nonvolatile compounds (in particular those with
food macromolecules like proteins, hydrocolloids, and carbohydrates) are major
factors that limit the release of volatiles, and thus their perception, as demonstrated
by many studies in this field.
In the case of carbohydrates, interactions with aroma compounds in general
demonstrate weak energy, depending on various factors such as the nature and
concentration of both aroma and carbohydrates that, in turn, affect both the kinetics
and the concentration in the vapor phase in equilibrium above the matrix. While
complex saccharides (e.g., starch) have been widely investigated for their role in
affecting the release–retention of volatiles in food matrices, scarce attention has
been given to mono- and disaccharides. In simple model systems, some studies
(Roberts et al. 1996; Nahon et al. 2000) demonstrated that the latter solutes can
induce an increase of release for some aroma compounds, especially those that are
more polar, while an opposite effect was observed for those that are more nonpolar.
In sucrose solutions at 65 % (w/v), Friel et al. (2000) found evidence of volatiles
that showed a “salting out” effect (linalool) and others that showed a “salting in”
effect (ethyl decanoate), while others remained not influenced by the solute
(acetaldehyde).
Physical (e.g., viscosity) and physicochemical properties (e.g., water activity) of
the matrix have been recognized as factors that could affect the liquid–vapor
partition and the kinetics of the release of aroma compounds, but these factors
have been studied only to a limited extent. This study was thus aimed to investigate
the release of volatiles having different hydrophilicities (ethyl acetate, ethyl buty-
rate, isopentyl acetate, ethyl hexanoate) from solutions made of either saccharides
(glucose, sucrose, trehalose) or sorbitol, at different solute concentrations as well as
the related kinetics.
The release and release rate of four volatile esters (ethyl acetate (EA), ethyl butyrate
(EB), isopentyl acetate (IA), and ethyl hexanoate (EH)) (Table 1) were investigated
in aqueous binary (water–saccharide) solutions individually prepared with glucose,
sucrose, trehalose, and sorbitol at solute concentration in the range 0–40 % w/v
(glucose and sorbitol) or 45 % w/v (sucrose and trehalose). A mixture of volatiles
was added in the sugar solution to make up a final concentration in the solution of
0.1 % v/v for EA, EB, and IA and of 0.05 % for EH.
Effect of Sugars on the Release of Aroma Compounds in Model Systems 501
where AH2O ¼ peak area of the volatile in water and Asol is the peak area of the same
volatile in the sugar solution.
Kinetic curves of the release of volatiles from sucrose and trehalose solutions
were obtained by analyzing the vapor phase of samples under static conditions at
different times after closure (0.5, 1, 3, 5, 10, 20, 30, 60 min). Only one sample of the
gas phase was obtained per vial, fitting the peak area of each aroma during time.
GC data were converted in partial pressure values according to Pittia
et al. (2006). Initial apparent rate of release, Rappi (min1), of each aroma com-
pound was computed in agreement with the experimental approach of Seuvre
et al. (2007), taking into account the first two experimental data points
corresponding to 0.5, and 1 min. Rappi was computed for each volatile compound
by using three repetitions for each sampling point.
502 P. Pittia et al.
Analyses were carried out using a PerkinElmer (mod. AutoSystem XL, Boston,
MA, USA) gas chromatograph connected with a flame ionization detector (FID)
detector and equipped with a capillary column Megawax (PerkinElmer, Boston,
MA, USA; inner dia., 250 μm; length, 25 m; particle size, 0.1–0.15 μm). Operating
conditions were as follows: gas carrier H2 at a pressure of 14 bar; injection
temperature, 200 C; and FID temperature, 200 C; temperature of the column
was kept at 40 C for 2 min, followed by an increase of 5 C/min until a final
temperature of 55 C was reached. Identification of compounds was carried out by
comparison with standards. Measurements were done in triplicate.
2.5 Viscosity
a b
50 50
0 0
-50 -50
-100 -100
-150 -150
-200 -200
1 0,99 0,98 0,97 0,96 0,95 1 0,99 0,98 0,97 0,96 0,95
aw aw
c d
50 50
0 0
-50 -50
-100 -100
-150 -150
-200 -200
1 0,99 0,98 0,97 0,96 0,95 1 0,99 0,98 0,97 0,96 0,95
aw aw
Fig. 1 Retention index R (%) of the volatile compounds in sweetener solutions as a function of aw
values of sugar solutions (concentration 0–45 % p/v). (a) EA, (b) EB, (c) IA, (d) EH
volatility of volatile compounds being more marked for the more polar EA than for
EH (Fig. 2). On the other hand, at the same aw value, sorbitol resulted in solute with
less effect on the release of all aroma compounds.
The effect of aw on aroma release is in agreement with previous studies that
indicate that a decrease of volume of nonassociated water at increasing solute
concentration enhances partition in the vapor phase (Rabe et al. 2003). However,
results obtained in these experiments provide evidence that the water state and the
physicochemical parameter aw cannot simply be considered as a good index to
explain the so called “salting out” effect in flavor release due to an increase of a
water binding solute.
To further investigate these results, liquid–vapor partition coefficients (k) of the
four volatile compounds in sucrose and trehalose solutions at increasing solute
concentration were obtained, and an increase of k for EH and IA at increasing sugar
concentration was observed; however, in solutions at sucrose concentration
>45 % w/v, a significant decrease of k occurred for EH, suggesting that changes
in physicochemical properties other than water–solute interactions in the solution
occurred which also affected the release–retention of more hydrophobic volatiles.
Experimental conditions allowed the study of the release rate of aroma com-
pounds from disaccharide solutions resulting from diffusion phenomena through
the matrices and transfer of aroma compounds toward the vapor phase and diffusion
in the vapor phase. The initial apparent rate of release Rappi (min1) of the four
aroma compounds from water and sucrose and trehalose solutions at increasing
concentrations was then computed.
504 P. Pittia et al.
6,0E+05
EA_GC area (a.u.) 5,0E+05
4,0E+05
3,0E+05
S_40%
2,0E+05 S_45%
1,0E+05 T_40%
W
0,0E+00
0 10 20 30 40 50 60
time (min)
1,4E+05
1,2E+05
EH_GC area (a.u.)
1,0E+05
8,0E+04
6,0E+04 S_40%
4,0E+04 S_45%
T_40%
2,0E+04
W
0,0E+00
0 10 20 30 40 50 60
time (min)
Fig. 2 Release of ethyl acetate (a) and ethyl hexanoate (b) as a function of time in water and in
sucrose (40 and 45 %) and trehalose (40 %) solution. W water, S_40% sucrose 40 %, S_45%
sucrose 45 %, T_40% trehalose 40 %, T_45% trehalose 45 %
From water (Table 1), the quantity of released flavor compounds, based on
partial pressures of the esters in the headspace, followed the volatility of the
volatiles, with the exception of EH, which was added in the flavored solution in a
lower concentration to avoid solubility constraints in the higher concentrated sugar
solutions. The initial apparent rate of release Rappi (min1) of flavor compounds in
water and solutions of sucrose and trehalose at different solute concentrations, as
well as the ratio Rappi in solution/Rappi in water, is reported in Tables 2 and 3,
respectively.
In water, it was observed that the rate of release of volatile esters was in
ascending order as follows: EB < IA < EA < EH. The most hydrophobic ester
(EH) showed a higher rate of release than the most hydrophilic (EA and EB) and
the less volatile ester (IA). On the contrary, these last esters were more able to
establish interactions with water molecules and have a lower tendency to partly to
the vapor phase. When disaccharide solutions were taken into account, a general
decrease of Rappi was observed, with the exception of the trehalose solutions above
40 % w/v, where the initial apparent rate of release of all the aroma compounds
showed a significant increase, likely related to major changes in the physicochem-
ical properties of the solution approaching its solubility limit.
Effect of Sugars on the Release of Aroma Compounds in Model Systems 505
Table 2 Initial apparent rate of release Rappi (min1) of flavor compounds in water and solutions
of sucrose and trehalose at different solute concentrations
Initial apparent rate of release Rappi (min1)
Aroma compound Ethyl acetate Ethyl butyrate Isopentyl acetate Ethyl hexanoate
Water 0.486 0.411 0.414 0.502
Sucrose, 10 % 0.327 0.261 0.248 0.285
Sucrose, 40 % 0.309 0.259 0.256 0.371
Sucrose, 45 % 0.296 0.247 0.256 0.352
Trehalose, 10 % 0.431 0.359 0.334 0.403
Trehalose, 40 % 0.304 0.258 0.264 0.401
Trehalose, 45 % 0.604 0.536 0.527 0.629
Table 3 Ratio of initial apparent rate of release Rappi of flavor compounds in sugar solution to
initial apparent rate of release Rappi of flavor compounds in water
Rappi in solution/Rappi in water ratio
Aroma compound Ethyl acetate Ethyl butyrate Isopentyl acetate Ethyl hexanoate
Water 1 1 1 1
Sucrose, 10 % 0.67 0.64 0.60 0.57
Sucrose, 40 % 0.64 0.63 0.62 0.74
Sucrose, 45 % 0.61 0.60 0.62 0.70
Trehalose, 10 % 0.89 0.86 0.81 0.80
Trehalose, 40 % 0.63 0.63 0.62 0.80
Trehalose, 45 % 1.24 1.30 1.27 1.25
In the case of sucrose solutions, in general, a steep and significant decrease with
respect to water of the Rapp of all the four aroma compounds was observed when
10 % w/v solutions were considered, while further increasing concentrations did not
induce major effects on the initial rate of release, although a slight decrease was still
observed for EA, EB, and EH. Furthermore, in the case of EH, a nonlinear
decreasing trend in the Rapp at increasing sucrose concentration was observed,
although the initial rate of release was lower than that in water.
For trehalose, a decreasing trend of the Rappi of the four aroma compounds at
increasing solute concentration was also shown; however, with respect to sucrose,
some differences were observed. In particular, the decrease in the initial apparent
rate was of lower entity (Table 3) in the 10 % trehalose solutions than that
determined in the sucrose solutions at similar concentration. On the contrary, at
higher solute concentration (40 %) for the Rapp, values were not significantly
affected by sugar type, with the exception of EH, which did not change its initial
rate of release due to an increased trehalose concentration.
Under these experimental conditions, results are attributed to the aroma mole-
cules that vaporize as they become present at the vapor–water interface or could
reach it in very short time (1 min). Thus, decrease in the initial apparent rate of
release could be related in general to a reduction of the diffusion of the flavor
506 P. Pittia et al.
compounds within the disaccharide solutions due to the increased viscosity. How-
ever, at low concentration (10 %), the viscosity of sucrose and trehalose is quite
similar, while a significantly different result occurred in the release rate of aroma
compounds in the two sugar solutions. In contrast, at higher concentration (40 %),
trehalose solutions are characterized by a higher viscosity than the correspondent
sucrose solutions, while the relative decrease of the Rapp of the compounds (Table 3)
was similar.
It could be suggested that the chemical and physicochemical properties of the
sucrose and trehalose aqueous solutions and their change in increasing concentra-
tion, other than viscosity, can affect aroma diffusion in the matrix and, thus, their
release kinetics, which may imply effects on water–aroma, aroma–sugar, and
sugar–water interactions, the nature of the aroma, and its specific affinity for the
solute and, eventually, retention phenomena.
4 Conclusions
This study, carried out with simple sugar model systems at increasing concentra-
tion, evidenced that release in the vapor phase of volatile compounds is affected by
the water state, but the nature of the solute, at the same water activity value, could
differently favor or hinder partition. Regarding the kinetics of the release, viscosity
of the matrix could be considered as a factor that limits the rate of release in the
vapor phase of aroma compounds in sugar solutions at increasing concentration,
although other physicochemical and physical properties must also be considered.
Based on these results, aroma compounds can be considered as showing reliable
proof of physicochemical changes occurring in the liquid phase of saccharide
solutions; thus, aroma release is an interesting tool for further investigation.
References
Roberts DD, Elmore JS, Langley KR, Bakker J (1996) Effects of sucrose, guar gum and
carboxymethylcellulose on the release of volatile flavour compounds under dynamic
conditions. J Agric Food Chem 44:1321–1326
Seuvre AM, Philippe E, Rochard S, Voilley A (2006) Retention of aroma compounds in food
matrices of similar rheological behaviour and different compositions. Food Chem 96:104–114
Seuvre AM, Philippe E, Rochard S, Voilley A (2007) Kinetic study of the release of aroma
compounds in different models systems. Food Res Int 40:480–492
Wetting Behavior of Chitosan Solutions
on Blueberry Epicarp With or Without
Epicuticular Waxes
Abbreviations
CA Contact angle
SFE Surface free energy
vOCG Theory of van Oss, Chaudhury, and Good
Wa Work of adhesion of a liquid on a solid surface
γ LV Surface tension of the liquid
θ Contact angle
γ LVLW Apolar component of liquid–vapor
γ SVLW Apolar component of surface–vapor
Mw Medium-molecular-weight chitosan
Wc Cohesive energy
Seq Spreading coefficient
Swith wax Spreading without chloroform wash
Swithout wax Spreading with chloroform wash
Sdiff Spreading difference
G Glycerol
O. Skurtys (*)
Department of Mechanical Engineering, Universidad Técnica
Federico Santa Marı́a, Santiago, Chile
e-mail: [email protected]
P. Velásquez
Departamento de Ingenierı́a Quı́mica y Bioprocesos, Pontificia Universidad Cat
olica de Chile
(PUC), Avenida Vicu~na Mackenna, 4860 Santiago, Chile
F. Osorio
Department of Food Science and Technology, Universidad de Santiago de Chile,
Avenida Ecuador 3769 Santiago, Chile
1 Introduction
Edible coatings can prevent microbial spoilage and quality loss of blueberries due
to mass transfer (e.g., moisture, gases, aroma compounds, etc.). The protection of
blueberry (and in foods in general) by edible coatings depends on controlling the
wetting of the coating solutions, which affects the adhesion and the thickness of the
film (Park 2002). Indeed, the thickness of edible coatings is an important parameter
since it directly affects the biological properties and the shelf-life of blueberries.
The central parameter that characterizes the wetting of a blueberry epicarp is the
equilibrium contact angle (CA) (Israelachvili 1992). By convention, the CA is the
angle, θ, at which a liquid–vapor interface meets the solid surface. Surface free
energy (SFE) of blueberry epicarp can be obtained using different approaches. All
of these methods are based on CA measurements of different sessile liquid drops
with known surface tension deposited on the blueberry epicarp, but discrepancies
have been shown in the obtained results (Etzler 2006). One of the main methods for
the determination of SFE is the “acid–base” theory of van Oss, Chaudhury, and
Good (vOCG) (van Oss et al. 1986, 1988).
Application of chitosan in the form of edible coatings on blueberries may be an
effective and affordable treatment to extend shelf-life and enhance quality and
safety of the produce (Park 2002). Chitosan shows antifungal and antibacterial
properties, which are believed to originate from its polycationic nature
(Tharanathan and Kittur 2003). Chitosan coatings, such as many polysaccharide-
based films, tend to exhibit fat and oil resistance and selective permeability to gases,
but lack resistance to water transmission (Bordenave et al. 2007). Dutta et al. (2009)
reviewed the many applications of chitosan coatings. Plasticizers are generally
required for polysaccharide-based edible films to improve elasticity and flexibility,
reduce brittleness and toughness, and prevent cracking during handling and storage.
The most common used plasticizers are polyols like propylene glycol, sorbitol,
polyethylene glycol, and glycerol. Surfactants can be incorporated into the film
formulation to reduce surface tension of the solution, improving the wettability and
adhesion of the film (Goodwin 2004). Surfactants are usually organic compounds
that are amphiphilic, meaning they contain both hydrophobic groups (apolar) and
hydrophilic groups (polar). A typically used surfactant is Tween 20, which has been
reported to decrease the water vapor permeability of coatings (Nieto 2009).
The aim of this study was to investigate the physical properties of blueberry
epicarp in order to increase the effectiveness of chitosan-based coatings. SFE of
blueberry epicarp with and without wax was calculated using the “acid–base”
theory. Various chitosan-based coatings were applied to the blueberry epicarp. In
particular, the effects of concentrations of a plasticizer (glycerol) and a surfactant
(Tween 20) on the wetting of blueberry epicarp were studied. For that, the cohesive
energy and the spreading coefficient were calculated for each solution. A mathe-
matical relationship between concentrations and spreading coefficient was pro-
posed. Finally, the optimum chitosan-based coating solutions were determined to
promote the adhesion of the edible film.
Wetting Behavior of Chitosan Solutions on Blueberry Epicarp With or Without. . . 511
Fresh highbush blueberries of the variety Duke (Vaccinium corymbosum L.) were
supplied by Chilean producers in the province of Santiago (Hortifrut SA, Chile).
Before measurements, blueberries were left at ambient temperature (~21 C). The
fruits were carefully selected to insure uniformity in maturity, size, color, and
physical appearance according to visual analysis. They were almost spherical,
between 11 and 15 mm in diameter, with a dark blue color. The blueberry cuticle
consists of a polyester matrix or cutin layer (about 5 μm thick) and an epicuticular
wax layer or waxy bloom. Epicuticular wax of undamaged fruits was removed by
gentle rinses of the blueberries’ surface with chloroform (Schreiber and Sch€onherr
2009).
SFE of each fruit epicarp was determined following wetting measurements using
the contact angle technique coupled with a visualization technique. The core of the
self-built instrument consists of an optic device and a light source mounted on a
rigid support. The optics comprised a zoom video lens (Edmund Optics, NJ, USA)
connected to a CCD camera (Pulnix, Inc., San Jose, CA, USA), which is piloted via
software. The drops were manually deposited on fruit epicarps using a precision
microliter pipette (Gilson Pipetman U2). The adequate drop volumes were chosen
between 0.2 and 2 μl to avoid two experimental problems: first, the drop must be
sufficiently smaller to be placed on a planar surface of the fruit epicarps; second, the
drop volume was sufficiently small to assume the effect of gravity to be negligible
on the drop shape. Five different liquids with known surface energy components
(polar and nonpolar liquids) were used for the surface energy measurements:
α-bromonaphthalene, ethylene glycol, formamide, glycerol, and distilled water.
These liquids span the entire range of fluids, from dispersive to polar. CA were
measured for each liquid on the two different surfaces: blueberry epicarp with and
without wax. For statistical approaches, for each liquid at least 10 (mostly 15) right
and left angles were measured. The root mean square of each set of measurements
was smaller than 2.3 . All of the CA measurements were made at room temperature
(~21 C). In order to avoid liquid adsorption by the fruit, the typical CA equilibrium
measurement was performed within 30–60 s after the drop placement on the
surface. CA values were determined using ImageJ program with the plug-in drop
shape analysis. The accuracy of the analysis is high since the plug-in used a method
based on B-spline snakes (active contours) (Drop-analysis 2010).
512 O. Skurtys et al.
and γ are the electron–donor (Lewis base) and the electron–acceptor (Lewis acid).
+
The material is considered as a polar if both γ and γ + are close to zero, monopolar
if one of these components is negligible, and bipolar if the two components have
appreciable values. The work of adhesion of a liquid on a solid surface is written
Wa ¼ γ LV ð1 þ cosθÞ ¼ 2ðγ LV LW γ SV LW Þ þ 2ðγ LV þ γ SV Þ þ 2ðγ LV γ SV þ Þ
1=2 1=2 1=2
where the surface tension (γ LV) of the liquid and the contact angle θ between the
solid surface and the liquid–vapor interface may be assessed experimentally.
However, concerning θ, the difficulty arises from the nonideal solid surfaces
where measured static CA values may exhibit a large dispersion (CA hysteresis).
In the last equation, γ SVLW, γ SV , and γ SV þ are unknown. But if one assumes that
the three dispersive and acid–base coefficients γ LV,i are known for each liquid
(i ¼ 1, 2, 3), a linear set of three equations can be obtained in the form
1=2 1=2 1=2
γ LV, i ð1 þ cosθÞ ¼ 2 γ LV, i LW γ SV LW þ 2 γ LV, i þ γ SV þ 2 γ LV, i γ SV þ .
This system of equations, Ax ¼ B, is generally ill conditioned: small differences in
experimental data or in the coefficients can induce strong variations in the results.
One apolar (C10H7Br) and two polar (CH3NO) and water liquids constitute a good
set of the three fluids with a condition number equal to 7.13, as reported by Della-
Volpe et al. (2004). Calculations were performed using the GNU free software
Octave with the front-end user interface QtOctave.
Interfacial tension measurements, γ LV, were carried out by the pendant drop
method. A small drop of liquid (about 10–20 μl) attached to the tip of a stainless-
steel needle (inner diameter of 1.651 0.005 mm) was suspended into air under a
constant temperature (~21 C). The drop was formed by a controlled syringe pump.
Images of the drop were taken as function of time with the optic device. The
solution of the Laplace equation was obtained with the free software Octave
3.2.4. The accuracy of the measurements was 0.3 mJ/m2.
Blueberry with epicuticular wax had a SFE γ SV ¼ 29.1 4.1 mJ/m2. In both cases,
blueberry surface is a low-energy surface. In comparison with other fruits (with
wax), like Granny Smith apple (γ SV ¼ 43.5 mJ/m2) or tomato (γ SV ¼ 36.9 mJ/m2),
blueberries (with wax) have a much lower surface energy, i.e., an epicarp more
hydrophobic. However, this value was close to the SFE of strawberry, measured by
the Owens–Wendt approach, 28.94 mJ/m2 (Ribeiro et al. 2007). After the wax
removal, SFE increased significantly to 38.3 2.8 mJ/m2. For a waterdrop (epicular
with and without wax), the contact angle was θ ¼ 124.9 and 100.9 respectively.
Therefore, epicarp with wax was more hydrophobic. Blueberry epicarps with or
without wax have mainly an apolar (Lifshitz–van der Waals) character with a small
polarity (Lewis acid–base). The apolar character is generally observed for the
hydrocarbons, which dominantly have almost zero polar, but van der Waals attrac-
tive forces between molecules. Apolar surfaces are surfaces where hydrophobicity
is observed; the interactions among the water molecules themselves exceed the
water–surface interactions (Israelachvili 1992). Moreover, in both cases, the results
indicate that the surfaces have low acidic and basic characters, implying that these
epicarps are slightly bipolar. The acidic ðγ SV þ Þ and basic ðγ SV Þ are 1 (with wax)–
0.9 mJ/m2 (without wax) and 1.6 (with wax)–1.8 mJ/m2 (without wax), respec-
tively. The wetting of blueberry epicarps varies not only with chemical texture
determined by the composition of the solid surface but also with the physical
texture or the roughness. The roughness of the blueberry epicarp was enhanced
by the three-dimensional wax structures and was measured at about 42 10 μm.
This value is in the range of measured values for three-dimensional epicuticular
waxes on plant surfaces between 0.5 and 100 μm (Jeffree 2006). Effect of surface
roughness on the wetting is easily understood by the Wenzel equation. According to
this equation, hydrophobicity is reinforced by roughness.
514 O. Skurtys et al.
Fig. 1 Cohesive energy Wc as a function of chitosan and (a) glycerol and (b) Tween
20 concentrations
ð2Þ
As strong adhesion and weak cohesion result in a high degree of wetting, the
minimum of Eqs. 1 and 2 was calculated using the software Octave (2010).
Wcglyc/chit ¼130 mJ/m2 for the solution [C]¼1 % and [G]¼20 % and
Wctween/chit ¼68 mJ/m2 for the solution [C]¼1 % and [T]¼20 %.
Wetting Behavior of Chitosan Solutions on Blueberry Epicarp With or Without. . . 515
Fig. 2 Spreading coefficient S as a function of chitosan and glycerol concentrations: (a) with wax,
(b) without wax, (c) spreading coefficient difference Sdiff ¼ Swith wax Swithout wax
516 O. Skurtys et al.
The reason why S was smaller on the wax component of cuticle than on the cutin
component of cuticle is that polar groups are capable of binding aqueous solutions
through hydrogen bonds, while hydrocarbon chains of wax only attract aqueous
solutions through the much weaker van der Waals forces (Israelachvili 1992). Thus,
CA between aqueous chitosan solution and the epicuticular wax was slightly larger
than those between water and the cutin component of cuticle. The influence of glycerol
[G] and chitosan [C] concentrations on the spreading factor followed the equations:
glyc=chit
pffiffiffiffiffiffi pffiffiffiffiffiffi pffiffiffiffiffiffi pffiffiffiffiffiffi
Swith wax ¼ 37:87 þ 11:86½C 3 ½T 48:11 3 ½Cþ0:83 3 ½C27:27 3 ½T ; R2 ¼ 0:64
ð3Þ
glyc=chit
p ffiffiffiffiffiffiffi pffiffiffiffiffiffi pffiffiffiffiffiffi pffiffiffiffiffiffiffi
Swithout wax ¼73:84þ13:21½C 5
½G11:58 5 ½Cþ0:59½G 5 ½C25:70 5 ½G; R2 ¼ 0:60
ð4Þ
Fig. 3 Spreading coefficient S as a function of chitosan and Tween 20 concentrations: (a) with
wax, (b) without wax, (c) spreading coefficient difference Sdiff ¼ Swith wax Swithout wax
With epicuticular wax [C]¼ 1.4 % and [T]¼20 %, and without epicuticular wax
[C]¼1 % and [T]¼ 20 %. However, at this concentration, a phase separation during
drying was observed, i.e., films obtained were of poor quality. This is due to the lack
of strong interactions between the surfactant and the chitosan.
4 Conclusions
In this paper, the physical properties of fresh highbush blueberry epicarp were
determined in order to increase the effectiveness of chitosan-based coatings. SFE
of blueberry epicarp, with and without epicuticular wax, was a low-energy surface
and presented an apolar character and a small polarity. Indeed, with wax
γ SV ¼ 29.1 4.1 mJ/m2, while after the wax removal, γ SV was increased signifi-
cantly to 38.3 2.8 mJ/m2. Various chitosan-based coating solutions were applied
to the blueberry epicarp with or without epicuticular wax. The effects of glycerol
[G] and Tween 20 [T] concentrations on the spreading coefficient S have been
studied. For the glycerol/chitosan coating solutions, the results showed that the
addition of glycerol to the chitosan solution diminished the wetting capacity of
518 O. Skurtys et al.
References
Bordenave N, Grelier S, Coma V (2007) Water and moisture susceptibility of chitosan and paper-
based materials: structure-property relationships. J Agric Food Chem 55:9479–9488
Della-Volpe D, Maniglio M, Brugnara S, Siboni MM (2004) The solid surface free energy
calculation. I. In defense of the multicomponent approach. J Colloid Sci 271:434–453
Drop-analysis (2010) Http://bigwww.epfl.ch/demo/dropanalysis/soft. Accessed March
Dutta P, Tripathi S, Mehrotra G, Dutta J (2009) Perspectives for chitosan based antimicrobial films
in food applications. Food Chem 114:1173–1182
Etzler FM (2006) Contact angle, wettability and adhesion. Chapter surface free energy of solids: A
comparison of models, edited by K- L. Mittal. Utrech, The Netherlands, V.S.P. Intl Science,
pp 215–236
Goodwin J (2004) Colloids and interfaces with surfactants and polymers: an introduction. Wiley,
West Sussex, England
Israelachvili J (1992) Intermolecular and surface forces. Academic, New York
Jeffree C (2006) Biology of the plant cuticle. The fine structure of the plant cuticle. Oxford, UK:
Blackwell Publishing Ltd, pp 11–125
Nieto M (2009) Edible films and coatings for food applications. Structure and function of
polysaccharide gum-based edible films and coatings. Springer, Berlin, pp 57–112
Octave (2010) http://www.gnu.org/software/octave/
Park HJ (2002) Fruit and vegetable processing: improving quality. Edible coatings for fruits.
Edited by: W. Jongen Woodhead Publishing Limited, Cambridge, pp 331–355
Ribeiro C, Vicente A, Teixeira J, Miranda C (2007) Optimization of edible coating composition to
retard strawberry fruit senescence. Postharvest Biol Technol 44:63–70
Schreiber L, Sch€ onherr J (2009) Water and solute permeability of plant cuticles. Springer Berlin
Heidelberg
Tharanathan RN, Kittur F (2003) Chitin—the undisputed biomolecule of great potential. Crit Rev
Food Sci Nutr 43:61–87
van Oss CJ, Chaudhury MK, Good RJ (1988) Interfacial Lifshitz-van der Waals and polar
interactions in macroscopic systems. Chem Rev 88:927–941
van Oss CJ, Good RJ, Chaudhury MK (1986) The role of van der Waals forces and hydrogen bonds
in hydrophobic interactions between biopolymers and low energy surfaces. J Colloid Interface
Sci 111:378–390
Influence of Brine Concentration on Moisture
and NaCl Transport During Meat Salting
Abbreviations
1 Introduction
et al. 2001). Thus, brining with low NaCl content produces growth of the WHC
linked to the protein solubilisation process, a phenomenon known as “salting in.”
On the other hand, when high brine concentrations are used, a decrease in WHC is
observed (“salting out”), probably due to the insolubilization of proteins (Graiver
et al. 2009). Therefore, the solution concentration used can be expected to influence
the mass transfer kinetics.
Mathematical modelling constitutes an interesting tool with which to quantify the
influence of process variables, such as brine concentration, on transport phenomena.
The transport of moisture and solutes in solid-liquid systems has been modelled
using empirical models, such as the Peleg or Weibull models (Garcı́a-Pascual
et al. 2006) or theoretical models, such as the diffusion models based on Fick’s
second law (Cárcel et al. 2007b). Diffusion models are more complex to solve but
provide a more reliable tool for process simulation under different conditions.
In the literature, a great number of research can be found about meat brining
processes using saturated or quasi-saturated brines (Fox 1980; Graiver et al. 2006,
2009; Cárcel et al. 2007a). Nevertheless, few studies exist on the effect of brine
concentration on water and NaCl transport during brining, and these aspects are the
main aim of this work.
Brining kinetics of pork meat samples (Longissimus dorsi) with slab geometry
(length 50 width 30 thickness 10 mm) were carried out at different brine
concentrations (50, 100, 150, 200, 240, and saturated at 5 1 C 280 kg/m3),
with each brining condition being tested in triplicate. In each experiment, samples
were extracted at pre-set times (15, 30, 45, 60, 90, and 120 min), immersed in
distilled water for 20 s to remove superficial brine, and then blotted. The moisture
content of the brined and fresh meat samples was measured in triplicate following
AOAC official method 950.46 (AOAC 1997). NaCl content was quantified by
determining the ion Cl concentration (Ciba Corning, mod. 926L; Halstead,
Essex, UK) in a meat (1 0.1 g) extract obtained with distilled water (100 mL)
of the salted samples (Cárcel et al. 2007a). Each sample from a single experiment
was analyzed in triplicate; thus, nine measurements were carried out to determine a
point in the kinetics. The high number of replicates was used to obtain significant
results by taking into account the variability of the raw material and experimental
procedure. The results were reported as kg NaCl/kg initial dry matter. The multi-
factor analysis of variance (MANOVA) and the estimation of least significance
difference (LSD) intervals were chosen to evaluate the influence of brine concen-
tration on moisture and NaCl experimental values. Statistical analysis was carried
out using the Statgraphics Plus 5.1 software package.
Influence of Brine Concentration on Moisture and NaCl Transport. . . 521
Diffusion models based on Fick’s second law were used to describe water and NaCl
transport, respectively. Equation (1) shows the diffusion equation for water trans-
port. An analogous equation was used for NaCl transport:
!
2
∂Cw ðx; tÞ ∂ Cw ðx; tÞ
¼ Dw ð1Þ
∂t ∂x2
where Cw is the local moisture content (kg water/kg dry matter), Dw is the average
effective moisture diffusivity (m2/s), t is the time (s), and x is the characteristic
coordinate in the infinite slab geometry (m).
Equation (1) was solved by assuming the symmetry of the solid, a uniform initial
moisture content and temperature, and that the effective moisture diffusivity and
the sample volume remain constant during brining (Graiver et al. 2006). The
effective diffusivity values of both moisture and NaCl were identified by using an
optimization procedure, i.e., the generalized reduced gradient (GRG), available in
Microsoft Excel™ spreadsheet from MS Office 2007. The explained variance (%
VAR) was computed to determine the ability of the model to describe the exper-
imental data (Cárcel et al. 2007b).
" #
S2tw
% VAR ¼ 1 2 100 ð2Þ
Sw
where S2w and S2tw are the variance of the sample and the estimation, respectively.
The evolution of sample moisture content during treatments was dependent on the
concentration of brine used, as shown in Fig. 1. It must be emphasized that, in this
figure, each point represents the average of nine measurements; three independently
brined samples were analyzed in triplicate.
As can be observed in Fig. 1, below 200 kg/m3 brine concentration, samples
showed an increase of moisture content, while those treated at brine concentrations
above 200 kg/m3 showed a decrease. For samples brined at 200 kg/m3, no net transport
of moisture was identified. Similar results were reported by Graiver et al. (2009). From
the ANOVA, it could be stated that both factors, time and brine concentration, were
522 C. Ozuna et al.
significant ( p < 0.05), and from the estimation of the LSD intervals, three significantly
different groups linked to treatments were identified. The samples showed a
significant ( p < 0.05) water gain in the first group (brine concentrations of 50 and
100 kg/m3), dehydration in saturated brine (280 kg/m3), and moisture content which is
not significantly different from that of fresh meat in the third group (intermediate brine
concentrations: 150, 200, 240 kg/m3).
From previous measurements, the NaCl content of fresh meat was considered
negligible. The posttreatment NaCl content in meat samples was dependent on
the brining time and the brine concentration (Fig. 2). As in the case of moisture
content, each point shown in this figure represents the average of nine measure-
ments; three independently brined samples were each analysed in triplicate.
The ANOVA showed brining time and brine concentration and their interaction
to be significant ( p < 0.05). From the estimation of LSD intervals, the average NaCl
content of samples from every brining concentration tested was found to be
significantly ( p < 0.05) different, except for 200 and 240 kg/m3.
The results of fitting the diffusion model to the experimental moisture data
are shown in Table 1. The low value of the percentage of explained variance
Influence of Brine Concentration on Moisture and NaCl Transport. . . 523
Table 1 Effective diffusivity and percentage of explained variance for water and NaCl transport
in pork meat brined in different brine concentrations (5 0.5 C)
Brine concentration Diffusivity (Dw) % Diffusivity (DNaCl) %
(kg/m3) (1010 m2/s) VAR (1010 m2/s) VAR
50 0.76 90.7 1.24 98.1
100 0.62 90.2 2.04 99.2
150 0.17 89.1 2.11 99.0
200 – – 1.99 96.4
240 1.04 81.1 1.73 96.1
280 1.83 90.0 1.96 96.1
(81.1–90.7 %) can be linked to the great variability of the raw material. The similar
trend between the experimental and calculated data (Fig. 3a) permitted the model to
be considered as adequate. No fit was possible for treatments carried out at a brine
concentration of 200 kg/m3 due to the fact that no net transport was identified.
The identified effective moisture diffusivity values decreased when brine con-
centration rose, from positive values (0.76 1010 m2/s) at the lowest brine
concentration, 50 kg/m3, to negative (1.83 1010 m2/s) at the highest,
280 kg/m3. The increase or decrease of the diffusion parameter shows the change
in the direction of moisture transport. The hydration of meat samples, when treated
at the lowest brine concentration tested (50 kg/m3), could be attributed not only to
osmotic pressure but also to the “salting in” phenomenon (Nguyen et al. 2010)
produced by protein solubilization. On the other hand, dehydration of samples
treated in saturated brine (280 kg/m3) could also be explained by an increase in
524 C. Ozuna et al.
Fig. 3 Experimental vs. calculated moisture (a) and NaCl (b) content of pork loin slices
(thickness 10 mm)
the osmotic effect due to the “salting out” phenomenon (Nguyen et al. 2010)
produced by the insolubilization of proteins. The variation in the effective diffu-
sivity identified for samples treated in intermediate brine concentrations (150, 200
and 240 kg/m3) from positive to negative values could indicate the superposition of
both phenomena. Thus, as brine concentration increased, the phenomenon of
protein solubilization disappears and the phenomenon of protein insolubilization
starts to be relevant. As this fact affected the structure of meat, the identified
diffusivity was also affected.
The considered diffusion model was adequate to describe NaCl transport, as can be
observed in the obtained values of the percentage of explained variance (Table 1).
Moreover, the close agreement between experimental and calculated data can also
be observed in Fig. 3b.
The identified effective diffusivity values were similar for all conditions tested,
with an average value of (1.90 0.28) 1010 m2/s. Because the diffusion model
considers the difference of concentration to be the driving force for transport, and
this variable is the only one that changes during experimentation, the identified
diffusivity should be similar for every treatment (Table 1). Therefore, the identified
constant value of diffusivity could indicate that although the NaCl brine concen-
tration affected the water transport rate, NaCl diffusion was not affected by
solubilization/precipitation of proteins. The values identified are similar to those
reported by other authors (Wood 1966; Fox 1980; Siro et al. 2008).
Influence of Brine Concentration on Moisture and NaCl Transport. . . 525
4 Conclusions
Brine concentration had a significant effect ( p < 0.05) on water and NaCl transport
during meat brining. Hydration or dehydration of samples could be achieved by
using low or high brine concentrations, respectively. The more concentrated the
brine, the greater the gain in NaCl. The effective diffusivity identified for moisture
transport showed that the different direction of moisture transport depended on the
brine concentration. The influence of NaCl concentration on the solubility of pro-
teins and, therefore, on the water-holding capacity of meat could explain this
behavior.
References
AOAC (1997) Official method of analysis of the association of official analytical chemists
international, 16th edn. AOAC, Washington, DC, 3rd revision
Barat JM, Grau R, Ibá~nez JB, Pagán MJ, Flores M, Toldrá F, Fito P (2006) Accelerated processing
of dry-cured ham. Part I. Viability of the use of brine thawing/salting operation. Meat Sci
72(4):757–765
Cárcel JA, Benedito J, Bon J, Mulet A (2007a) High intensity ultrasound effects on meat brining.
Meat Sci 76:611–619
Cárcel JA, Benedito J, Rossello C, Mulet A (2007b) Influence of ultrasound intensity on mass
transfer in apple immersed in a sucrose solution. J Food Eng 78:472–479
Fox JB (1980) Diffusion of chloride, nitrite and nitrate in beef and pork. J Food Sci 45:1740–1744
Garcı́a-Pascual P, Sanjuán N, Melis R, Mulet A (2006) Morchella esculenta (morel) rehydration
process modelling. J Food Eng 72:346–353
Graiver N, Pinotti A, Califano A, Zaritzky N (2006) Diffusion of sodium chloride in tissue. J Food
Eng 77:910–918
Graiver N, Pinotti A, Califano A, Zaritzky N (2009) Mathematical modeling of the uptake of
curing salts in pork meat. J Food Eng 95:533–540
Lambert IH, Nielsen JH, Andersen HJ, Ortennland N (2001) Cellular model for induction of drip
loss in meat. J Agr Food Chem 49:4876–4883
Nguyen MV, Arason A, Thorarinsddottir KA, Thorkelsson G, Gudmundsd ottir A (2010) Influence
of salt concentration on the salting kinetics of cod loin (Gadus morhua) during brine salting.
J Food Eng 100:225–231
Siro I, Vén C, Balla C, Jonás G, Zeke I, Friedrich L (2008) Application of an ultrasonic assisted
curing technique for improving the diffusion of sodium chloride in porcine meat. J Food Sci
91:353–362
Wood FW (1966) The diffusion of salt in pork muscle and fat tissue. J Sci Food Agr 17:138–140
Relationship Between Electrical Conductivity
and Water Activity of Starch-Water
Composites
Abbreviations
aw Water activity
DSC Differential scanning calorimeter
ANOVA Analysis of variance
EC Electrical conductivity
1 Introduction
swelling, increase of viscosity, and reduction of the area for starch particle movement.
Decreasing water content in the material decreases EC (Lewicki Piotr 2004).
All of these works have shown the important role of water on EC of foods and
particularly on starchy foods. It is known that water is contained in composites as
free water or as bound water, and one way to determine the state of water is to
measure the water activity (Fennema 1985). In spite of abundant reported works
about EC in food material, no studies relating water activity with EC were found in
literature, with the exception of a brief report by Lewicki Piotr (2004).
The present work studies the relationship among electrical conductivity, water
activity, and enthalpy of maize, potato, and rice starch-water composites heated at
different temperatures.
Corn, rice, and potato starches (Sigma Chemical Co.) were used as study systems.
The materials were suspended at different starch-water ratios (30:70, 50:50, and
70:30 w/w). Water activity (aw) was measured using an AquaLab® model C2
(Decagon Devices, Inc. Pullman, USA) at room temperature.
Electrical conductivity was measured using a self-designed resistive heating system
(Morales-Sánchez et al. 2009) composed of a resistive heating cell, a type T thermo-
couple, a Watlow ramp-programmable temperature controller (Model 981, Watlow
Electric Manufacturing Co., St. Louis, MO), an HP 34410A 6½ digital multimeter
(Agilent Technologies), and a PC computer. Heating rate program was set at 5 C/min,
and the studied temperature range was from 30 to 90 C to avoid boiling of water.
A differential scanning calorimeter (DSC Mettler® Toledo model 821e)
equipped with a thermal analysis data station was used. The samples were heated
in a 40 μL pan from 30 to 90 C at a rate of 5 C/min. Enthalpy was computed
automatically in duplicated samples.
Data were analyzed using analysis of variance (ANOVA) procedures. Means
were compared using Tukey’s test (P ¼ 0.05). Statistical analyses were performed
using SAS software. Pearson’s correlation coefficient between EC, aw), and
enthalpy were determined.
Table 1 shows the electrical conductivity, water activity, and enthalpy of corn,
potato, and rice starch-water composites at 50:50 w/w and 25 C.
Rice starch showed higher electrical conductivity than cornstarch, and the latter
was higher than that for potato starch; the same order was observed in water activity
values. Cornstarch had the minimum water activity value, indicating that this type
of starch establishes stronger interactions with water than the other two; it also has
the minimum enthalpy value. Rice and potato starches had no significant (P > 0.05)
Relationship Between Electrical Conductivity and Water Activity. . . 529
Table 1 Electrical conductivity, water activity, and enthalpy of [starch-water] composites for
corn, potato, and rice at 50:50 w/w and 25 C
Starch EC (S/m) aw Enthalpy (J/g)
Corn 0.25178c 0.965b 2.04b
Potato 0.29969b 0.974a 4.00a
Rice 0.99639a 0.973a 2.38b
Different letters in the same column indicate significant difference (P < 0.05)
EC electrical conductivity, aw water activity
4 Conclusions
Acknowledgements The authors want to thank CONACyT and IPN for their financial support
through project SIP-20090967 and SIP 20100209.
References
Fennema OR (1985) Food chemistry, 2nd edn. Marcel Dekker Inc., New York, pp 46–50
Karapantsios TD, Sakonidou EP, Raphaelides SN (2000) Electrical conductance study of fluid
motion and heat transport during starch gelatinization. J Food Sci 65(1):144–150
Lewicki Piotr P (2004) Water as a determinant of food engineering properties. A review. J Food
Eng 61:483–495
Fa-De Li, Li-Te Li, Zaigui Li, Eizo Tatsumi (2004) Determination of starch gelatinization
temperature by ohmic heating. J Food Eng 62:113–120
Relationship Between Electrical Conductivity and Water Activity. . . 531
Abbreviations
Ala-Pro-pNA Alanine-proline-p-nitroanilide
aw Water activity
CP Carboxypeptidases
DCP Dry cacao powder
PVPP Polyvinylpolypyrrolidone
U One unit of enzyme
X Water content
Xaa-Pro-DAP Xaa-prolyl-dipeptidyl-peptidase
1 Introduction
Cacao seeds demonstrate epigeal (or epigeous) germination in which the hypocotyl
elongates and forms a hook, pulling rather than pushing the cotyledons and apical
meristem through the soil. Once it reaches the surface, it straightens and pulls the
cotyledons and shoot tip of the growing seedlings above the ground (Bewley 1997).
After the cocoa seeds are removed from the fruit they germinate rapidly, 4–6 days
after planting. The root and the hypocotyle emerge first, which causes the cotyle-
dons to elevate above the substrate (10 or 15 days after planting). The cotyledons
then open and expose the plumule, which begins to grow at the same time as the
root, but it is much smaller. The first phase of growth is completed with the
maturation of the first leaves (Lopez-Andrade 2003). This means that germination
commences with the uptake of water by dry seed imbibition and is completed when
a part of the embryo, usually the radicle, extends to penetrate the structures that
surround it (Bewley 1997).
Water is the principal factor in germination stages. Uptake of water by a mature dry
seed is developed in three phases, with a rapid initial uptake (imbibition) followed
by a plateau phase (metabolic activation). A further increase in water uptake occurs
only after germination is completed, as the embryonic axes elongate. Because
dormant seeds do not complete germination, they cannot enter to the third phase
(i.e., growth of the radicle).
Phase I: Imbibition. Important structural alterations are temporarily produced with
the entrance of water into the seed, especially at the membrane level. In the dry
seed, phospholipids of the membrane are presented as a gel, which lead to a rapid
leakage of solutes and low-molecular-weight metabolites into the surrounding
imbibition solution. Nevertheless, with re-hydration, the membranes return to
crystalline hydrous status, a stable condition, at which time solute leakage is
curtailed. The rehydration allows turnover or replacement of components occurring
over several hours as full metabolic status is achieved (Bewley 1997).
Phase II: Metabolic activation. During this phase, the synthesis occurs. All of the
components necessary for the resumption of protein synthesis upon imbibition are
present. In the case of reserve proteins, these components are reduced to amino
acids and peptides due to exopeptidases (carboxypeptidase) and endopeptidases.
These enzymes break down the peptides to produce small polypeptides, which are
degraded by a group of enzymes known as peptide hydrolases to produce amino
acids (Herrera 2006).
Phase III: Growth of the radicle. Finally, radicle extension through the structures
surrounding the embryo is the event that terminates germination and indicates the
beginning of seedling growth (Bewley 1997).
Water Content, aw, and Enzyme Activity. . . 535
Seed germination depends on both internal and external conditions. The most
important external factors include temperature, water, oxygen, and sometimes
light or darkness. In the absence of any of these factors, most seeds remain in a
quiescent state, even without a rest state. In the case of recalcitrant seeds, such as
cocoa, there can be a rapid decline in seed longevity. The entry of water into the
seed is due to a difference in water potential between the seed and the surrounding
environment. Normally, the water potential is lower in dry seeds than in the external
environment. Therefore, until the radicle emerges, water reaches the embryo
through the cell walls of the seed coat. Although water is necessary for the
rehydration of seeds, an excess of water can adversely affect germination, as well
as hinder the delivery of oxygen to the embryo (Herrera 2006; Vashisth and
Nagarajan 2010).
Theobroma cacao L. has been described as a plant whose seed contains vicilina-
class globulin, but not legumin, as vacuolar storage protein (Voigt et al. 1993).
Proteolysis is initiated by an endopeptidase with high specificity for seed reserve
proteins, which would be degraded partially as a consequence of the rupture of
some specific peptide bonds. The resulting polypeptide chains from this action
would then be apt for the action of endopeptidases that had remained inactive, and,
eventually, for the action of carboxypeptidases (CP) (Abecia-Soria et al. 2005;
Donaevsky et al. 1989). Among cacao proteins, proline content (0.72–1.97 g/100 g
of cacao) stands out, and because of its specific conformation it possesses many
restrictions on the structural aspects of peptides and proteins, and grants particular
biological properties to a large range of physiologically important biomolecules
(Kalvatchev et al. 1998; Kratzer et al. 2009). In contrast to the abundant reports
regarding Xaa-prolyl-dipeptidyl-peptidase (Xaa-Pro-DAP) [EC 3.4.14.5] in lactic
acid bacteria, there are few reports on this enzyme in plants (Besanova et al. 1987;
Stano et al. 1994, 1997; Davy et al. 2000), and given the aforementioned proline
content, this is one of the first works in regard to the Xaa-Pro-DAP activity in cacao.
536 M.L. Sánchez-Mundo et al.
2 Methods
Cacao (Theobroma cacao L.) seeds of the white almond criollo genotype, culti-
vated in the municipality of Cunduacan, state of Tabasco, Mexico, were used as
study material. For analysis and preparation, only cotyledons were used. Mucilage
was removed from seeds and these were placed in wet agrolite to germinate at 30 C
for 15 days (Sánchez-Mundo et al. 2010). Germinated seeds were collected every
24 h during all germinating processes. Water content (X), enzymatic activity
(Xaa-Pro-DAP), and water activity (aw) of germinating cocoa beans were moni-
tored for 15 days.
Seed water content (X) was measured in triplicate by drying seeds at 100 C for 2 h,
according to international standards (International Seed Testing Association 1999)
using a thermobalance (Brainweigh, model MB300, USA) until constant weight
(Nollet 1996). Results are expressed on a dry-weight basis.
Dry cacao powder (DCP) was obtained according to the method described by
Hansen et al. (1998).
Water Content, aw, and Enzyme Activity. . . 537
During the study, three stages of germination with presence of peaks of water
content and enzyme activity were observed (Fig. 1a, b): I. The first stage (1–7 days
of germination), in which values of water contents (0.58 gH2O/g ds) were detected:
In this period, it was observed that starting from the third day of imbibition of the
seed, the cotyledons emerged from substrate due to the growth of the hypocotyle. In
addition, the white color disappeared from the seed, and clear green color was seen
(Fig. 1a). These changes are attributed to the differentiation of the chloroplast in the
cotyledons, transforming them in photosynthetic organs acting as leaves,
corresponding to the epigea germination type which presents the cacao (Herrera
2006). In day 7, at the end of stage I, the lowest levels in moisture (0.28 gH2O/g ds)
and water activity (0.9695) were identified; however, this amount was sufficient to
continue the germination process. On the other hand, the enzymes also began to
increase in activity (3.55 U). This phenomenon was more evident in the next few
days.
II. In the second stage (II), during 8–10 days of germination, an increment of the
enzymatic activity was detected, specifically on day 10 of the process, with 13 U of
Xaa-Pro-DAP activity coinciding with the maximum value of water content and
water activity, 1.32 gH2O/g ds and 0.98, respectively (Fig. 1b). These results
indicate that, according to the germination conditions of this specific cacao criollo
type, at day 10 exists a proteolytic activity stimulated by moisture and water content
538 M.L. Sánchez-Mundo et al.
III
II
b
16
II 0.988
III
Moisture content/10 (gH2O/gss)
I
14 0.986
Enzymatic activity (U)
0.984
6 0.976
0.974
4
0.972
2 0.97
0 0.968
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Days of germination
Fig. 1 (a) Visible changes during germination process of cacao criollo seeds during 15 days.
(b) Graphic representation of Xaa-Pro-DAP enzymatic activity profile, moisture content, and
water activity
Water Content, aw, and Enzyme Activity. . . 539
References
Abecia-Soria p, Pezoa-Garcı́a NH, Amaya-Farfan J (2005) Soluble albumin and biological value
of protein in cocoa (Theobroma cacao L.) beans as a function of roasting time. J Food Sci 70
(4):294–298
Anonymous (1999) International Seed Testing Association (ISTA). International rules for seed
testing. Seed Sci Technol 27(Suppl): 27–32
AOAC (2005) Official methods of analysis of AOAC international, 18th ed. Method No. 978.18
Besanova M, Kovacs P, Psenak M, Barth A (1987) Dipeptidyl peptidase of poppy seedlings.
Biologia 42:779–787
Bewley DJ (1997) Seed germination and dormancy. Plant Cell 9:1055–1066
Biehl B, Wewetzer C, Passern D (1982) Vacuolar (storage) proteins of cocoa seeds and their
degradation during germination and fermentation. J Sci Food Agric 33:1291–1304
Davy A, Thomsen KK, Juliano MA, Alves LC, Svendsen I, Simpson DJ (2000) Purification and
characterization of barley dipeptidyl peptidase IV. Plant Physiol 122:425–431
Donaevsky YE, Sarbakanova ST, Belozersky MA (1989) Wheat seed carboxypeptidase and joint
action on gliadin of proteases from dry and germinating seeds. J Exp Bot 40:1323–1329
Hansen CE, del Olmo M, Burri C (1998) Enzyme activities in cocoa beans during fermentation.
J Sci Food Agric 77:273–281
540 M.L. Sánchez-Mundo et al.
Abbreviations
K Consistency index
n Flow behavior index
r² Coefficient of determination
γ Shear rate
η1 High shear rate-limiting viscosity
σ Shear stress
σ0 Yield-stress parameter
1 Introduction
F. Santoyo
Departamento de Ciencias Exactas, Tecnologı́as y Metodologı́as, CUSur, Universidad de
Guadalajara, Ciudad Guzmán, Guadalajara, Jalisco, Mexico
J. Vigano • J. Telis-Romero (*)
Departamento de Engenharia e Tecnologia de Alimentos, Universidade Estadual Paulista,
São José do Rio Preto, São Paulo, Brazil
e-mail: [email protected]
et al. 2008). It has been estimated that hot peppers are the second most consumed
vegetable by the Mexican population, after tomatoes. Approximately 75 % is
consumed as fresh product for the preparation of different dishes. Other commonly
consumed presentations of hot peppers include pickled, dried, and smoked, and in
sauces. There are several varieties of hot peppers consumed in Mexico; two of the
most popular varieties are jalape~no and serrano, in both fresh and processed forms
(Álvarez-Parrilla et al. 2011).
Fresh spice vegetables, such as hot peppers, are mainly processed to obtain
dehydrated products. The distinct aromatic odor, color, and pungency are seldom
carried over to the processed products (Giridhar et al. 1996). The chemicals
responsible for the delicate odors and attractive colors of spice vegetables are
highly heat sensitive and are lost during dehydration (Pezzutti and Crapiste
1997). Spice-vegetable pulps, classified as minimally processed food (King and
Bolin 1989), could serve the purpose by retaining the fresh spice odor in a form
acceptable to consumers (Baranowski 1985).
During production, the vegetable pulps industry deals with a variety of concen-
trations of water and temperatures, with pulps being submitted to unit operations
such as pumping, heat exchange, evaporation, and spray-drying, drum-drying, or
freeze-drying. In order to allow adequate process design, operation, and control,
knowledge of the rheological behavior of the vegetable pulps as affected by water
fraction and temperature is of fundamental importance.
Several works are available concerning the influence of particle size and tem-
perature on the rheology of green chili puree (Ahmed et al. 1999, 2000; Cepada and
Gomez 2002). On the other hand, there is little information on the rheological
parameters of pepper pulp under different conditions of water fraction. Under these
considerations, the main purpose of this work was to determine rheological prop-
erties of jalape~ no pepper pulp, as well as to develop simple correlations for
predicting these properties under different conditions of water fraction.
2 Rheological Properties
The flow properties of pulp and puree are essential for the design and evaluation of
food processing equipment. Rheological studies would provide information on how
best to control the flow properties of the product. In addition, knowledge of the
rheology of some suspensions also contributes to a better understanding of under-
lying mechanisms of momentum and heat transfer process.
In this chapter, results of investigations in rheology of jalape~no pepper pulp are
presented. The pepper pulp was prepared at different water fractions, from 0.54 to
0.76. In order to obtain the experimental data of the jalape~no pepper pulp, rheo-
logical tests were carried out using a rheometer, AR-2000EX (TA Instruments,
Delaware, USA), with concentric cylinder geometry under controlled shear rate and
temperature. Flow curves were obtained in duplicate with shear rate ramps from
1 to 442 s1, at a temperature of 30 C.
Water Fraction Effect in the Rheological Behavior of Jalape~
no Pepper Pulp 543
400
350
300
250
Shear stress (Pa)
200
150
100
50
0
0 100 200 300 400 500
Shear rate (s-1)
Fig. 1 Flow curves of jalape~no pepper pulp at 30 C and different water fraction (water fraction:
filled square ¼ 0.76; open square ¼ 0.68; filled circle ¼ 0.61; open circle ¼ 0.54)
The flow curves of jalape~no pepper pulp at 30 C and different water fraction are
shown in Fig. 1. The points of each curve represent the average of the duplicates.
The effect of the shear rate and water fraction in shear stress can be clearly
observed. The shear stress increased with increasing shear rate and decreasing
water fraction. In addition, we can see that the flow curves indicate a
non-Newtonian fluid with pseudoplastic behavior.
Flow curve modelling is interesting because simple equations are able to describe
the shear stress behavior of a fluid when different strain rates are applied. To
describe the shear stress (σ)–shear rate ( γ_ ) dependence of the jalape~no pepper
pulps, a number of empirical rheological models were tested. The two-parameter
power law, Casson and Bingham plastic, and the three-parameter Herschel-Bulkley
and Sisko models were tested. These are given by the following equations:
Power law
σ ¼ K γ_ n ð1Þ
544 F. Santoyo et al.
Casson
p ffiffiffi p ffiffiffiffiffi pffiffiffi
2
σ ¼ 2 σ 0 þK 2 γ_ ð2Þ
Bingham plastic
σ ¼ σ 0 þ K γ_ ð3Þ
Herschel-Bulkley
σ ¼ σ 0 þ K γ_ n ð4Þ
Sisko
σ ¼ η1 γ_ þ K γ_ n ð5Þ
In Eqs. (1)–(5) σ is the shear stress, γ_ is the shear rate, K is the consistency index,
n is the flow behavior index, σ 0 is the yield-stress parameter, and η1 is the high
shear rate-limiting viscosity.
This investigation provided important values for the process and facilitating the
marketing of jalape~no pepper pulp for export and internal consumption. Fitted
rheological models for the dependence of shear rate on shear stress and for the
dependence of the obtained rheological parameters on water fraction were obtained
by nonlinear estimation procedure from Statistica 8, minimizing the sum of squared
errors. The suitability of the equations was evaluated by the coefficient of determi-
nation (r2).
Table 1 shows the obtained parameters and coefficients of determination of the
fits. Based on the coefficient of determination, the better models to describe the flow
curves of jalape~ no pepper pulp, at 30 C and in the range of studied water fraction,
are power law, Herschel-Bulkley, and Sisko model (r2 > 0.995). However, among
these three models, the Herschel-Bulkley and Sisko models are more appropriate,
because they present coefficient of determination very close to unity (r2 > 0.999),
and the Herschel-Bulkley model is widely used to describe fluid foods that present
yield stress.
The presence of yield stress, which represents a finite stress required to achieve
flow, is an important characteristic of the Herschel-Bulkley plastic material. Below
the yield stress, a material exhibits solid-like characteristics; that is, it stores energy
and does not level out under the influence of gravity to form a flat surface. This
characteristic is very important in process design and quality assessment for
materials (Steffe 1996) such as pepper pulp.
Results of some research have also shown that pepper purées presented
pseudoplastic behavior. However, the power law model was fitted to experimental
data only, which presented good coefficient of determination results. Cepada and
Gomez (2002) fitted the power law model to experimental data of purée of three
Water Fraction Effect in the Rheological Behavior of Jalape~
no Pepper Pulp 545
Table 1 Variation of the five model parameters with the water fraction for jalape~
no pepper pulp
Water Water
fraction Model Parametera r2 fraction Model Parametera r2
0.76 Power K 26.557 0.997 0.61 Power K 32.717 0.997
law n 0.356 law n 0.379
Casson σ0 59.025 0.902 Casson σ0 76.958 0.910
K 0.697 K 0.840
Bingham σ0 59.025 0.902 Bingham σ0 76.958 0.910
plastic K 0.486 plastic K 0.705
Herschel- σ0 15.566 0.999 Herschel- σ0 19.243 0.999
Bulkley K 16.906 Bulkley K 21.466
n 0.422 n 0.442
Sisko η1 0.110 0.999 Sisko η1 0.162 0.999
K 29.983 K 37.289
n 0.304 n 0.325
0.68 Power K 29.625 0.997 0.54 Power K 33.233 0.995
law n 0.375 law n 0.387
Casson σ0 68.975 0.907 Casson σ0 79.673 0.920
K 0.788 K 0.870
Bingham σ0 76.958 0.910 Bingham σ0 79.673 0.920
plastic K 0.705 plastic K 0.757
Herschel- σ0 16.437 0.999 Herschel- σ0 23.860 0.999
Bulkley K 19.874 Bulkley K 19.756
n 0.435 n 0.466
Sisko η1 0.134 0.999 Sisko η1 0.225 0.999
K 33.439 K 39.695
n 0.324 n 0.313
In this work, K is in Pa sn, η1 is in Pa s, σ 0 is in Pa, and n is dimensionless
a
pepper varieties. The behavior index (n) at 30 C was very close to this research,
from 0.30 to 0.37, while the consistency index (K ) was very diverged (4.42–
13.9 Pa sn). Ahmed et al. (1999) also fitted the power law model to experimental
data of pepper purée. Their n results at temperature of approximately 30 C were
lower than those of this work (0.228–0.300), while K values were similar (36.482–
29.703 Pa sn). These differences are possibly due to differences in strain rate range
used (0–100 s1), differences of water fraction, or to particle size of the suspension.
The water fraction influence in the values of Herschel-Bulkley parameters was
evaluated. Figures 2, 3, and 4 show the effect of the water fraction in the yield
stress, consistency, and behavior index, respectively. The increase of yield stress
values exponentially with decreasing water fraction can clearly be seen. The same
effect was observed in the behavior index, while the consistency index increased
linearly until water fraction of 0.61 and then decreased with decreasing water
fraction.
546 F. Santoyo et al.
24
22
Yield-stress (Pa)
20
18
16
14
0.50 0.55 0.60 0.65 0.70 0.75 0.80
Water fraction
22
21
Consistency index (Pa.s")
20
19
18
17
3 Conclusion
The shear stress increased with increasing shear rate and decreasing water fraction
of jalape~
no pepper pulp. Jalape~no pepper pulp behaved as a pseudoplastic fluid and
the Herschel-Bulkley and Sisko models described well the shear stress–shear rate
Water Fraction Effect in the Rheological Behavior of Jalape~
no Pepper Pulp 547
0.47
0.46
0.45
Behavior index
0.44
0.43
0.42
Acknowledgements The authors would like to acknowledge and thank FAPESP (Process Num-
ber 2009/11675-3) for their financial support.
References
Steffe JF (1996) Rheological Methods in Food Process Engineering, 2nd edn. Freeman, East
Lansing, p 418
Topuz A, Ozdemir F (2007) Assessment of carotenoids, capsaicinoids and ascorbic acid compo-
sition of some select peppers cultivars (Capsicum annuum L.) grown in Turkey. J Food
Compos Anal 20:596–602
Study of Water Quality Through
Hydro-Chemical Signature in Leon,
Guanajuato, Mexico
Abbreviations
1 Introduction
The aquifer system of Valle de Leon (VL) is a free aquifer that consists mainly of
sedimentary Tertiary rocks of lacustrine origin (Fig. 1). This valley presents
different hydraulic conductivity stratifications that store water with various chem-
ical and isotopic characteristics, so this water represents its different origins.
According to the most recent census (Instituto Nacional de Estadı́stica y Geografı́a,
INEGI), VL has a population of 1,436,480 inhabitants, living in 11 population
S.A. Cortés
Instituto de Geofı́sica, Universidad Nacional Autonoma de México (UNAM), Circuito de la
investigacion cientı́fica s/n, Ciudad Universitaria, Delegacion Coyoacán, C.P. 04510 México,
D.F., Mexico
G.A. Lozano (*)
Centro de Investigaci on en Ciencia Aplicada y Tecnologı́a Avanzada (CICATA, Unidad
Querétaro), Instituto Politécnico Nacional (IPN), Cerro Blanco No. 141, Col. Colinas del
Cimatario, C.P. 76090 Santiago de Querétaro, Querétaro, Mexico
e-mail: [email protected]
J. Pérez
Centro de Geociencias, Universidad Nacional Autonoma de México (UNAM), Campus
Juriquilla, Querétaro, Mexico
centers; it is the region’s most important industrial corridor and is included in one of
Mexico’s most productive agricultural zones. VL covers approximately 687 km2
surrounded by higher altitude topographic zones. The water supply of the popula-
tion is mainly groundwater. This comes from deep wells, which are located mostly
in irrigation zones but also in urban and rural zones of VL.
The Guanajuato State Water Commission (Comision Estatal de Agua de
Guanajuato, CEAG) reports that there are 1,154 active wells in VL, averaging
300–400 m depths. From this number of wells, 797 goes to irrigate 15,000 ha,
245 supply drinking water, 100 goes to cattle ranches, and 12 of them supply
industry. The VL drainage system covers 2,672 km. Drainage comprising urban
wastewater and pluvial water is 2,540 km long. Leon city (Fig. 1) itself generates
2,300 l/s of wastewater, generated by housing gray water and sewage released by
urban industries.
2 Data Collected
Major and minor ions, heavy metals, and physicochemical parameters were col-
lected from 1996 to 2007, resulting in 1,381 data sets (SAPAL 2007). From these
sets, all incomplete sets were eliminated. The elimination criterion was a lack
of information related with major ions, which are the ones used for ionic balance.
Study of Water Quality Through Hydro-Chemical Signature. . . 551
Only data sets with a load balancing less or equal to 16 % were taken into account.
Based on this information, the behavior of the chloride, nitrate, and fluoride ions
and the metals such as lead, chrome, cadmium, and arsenic was determined
(Ramos-Leal et al. 2005). Analysis of these data showed concentration changes in
time though their hydro-chemical signature showed that historical values tend to
converge to those obtained for the 2007 sampling (Ramos-Leal et al. 2007).
3 Results
From all the studied wells, it was found that 92 % of them have water where
bicarbonate ion is present. On the rest 8 %, chloride ion was predominant. The
highest values of this ion were found in wells situated on zones irrigated with
wastewater. It is noted that 10 % of this group are above the Mexican Drinking
Water Norm (MDWN) (NOM-127-SSA1-1994). The anomalous situation appears
in a group of wells located in the south of the study area and its origin is associated
to the presence of faults and/or components of deep water. Nitrate high values are
mainly in wells located on urban zones; quality is related to leaks in the sewer
system. Results for heavy metals, in general, do not present inconsistencies, and
most data are within the MDWN. Figure 2 shows piper diagram for the 2007 well
sampling evenly distributed within VL. This diagram allows identifying four
probable facies (water families). Facie I has the characteristic of having ion
Na > 80 % related to its relative abundance, and its principal anion is HCO3 > 80 %.
Facie II is characterized by the relative abundance of Na+ > Ca2+ cations. This facie
can be described as Na+ > Ca2+ HCO3 > SO24. Anion HCO3 is present in all
wells related with facie III. For this facie, cations show high variability though in
15 wells predominant relation is Ca > Na. Facie IV presents HCO3 > Cl and
predominant cations are Ca > Mg. Figure 3 shows the relation between field-
measured water average temperature against milliequivalents per liter concentra-
tions for SO42, Na+, Li, F, and boron ions for the previously described facies.
Figure 4 describes behavior of N-NO32 ion for urban zone-located wells.
Allowable value for this ion is 10 mg/l. From this figure, a gradual increment of
this ion’s value can be seen. This increment causes the allowable value for drinking
water to be surpassed and, in extreme cases, the well to be closed. Within the Leon
city might be illegal wells pumping out contaminated water without any control.
So, this issue lives open the question on how exposed is the population to this kind
of water.
552 S.A. Cortés et al.
Fig. 3 Water temperature versus milliequivalent per liter relation for facie identified at the
studied zone
554 S.A. Cortés et al.
4 Conclusions
Results up to date show a significant variation, so time sampling and time of well
operation seem to be of prime importance. Related to water quality, generally
speaking, it can be said that its quality is acceptable and that drinking water is
within the current standard. It has to be taken into account that sampling is carried
out at the wells’ output before chlorination takes place. So, sampled water is aquifer
representative and different from that coming out from a house tap. It can be said
that although there are pollution problems, both natural and anthropogenic, they can
be kept controlled with a proper aquifer management.
References
Abbreviations
CI Crystallinity index
DSC Differential scanning calorimetry
FAAT Fibers submitted to acid-alkaline hydrolysis at elevated temperature
FAATP Fibers submitted to acid-alkaline hydrolysis with pressure and at
elevated temperature
FAWT Agave fibers without any treatment
FAWT Fibers without treatment
FTIR Spectroscopy Fourier transform infrared
TGA Thermogravimetric analysis
XRD X-ray diffraction
1 Introduction
Agave tequilana crop waste provides lignocellulosic material that can be exploited
through the surface treatment of the fibers, such as by soaking in acid-alkali media,
which allows for the separation of water-soluble components such as pectins,
sugars, and fructooligosaccharides from insoluble components such as hemicellu-
lose and lignin fibers. This treatment also improves the properties of the fibers in
terms of their thermal stability and tensile strength, which is necessary for devel-
oping composites because it allows for the removal of impurities such as waxes,
pectin, and minerals, increasing the fibers’ surface area (Ben Sghaier et al. 2012;
Rosli et al. 2013). The objective of this research was to characterize lignocellulose
fibers from A. tequilana Weber by calorimetric and thermogravimetric techniques.
2.1 Materials
Agave tequilana Weber leaves aged approximately 8–10 years were collected from
Morelos State, Mexico. The leaves were washed, weighed, had their cuticle
excised, pressed, dried at 70 C for 12 h, freeze-dried for 12 h, and finally milled
into a powder with a mean particle size of 150 μm.
The leaf powder (FAWT) was subjected to two different treatments in an attempt to
separate water-soluble components such as pectins, sugars, and fructooligosac-
charides from fibrous components such as lignocellulosic materials. In treatment
1, a modified acid-alkaline hydrolysis process was used (Saucedo et al. 2010); the
fibers obtained using this treatment was denoted as FAAT. Treatment 2 was
conducted by applying the technique reported by Hernández-Salas et al. (2009).
This treatment involved the application of elevated temperature and pressure (121
C/1.1 kg/cm2) over a period of 4 h; the fibers obtained using this treatment were
denoted as FAATP.
Agave tequilana fibers were submitted to chemical analysis to determine their ash
content (AOAC—31 013, 1995) and content of total reducing sugars (Ting 1956).
Calorimetric and Thermogravimetric Analysis of Agave tequilana Weber Fibers 559
Samples of agave fibers not submitted to any treatment (FAWT) and modified fibers
(FAAT and FAATP) were analyzed by a thermogravimetry device (Equinox,
Bruker, Finland) coupled to a thermoanalyzer (STA 409 PC Luxx, Netzsch, Ger-
many). Thermograms were obtained at a heating rate of 10 C/min and over a
temperature range of 30–80 C.
The FAWT, FAAT, and FAATP samples were analyzed using a differential
scanning calorimetry (DSC) 2010 calorimeter (TA-Instruments Inc., USA) over a
temperature range of 30–620 C in increments of 10 C/min. Each scan was carried
out separately under a nitrogen atmosphere at a flow rate of 20 cm3/min.
The FAWT, FAAT, and FAATP samples were analyzed using a diffractometer
(D8 advance Vantec Bucker, Germany) with CuKα radiation source (1.5418 Å), a
radioactive Ni filter, and a scintillation counter as a detector at 35 kV and 30 mA
over a range of 2θ angles of 3–37 via the technique reported by Singha and Rana
(2010). The crystallinity index (CI) was determined using the diffraction signals
obtained at 2θ angles near 15 and 22 .
The chemical compositions of the Agave tequilana leave powder (FAWT) and the
fibers obtained by the modified acid-alkaline hydrolysis process without pressure
(FAAT) and with pressure (FAATP) are shown in Table 1. These treatments
induced the chemical breakdown of compounds due to reaction with acid and
alkaline substances, allowing for the breaking of lignin and hemicellulose bonds
560
(Rosli et al. 2013). Pressure also facilitated the hydrolysis of cellulose bonds,
reducing the content of this component in FAATP.
Thermogravimetric analysis (TGA) revealed that the two extraction treatments
affected the fibers in different ways; in fact, three main stages of weight loss of the
fiber materials could be recognized at different temperatures. The FAWT showed a
gradual release or loss of components as the temperature increased. During the first
stage (between 25 and 104 C), FAWT lost free water weakly linked to the structure
and water-soluble solids such as simple sugars. Subsequently, in the second step
(104–241 C), the mass loss was attributed to the degradation of fructooligosac-
charides, pectins, sugar monomers (maltose), and disaccharides such as sorbitol. In
the third step (241–302.08 C), the FAWT reached 60 % total weight loss. At this
stage, the residual thermal degradation of hemicellulose occurred. With an increase
in temperature, chemical changes in the hemicellulose components occurred due to
cell wall rupture. This finding is consistent with previous studies indicating that
hemicellulose degrades between 180 and 350 C. Cellulose is more thermally stable
due to its crystalline nature, which is composed of a network of hydrogen bonds that
form microfibrils; it has been reported that the thermal degradation of cellulose is
initiated at 350 C (Yang et al. 2007). FAATP showed mass loss up to a temperature
of 241 C. These fibers were composed mainly of cellulose and lignin in lower
proportion. The FAAT fibers exhibited a high loss of mass during the first stage,
mainly due to the removal of large amounts of water, simple sugars, and water-
soluble solids. The behavior observed in the second stage was similar to that
observed for FAATP and FAWT. In the third stage, FAATP showed a higher
percentage of mass loss (39 %) than that observed for the other fiber materials.
The high thermal resistance of FAATP and FAAT is due to the materials’ high
lignin content; thus, these fibers were observed to thermally degrade between
200 and 700 C at a mass loss rate much lower than the rates observed for cellulose
and hemicellulose. This finding agrees with the lignin and cellulose contents
observed for FAATP, FWAT, and FAAT, namely 19.85 and 31.53, 14.58 and
68.37, and 13.52 and 80.35, respectively (Fig. 1).
Differential scanning calorimetric analysis revealed two endotherms for FAWT
(Fig. 2). The first one began at 94.91 C with a plateau of 120.87 C and
ΔH ¼ 140.1 J/g. The second endotherm occurred at 206.79 C with a plateau of
220.33 C and ΔH ¼ 23.09 J/g. In addition, the FAAT cellulosic material showed
two endotherms, one beginning at 105.79 C with a plateau at 127.42 C and
ΔH ¼ 218.0 J/g and the other beginning at 205.53 C with a plateau at 225.05
and ΔH ¼ 17.78 J/g. The first endotherm was likely due to the presence of
D-fructose, whereas the second was likely due to hydrolyzed chains of fructans,
based on the fact that agave fructans thermally decompose at temperatures near
200 C (Espinosa-Andrews and Urias-Silvas 2012). Moreover, agave fructans show
a decrease in their thermal decomposition rate at approximately 230 C, which
could be associated with the decomposition of branched chains of agave fructans.
Chui et al. (2002) reported that the decomposition rate of cross-linked inulin is
lower than that of native inulin.
562 C. De Dios-Naranjo et al.
Fig. 1 Thermogravimetric analysis of Agave tequilana Weber: sample of leaf powder (FAWT),
fibers submitted to modified acid-alkaline hydrolysis without pressure (FAAT) and fibers submit-
ted to modified acid-alkaline hydrolysis with pressure (FAATP)
Fig. 2 DSC results for different lignocellulosic materials of Agave tequilana Weber
Calorimetric and Thermogravimetric Analysis of Agave tequilana Weber Fibers 563
FAATP showed two endotherms. The first began at 77.47 C with a plateau of
113.68 C and ΔH ¼ 139.1 J/g. This behavior is likely due to sugar alcohols such as
D-sorbitol due to the fructooligosaccharide decomposition that generates this mono-
saccharide (Roos 1995), whose onset occurs at 86 C with a plateau at approxi-
mately 110–112 C and ΔH ¼ 150 J/g. The second endotherm occurred at
182.37 C with a peak at 217.48 C and ΔH ¼ 74.80 J/g. This endotherm could
be due to the decomposition of α,α-trehalose, whose onset occurs a 173 C with a
plateau at 214–216 C and ΔH ¼ 118 J/g (Roos 1995). The ΔH reported may vary
because sugars do not exhibit exact melting temperatures; thus, it is important to
note that the fusion temperatures and ΔH values may slightly differ between sugar
monomers (Roos 1995). Waleckx et al. (2008) mention that A. tequilana Weber var.
Azul and other Agave species contain a complex mixture of highly branched
fructans and neo-fructans (referred to as agavins), with each type of linkage
(β-(2–1) and β-(2–6)) between fructose moieties.
Changes in the structural arrangement of the lignocellulosic material determined
by X-ray diffraction (XRD) are shown in Fig. 3. The chemical solutions of FAATP
and FAAT showed changes in the crystallinity indexes (CIs). FAATP lignocellu-
lose showed a CI of 41.49 %, 20.4 % greater than that of fibers without treatment.
FAAT showed a similar CI, differing by 2.51 % compared to the CI of FAATP. It is
possible that the diluted acid destroyed the amorphous regions of the fibers, whereas
the alkali component helped increase the stiffness as well as remove impurities
(Fengel and Wegener 1984; Mwaikambo and Ansell 2006).
Fig. 3 XRD results for different lignocellulosic materials of Agave tequilana Weber
564 C. De Dios-Naranjo et al.
4 Conclusions
Applying extraction processes that involve the use of acid-alkali solutions with or
without pressure can enhance the removal of components within Agave fibers as
well as the components of more open microfibrils within fiber complexes. At the
beginning of the extraction process, important water-soluble components are
removed; these include reducing sugars, chlorophylls, and pectins that are widely
used in the food industry.
TGA coupled with Fourier transform infrared (FTIR) spectroscopy, DSC, and
XRD is a useful tool for analyzing any changes that may occur during the dissolu-
tion of lignocellulosic materials contained in Agave fibers.
Acknowledgments The authors acknowledge the financial support of research projects SIP
20140253 and 20140100. The authors are fellows of EDI-IPN, COFAA-IPN, and SNI.
Cuauhtémoc de Dios Naranjo was awarded a CONACyT grant.
References
AOAC (1995) Official method of analysis, 14th edn. Association of Official Analytical Chemists,
Washington, pp 23–26
Ben Sghaier AEO, Chaabouni Y, Msahli S, Sakli F (2012) Morphological and crystalline charac-
terization of NaOH and NaOCl treated Agave americana L. fiber. Ind Crops Prod 36
(1):257–266
Chui HC, Hsu YH, Lin PJ (2002) Synthesis of pH-sensitive inulin hydrogels and characterization
of their swelling properties. J Biomed Mater Res 61(1):146–152
Espinosa-Andrews H, Urias-Silvas JE (2012) Thermal properties of agave fructans (Agave
tequilana Weber var. Azul). J Carbohydr Polym 87:2671–2676
Fengel D, Wegener G (1984) Wood chemistry, ultrastructure, reactions. In: Chemical composi-
tion. Walter de Gruyter Press, Berlin. pp. 256–265
Hernández-Salas JM, Villa-Ramirez MS, Veloz-Rendon JS, Rivera-Hernández KN, González-
César CRA, Plascensia EMA, Trejo ESR (2009) Comparative hydrolysis and fermentation of
sugarcane and agave bagasse. Bioresour Technol 100:1238–1245
Moller J (2009) Gravimetric determination of acid detergent fiber and lignin in feed:
interlaboratory study. AOAC Int J 92(1):74–90
Mwaikambo LY, Ansell MP (2006) Mechanical properties of alkali treated plant fibres and their
potential as reinforcement materials. I Hemp fibres. J Mater Sci 41:2483–2496
Roos YH (1995) Phase transitions in foods. Food components and Polymers. Academic, USA, pp
113–115
Rosli NA, Ahmad I, Abdulllah I (2013) Isolation and characterization of cellulose nanocrystals
from Agave angustifolia fibre. Bioresour Technol 8(2):1893–1908
Saucedo LJ, Castro MAJ, Rico JL (2010) Optimization of acid Hydrolysis if Bagasse from Agave
tequilana Weber. Rev Mex Ing Quı́m 91(1):91–97
Singha AS, Rana RK (2010) Graft copolimerization of methyl methacrylate (MMA) onto Agave
Americana fibers and evaluation of their physicochemical properties. Int J Polym Anal Charac
15:27–42
Ting SV (1956) Fruit juice assay, rapid colorimetric methods for simultaneous determination of
total reducing sugars and fructose in citrus juices. J Agric Food Chem 4(3):263–266
Calorimetric and Thermogravimetric Analysis of Agave tequilana Weber Fibers 565
Abbreviations
1 Introduction
Allspice (Pimenta dioica L. Merrill) is a spice native to Southern Mexico that can
be used in different ways, from the pharmaceutical industry to the food industry,
and it is well known for its flavoring properties from antiquity to today (Macı́a
1998). Allspice contains a number of antioxidants, which has caught the attention of
researchers in this field.
The drying operation is a process which involves mass transfer between a gas
and a solid product, where the moisture content of the solid is transferred by
evaporation from the gas phase (Perry et al. 1997). The main purpose of drying is
to extend the half-life of food and to reduce water activity in order to inhibit
microbial growth and enzyme activity (Hui 1992; Rahman and Perera 1999).
Food dehydration presents problems related to the food’s structural characteris-
tics, which affects rehydration, retention of aroma and sensorial attributes, as well
as mass and heat transport phenomena during the process (Aguilera 2003).
In a previous work Flores et al. (2009) it was found that during the drying of the
allspice at 60, 70, and 80 C, the product presents shrinking, but no deformation.
This fact indicates that the operation is isotropic; this phenomenon is not common
in foods. In addition, they evaluated the biochemical quality of this operation, and it
was found that at 70 C there was major retention of ascorbic acid, while at 80 C
lower retention of ascorbic acid was shown. Thus, the aim of this study is to
evaluate the changes in the structure during drying with the help of scanning
electron microscopy (SEM), as well as the determination of enzyme activity of
peroxidase, polyphenol oxidase, catalase, glucose oxidase, and total phenols.
The raw material was obtained from Cuetzalan, in the state of Puebla, having an
initial moisture content of 0.627–0.617 db.
The equipment used was a fluidized bed dryer. Drying curves were carried out
based on the measure of the broken record of loss weight by using the relation
L/D ¼ 0.5 (145 g of allspice). Air temperature was 70 and 80 C, reaching a final
moisture content of 0.11 db. Surface temperature was also evaluated during
dehydration.
The effective diffusion coefficients (Deff) were determined using Fick’s second
law for a spherical geometry (Eq. 1). For the geometry calculation, the expression
of logarithmic mean radius was used (Eq. 2):
X X*
¼ eDL tðπ=req Þ
2
ð1Þ
Xt1 X *
r2 r1
r¼ ð2Þ
lnrr21
and where r ¼ logarithmic mean radius; r1 ¼ initial radius; and r2 ¼ final radius.
For the analysis of texture, relief, size, and material composition, the scanning
electron microscope (SEM) of the Microscopy Center at Escuela Nacional de
Ciencias Biol
ogicas (IPN) was used. In order to take a sample of the allspice
which contained cortex and seed, the allspice was cut in half and then a layer
from the center was cut. The layer was fixed with 2 % glutaraldehyde and 1 %
osmium tetroxide and sequentially dehydrated by using different ethanol solutions
(10–100 %). Then was critical-point dried, cut, and assembled on slides which were
covered with gold-palladium. In the case of green allspice, the samples were
assembled on aluminum slides and coated with gold-palladium as well.
Temperature of the allspice increased in the first 5 min, and reached the temperature
of the dry bulb at 70 C within 45 min of aeration, which indicates that drying
occurs exclusively in the decreasing rate period. This behavior was also found at
80 C.
Figure 1 shows the conventional drying curves, which can be observed with no
significant difference because the standard deviation of the curves overlaps; this is
570 A.E. Carpinteyro-Dı́az et al.
a 1
70ºC 80ºC
0.9
0.8
X* (dimensionless)
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 1000 2000 3000 4000 5000 6000 7000 8000
Time (s)
Time (s)
b 0
0 1000 2000 3000 4000 5000 6000 7000 8000
-0.5
-1 70ºC 80ºC
LnX* (dimensionless)
-1.5
-2
-2.5 R2 = 0.95283
-3
-3.5 R2 = 0.95981
-4
-4.5
-5
Fig. 1 Conventional drying curves of allspice by fluidized bed drying (a) and curve of lineariza-
tion carried out by the diffusion coefficient with the second Fick’s law equation (b)
due to the way the allspice reaches the temperature of the bulb dry air; therefore, at
70 C after 40 min, being a lower temperature, the drying extends 25 min more
(Fig. 1a).
Figure 1b shows the corresponding line to the evaluation of the diffusion
coefficient with the second Fick’s law equation. The model used is consistent
with the moisture loss of the allspice at 80 C from 0 to 4,800 s, and at 70 C the
highest rate of water migration is at 6,300 s. In the linearization, it was found that, in
these periods, the correlation coefficient is the highest; after these periods the slope
decreases, extending the drying. This decreasing period is no longer a Fickian
model. Table 1 shows values of logarithmic mean radius with no significant
difference, as well as values of the effective diffusion. This last parameter is
Study of Allspice Fluidized Bed Drying (Pimenta dioica L. Merrill). . . 571
Figure 2 shows the micrographs of green allspice, where the seeds and the cortex of
the product can be observed. In Fig. 2a, some holes can be seen near the outer cell
wall of the cortex. These holes are cavities of volatile oil; the epidermis can be seen
as well (Winton and Barber-Winton 1939). The cell walls of the seed are reinforced
with starch, which is found in clusters, and are composed of tiny spheres of this
macromolecule (Fig. 2b). Also found in the cortex was the presence of starch
having a spherical shape (Fig. 2c).
By comparing the allspice structure with the dehydrated products (Fig. 3),
structure loss by cell compaction was found, which leads to fragility in the cortex
that may cause fractures. When the cavities were compacted, the structure released
volatile oils that give the characteristic smell of allspice (Fig. 3a). During drying,
the structure of the starch begins to present a more elongated conformation that is
probably caused by thermodegradation (Karlsson and Eliasson 2003) (Fig. 3b),
while the appearance of calcium oxalate crystals in the vacuoles is the result of
metabolic reactions of the product after dehydration (Nakata and McConn 2003)
(Fig. 3c). When comparing the samples dried at 80 C with the ones at 70 C
(Fig. 4), the latter show less cell compaction (Fig.4a) because volatile oil cavities
can be better seen. The presence of elongated starch (Fig. 4b) and the appearance of
oxalate crystals can also be seen (Fig. 4c).
The enzymes evaluated in this study are from the group of oxidoreductases.
Catalase, peroxidase, and polyphenol oxidase regulate the formation of free radicals
like superoxide, hydrogen peroxide, and lipid peroxides. Figure 5 shows the results
of enzyme activity in drying temperatures. It was found that peroxide and polyphe-
nol oxidase are activated mostly at 80 C because of the structural damage of the
membrane and the release of enzymes present in the same, while catalase showed
572 A.E. Carpinteyro-Dı́az et al.
Fig. 2 Micrographs of green allspice obtained by SEM. Micrograph at 35 (a), 500 (b), and
220 (c) of microscopic magnification
UPFO
200
2000
150 1000
1500
100 1000
500
50 500
0 0 0
G.A. C. 70°C C. 80°C G.A. C. 70°C C. 80°C G.A. C. 70°C C. 80°C
Fig. 5 Peroxidase, polyphenol oxidase, and catalase in allspice. G.A. ¼ green allspice,
C. 70 C ¼ conventional at 70 C, C. 80 C ¼ conventional at 80 C
Study of Allspice Fluidized Bed Drying (Pimenta dioica L. Merrill). . . 573
300
250
200
mMol H2O2 / mL
150
100
50
0
G.A. C. 70°C C. 80°C
Fig. 6 Glucose oxidase enzyme in the allspice. G.A. ¼ green allspice, C. 70 C ¼ conventional at
70 C, C. 80 C ¼ conventional at 80 C
References
Aguilera JM (2003) Drying and dried products under the microscope. Food Sci Technol Int
9:137–143
Flores NL, Cruz y Victoria MT, Anaya I (2009) Dehydration in fluid bed with tempering cycles,
using residual enzymatic activity as a quality parameter. Paper presented at 3rd international
congress on food science and food biotechnology in developing countries, Querétaro, México,
2009
Gutiérrez DM, Ortiz CA, Mendoza A (2008) Medicion de fenoles y actividad antioxidante en
malezas usadas para alimentacion animal. Paper presented at simposio de metrologı́a,
Querétaro, México
Hui YH (1992) Encyclopedia of food science and technology. Wiley-Interscience, New York
Karlsson ME, Eliasson AC (2003) Gelatinization and retrogradation of potato starch in situ as
assessed by differential scanning calorimetry. Lebens Wiss Technol 36:735–741
Kimberly WW, Lee CY (1981) Characterization of polyphenoloxidase. J Food Sci 46:506–514
Macı́a MJ (1998) La pimienta de Jamaica (Pimenta dioica L., Merrill, Myrtaceae) en la Sierra
Norte de Puebla (México). An Jard Bot Madr 56:337–349
574 A.E. Carpinteyro-Dı́az et al.
Mahely AC, Chance B (1954) The assay of catalases and peroxidases. In: Glick D (ed) Methods of
biochemical analysis, vol I. Churchill, London, pp 408–424
Nakata PA, McConn MM (2003) Calcium oxalate crystal formation is not essential for growth of
Medicago truncatula. Plant Physiol Biochem 41:325–329
Perry RH, Green DW, O’Mahoney JH (1997) Perry’s chemical engineers’ handbook, 7th edn.
McGraw-Hill, New York
Rahman MS, Perera CO (1999) Drying and food preservation. In: Rahman MS (ed) Handbook of
food preservation. Marcel Dekker, Inc., New York, pp 173–216
Voget CE, Maza LA, Balatti A (1987) Determinacion colorimétrica de glucosa oxidasa en
preparado enzimático y extractos celulares. Rev Latinoam Microbiol 30:47–52
Whitaker JR (1994) Principles of enzymology for the food science, 2nd edn. Marcel Dekker Inc.,
New York, p 648
Winton AL, Barber-Winton K (1939) The structure and composition of foods. Sugar, sirup, honey,
tea, coffee, cocoa, spices, extracts, yeast, baking powder, vol IV. Wiley, New York, pp
408–413
Refined Hemisphaericin Stabilization
by Microencapsulation with Arabic
Gum and Spray Drying
Abbreviations
AG Arabic gum
CCD Central composite design
PC Protein concentration
RH Refined hemisphaericin
RHME Refined hemisphaericin microencapsulated
RS Response surface
SEM Scanning electron microscopy
T Spray drying feed temperature
TU Tyrosine unit
YIELD Microencapsulation yield
1 Introduction
RS design and analysis were carried out using Design Expert 5 software. A central
composite design was used to investigate the variation of microencapsulation with
respect to the factors arabic gum concentration and drying inlet temperature. The
response variables were protein concentration, proteolytic activity, and microen-
capsulation efficiency, resulting in 13 experimental settings (runs) with five repe-
titions on the central point. The experimental conditions are listed in Table 1.
The protein content in the refined hemisphaericin preparations and the microcap-
sules were determined by the Bradford method (1976) and using bovine serum
albumin as a standard.
Hemisphaericin activity was measured by the method of Ortega and del Castillo
(1966), based on the hydrolysis of casein. A standard curve was generated with
tyrosine. One unit of protease activity, tyrosine unit (TU), was defined as the
amount of enzyme required to liberate 1 μg of tyrosine per minute per mg protein,
under the used conditions.
where X1 is the temperature ( C), and X2 is the concentration of arabic gum [g].
Figure 1 shows response surface plots of refined hemisphaericin
microencapsulated (RHME) variations as a function of AG concentration and
feed temperature (T ): the effect on proteolytic activity assayed with cysteine
(Fig. 1I) and the effect on proteolytic activity without cysteine (Fig. 1II). Also
described in this figure are the observed changes in the microcapsules as a conse-
quence of storage time at 5 C: (A) t ¼ 0, (B) 1 month, and (C) 2 months.
Higher proteolytic activity (816 TU), assayed in the presence of cysteine, was
found for microcapsules obtained with AG concentration between 7 and 10 % dried
at 120–130 C. On the other hand, the response surface plot of proteolytic activity,
without cysteine, exhibited the best value (117 TU) under the central point condi-
tions: 120 C and 15 % AG (Fig. 1I).
Surface response results showed that better conditions were located at 7–15 %
AG and 120–130 C of T, where greater activity levels (with and without cysteine)
were achieved. Statistical analysis of the results generated the following equations:
After 1 month of storage, better proteolytic activity for RHME, without the
reducing agent cysteine, was found in samples corresponding to experimental
runs number 7 and 9 and in the central point, namely 10 % AG-130 C, 15 %
AG-135 C, and 15 % AG-120 C, respectively. Meanwhile, the better activity
(almost 100 TU) of RH microcapsules stored for 2 months corresponded to capsules
produced with 10 % GA-130 C, the conditions in the experimental run
7 (Fig. 1IIC).
580 L.A. Reyes et al.
a 800
120
750
700 100
650 80
PA (UT)
PA (UT)
600 60
550
40
500
20
120.0
10.0
12.0 118.0
120.0 10.0
14.0 118.0 12.0 116.0
16.0 116.0 14.0 114.0
AG (%) 18.0 114.0 16.0 112.0 T (°C)
112.0 18.0
20.0 110.0 T (°C) AG (%) 20.0 110.0
b
160
850
140
800 120
750 100
80
PA (UT)
PA (UT)
700
60
650 40
20
600
120.0
118.0 120.0
10.0 118.0
10.0 116.0 12.0
12.0 14.0 116.0
14.0 114.0 T (°C) 114.0
16.0 16.0
18.0
112.0 18.0 112.0 T (°C)
AG (%) 20.0 110.0
20.0 110.0
AG (%)
c
90
780
760 80
740 70
720 60
PA (UT)
PA (UT)
700
50
680
40
660
640 30
120.0
120.0
118.0 118.0
10.0 116.0 10.0 116.0
12.0 12.0 114.0
14.0 114.0 14.0
16.0 T (°C) 16.0
18.0
112.0 T (°C)
112.0 20.0 110.0
18.0 AG (%)
AG (%) 20.0 110.0
Fig. 1 Response surface analysis showing the effect on proteolytic activity (PA) assayed with
cysteine (I), and on proteolytic activity without cysteine (II), of arabic gum (AG) concentration
and microencapsulation feed temperature (T). (a) t ¼ 0; (b) 1-month storage at 5 C; (c) 2-month
storage at 5 C
The response surface graph for proteolytic activity of RHME with cysteine after
1 month of storage (Fig. 1IB) shows an increase for the experimental runs 4, 5, 7, 9,
and 13, especially for the 7.93 % AG and 120 C conditions of run 13. After the
second month of storage, better activity (around 900 TU) was found for those
conditions assayed in runs 7 and 13, that is, 10 % AG-130 C and 7.9 % AG-120 C,
Refined Hemisphaericin Stabilization by Microencapsulation with Arabic Gum. . . 581
The surface morphology and microcapsule size were observed by SEM of RHME
preparations. The results are shown in Fig. 4. Particles were dispersed within sizes
from 5 to 50 μm; greater diameters were observed for those dried under 135 C and
22 % AG. The microcapsules were spherical and irregularly shaped, and small-
sized particles were susceptible to agglomeration.
Refined Hemisphaericin Stabilization by Microencapsulation with Arabic Gum. . . 583
4 Conclusions
Better stability at the assayed time of storage was found for proteolytic preparations
obtained with 7.9 % AG at 120 C (experimental run 13), which improved the
initial activity by 14.8 %; similarly, capsules prepared with 10 % AG at 130 C
(run 7) showed a 14.2 % greater activity than the initial sample (zero time). The
sample with the most stability was that obtained at the central point, with 15 % AG
and 120 C; nevertheless, this sample presented lower activity (711 TU) than those
obtained under the assayed conditions in runs 7 and 13 (932 and 890 TU).
References
Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of
protein utilizing the principle of dye binding. Anal Biochem 72:248–254
Briones-Martı́nez R, Cortés-Vázquez MI (2000) Avances en los estudios cientı́ficos y tecnologicos
de la hemisfericina: Una nueva proteinasa de interés industrial. In: Tapia M
(ed) Contribuciones a la Investigacion Regional en el Estado de Morelos. CRIM-UNAM,
México, pp 425–467
Briones-Martı́nez R, Oliver-Salvador MC, Cruz MT, Cortés MI (1994) Preparaciones enzimáticas
de interés industrial con proteinasas de plantas mexicanas. Inf Tecnol 5–1:29–38
Cortés-Vázquez MI, Mu~noz-Sánchez JL, Briones-Martı́nez R (2008) Substrate specificity of a
cationic peptidase from Bromelia hemisphaerica L. Nat Prod Commun 3(3):351–355
Desai KGH, Park HJ (2005) Recent developments in microencapsulation of food ingredients.
Drying Technol 23:1361–1394
Dickinson E (2003) Hydrocolloids at interfaces and the influence on the properties of dispersed
systems. Food Hydrocolloids 17:25–39
Gardu~ no R, Soriano M, Chávez E, Cruz VMT, Del Castillo LM, Casta~ neda-Agullo M (1974)
Proteinasas de plantas mexicanas II. Puntos isoeléctricos y caracterizaci on de formas
moleculares múltiples en enzimas de bromeliáceas. Rev Lat Quı́m 5:243–248
Ortega ML, del Castillo LM (1966) Actividad de la mexicana en presencia de altas
concentraciones de urea. Ciencia Méx 24(5/6):247–251
Palma-Rodrı́guez HM (2008) Efecto de la hemisfericina refinada en la hidr olisis de proteı́na de
suero lácteo. Master science thesis, CEPROBI-IPN, México
Yá~nez-Fernández J, Méndez-Montealvo G, Menéndez-Aguirre O, Cortés-Vázquez MI, Briones-
Martı́nez R (2001) Spray drying of hemisphaericin from Bromelia hemisphaerica. In: Barbosa-
Canovas GV et al (Eds). Proceedings of eight international congress of engineering and food,
Technomic Publishing Co Inc, pp 1156–1160
Salt Crystallization in Chitosan Films
as Affected by Solvent pH and Moisture
Content
Abbreviations
1 Introduction
Chitosan (CH) is a cationic biopolymer used as a matrix to develop edible films and
coatings for food applications. When developing chitosan-based film-forming
dispersions and films, CH is usually dissolved in aqueous acetic acid solutions
(pH ¼ 3.8); sodium hydroxide is often added to increase the pH of the solvent, thus
preventing film solubilization at high relative humidity conditions, which can cause
changes in the solvent properties (pH and ionic strength) affecting the structure and
properties of the obtained films. This is particularly significant for chitosan, since its
conformation in solution greatly depends not only on structural parameters like the
degree of acetylation and chain length, but also on solution parameters such as ionic
strength, solvent, temperature, and pH (Sorlier et al. 2002). During film drying,
changes in solvent properties occur in line with water evaporation: increase in ionic
strength, change in pH, and conformational modifications in the polymer chain.
Moreover, when sodium hydroxide is added to chitosan-based film-forming disper-
sions, sodium-acetate tri-hydrate salts can crystallize, since these are oversaturated
when the film reaches determined water content, depending on molecular mobility,
which in turn is greatly affected by film moisture content.
The purpose of this work is to study the influence of solvent conditions in the film-
forming dispersion (ionic strength and pH) on CH film properties such as water sorption
behavior, and thermal (phase transitions), mechanical, and microstructural properties.
For the adsorption experiments, dry film specimens, in triplicate, were placed at
20 C in hermetic chambers containing oversaturated salt solutions with different
water activity (aw): LiCl (0.11), MgCl2 (0.34), K2CO3 (0.43), Mg(NO3)2 (0.54), KI
(0.72), and KCl (0.85). Samples were weighed periodically until a constant value
was reached, where the equilibrium was assumed to be achieved (Spiess and Wolf
1983). Finally, equilibrium moisture content was determined using a vacuum oven
at 70 C for 24 h.
Salt Crystallization in Chitosan Films as Affected by Solvent pH. . . 587
2.4 Microstructure
Dry films were stored in a desiccator with P2O5 for at least 15 days to ensure that no
water was present in the film. Microstructural analysis of cross sections of the films
was carried out using SEM technique in a JEOL JSM-5410 (Japan) electron
microscope. Pieces of 5 1 mm were cryo-fractured from films and mounted in
copper stubs. Samples were gold coated and observed using an accelerating voltage
of 10 kV.
Water sorption isotherms of films are shown in Fig. 1. Water sorption isotherms for
CH film prepared with solvent I showed a typical sigmoid shape, increasing slowly
in line with aw up to 0.7, beyond which a steep rise in moisture content was
observed, owing to the solubilization phenomenon. This sigmoid shape was also
observed by other authors in pure CH films (Del Nobile et al. 2004; Vargas
et al. 2009). Experimental adsorption data of films prepared with solvent I were
fitted to the Guggenheim–Anderson–de Boër (GAB) model; the predicted sorption
isotherm is shown in Fig. 1. Monolayer moisture content was 0.09 g water/g dry
solids, which is in the range of the values reported in the literature (Fernández
Cervera et al. 2004). For films prepared with solvent II, equilibrium moisture
content values were higher and almost constant in films equilibrated at aw values
up to 0.7. The greater water sorption capacity of the films prepared with solvent II
can be attributed to gain of crystallization water molecules of sodium acetate
tri-hydrate, which supposed 0.256 g of crystallization water/g total dry solids if
all of the present ions form crystals. At the lowest aw values, no differences in
equilibrium moisture contents were observed between films obtained with solvents
I and II, but at aw 0.3 onwards, when molecular mobility in the film reaches a
critical level, salt crystallization promotes the gain of water crystallization mole-
cules. From aw 0.7 onwards, films prepared with solvent II also had greater water
sorption capacity, showing more intense solvent effects, due to the presence of a
higher amount of ions that can bond water molecules.
Fig. 1 Water sorption isotherms (experimental values and GAB fitted model for solvent I) of CH
films prepared with solvent I (empty symbols) or solvent II ( filled symbols) at 20 C
Salt Crystallization in Chitosan Films as Affected by Solvent pH. . . 589
Fig. 2 DSC thermograms of chitosan films prepared with solvent I (solid line) and solvent II
(dashed line) equilibrated at aw ¼ 0.54
Table 1 Initial melting temperature and melting enthalpy (ΔH ) of chitosan films prepared with
solvent II and equilibrated at different water activity values
aw Tm ΔH (J/g film) ΔH (J/g sodium acetate tri-hydrate)
0.11 54 (1) 10 (3) 12 (1)
0.33 42 (3) 152 (2) 238 (1)
0.43 35 (3) 156 (8) 250 (12)
0.54 34 (1) 157 (7) 231 (16)
0.72 20 (5) 172 (26) 288 (47)
Mean values and standard deviation (in brackets)
DSC thermograms of chitosan films obtained with solvent I and II are shown in
Fig. 2, where the characteristic melting endotherm of acetate tri-hydrate (melting
temperature 55.6–56.5 C, melting enthalpy 237–243 J/g, Cabeza et al. 2003) can
be observed for films prepared with solvent II. This result corroborates salt crys-
tallization in the film when ions are present at high concentration in the film-
forming dispersion (solvent II), which notably does not occur at the lowest ionic
strength (solvent I). Table 1 shows the values of initial melting temperature and the
melting enthalpy of chitosan-based films prepared with solvent II as a function of
their water activity. The initial melting temperature of sodium acetate tri-hydrate
crystals decreases with increase in the sample water activity, whereas the melting
enthalpy tends to increase. This result indicates that a different degree of salt
crystallization occurs as a function of the film moisture content. When water uptake
rises in the film and molecular mobility is promoted, crystallization occurs to a
590 M. Vargas et al.
Fig. 3 SEM micrographs of cross sections of chitosan films prepared with solvent I (a) or solvent
II (b)
Table 2 Mechanical properties of chitosan-based films (aw ¼ 0.54) obtained with solvent I and II
Solvent EM (MPa) TS (MPa) E (%)
I 1,383 (150)a 35 (3)a 12 (5)
II 392 (219)b 12 (7)b 15 (7)
Mean values and standard deviation (in brackets)
a,b
Different letters indicate significant differences ( p < 0.05)
greater extent and the area of the crystal melting endotherm increases. The values of
enthalpy of melting were expressed per gram of salt, assuming that all of the present
ions crystallized in accordance with the stoichiometric ratio. Enthalpy values were
within the range reported for sodium acetate tri-hydrate pure salt when aw in the
film was 0.33 or higher; the latter indicates that the total amount of ions can
crystallize when the threshold moisture content is reached, in accordance with the
obtained equilibrium sorption data (Fig. 1).
Figure 3 shows SEM micrographs of chitosan films, where microcrystals of
sodium acetate tri-hydrate can be clearly detected in films prepared with solvent
II. These microcrystals interrupt the polymer matrix and hinder the regular and
linear aggregation of the polymer chains, having a significant effect on the mechan-
ical properties of the films (Table 2). Elastic modulus (EM), tensile strength (TS),
and percentage of elongation at break (%E) values were lower than those reported
in previous studies carried out with chitosan films prepared with acetic acid
solutions (pH ¼ 3.8) without adjusting pH with NaOH (Sánchez-González
et al. 2010). Moreover, all parameters drastically decreased in films prepared with
solvent II, where crystallization of sodium acetate tri-hydrate led to softer and less
stretchable films with less resistance to break. This finding is in agreement with
both the occurrence of interruptions in polymer chain aggregation in the film matrix
and the presence of more discontinuities in the polymer network (Fig. 3b).
Salt Crystallization in Chitosan Films as Affected by Solvent pH. . . 591
4 Conclusion
References
ASTM (2001) Standard test method for tensile properties of thin plastic sheeting Standard D882.
In: ASTM (ed) Annual book of ASTM. ASTM, Philadelphia, pp 162–170
Cabeza LF, Svensson G, Hiebler S, Mehling H (2003) Thermal performance of sodium acetate
trihydrate thickened with different materials as phase change energy storage material. Appl
Therm Eng 23:1697–1704
Del Nobile MA, Buonocore GG, Conte A (2004) Oscillatory sorption tests for determining the
water vapour transport properties of chitosan based edible films. J Food Sci 69:44–49
Fernández Cervera M, Karjalainen M, Airakisan S, Rantanen J, Krogars K, Heinäimäki J, Iraizoz
Colarte A, Yliruusi J (2004) Physical stability and moisture sorption of aqueous chitosan-
amylose starch films plasticized with polyols. Eur J Pharm Biopharm 58:69–76
Sánchez-González L, Cháfer M, Chiralt A, González-Martı́nez C (2010) Physical properties of
chitosan films containing bergamot essential oil and their inhibitory action on Penicilium
Italicum. Carbohydr Polym 82:277–283
Sorlier P, Viton C, Domard A (2002) Relation between solution properties and degree of
acetylation of chitosan: role of aging. Biomacromolecules 3:1136–1342
Spiess WEL, Wolf WR (1983) The results of the COST 90 project on water activity. In: Jowitt R
et al (eds) Physical properties of foods. Applied Science Publishers, London, pp 65–91
Vargas M, Albors A, Chiralt A, González-Martı́nez C (2009) Characterization of chitosan-oleic
acid composite films. Food Hydrocolloids 23:536–547
Water Antiplasticization Effect in Biscuits
as Affected by Glucose and Sucrose Addition
Abbreviations
aw Water activity
B0 Short dough biscuit original formula
B12 Short dough biscuits with sucrose
Cg Measure of the strength of binding of water to the primary binding sites
ERH Equilibrium relative humidity
G12 Short dough biscuits with glucose
GAB Guggenheim-Anderson-de Boer equation
Tg Glass transition temperature
Tgdry Matrix glass transition temperature
Xm Amount of water capable of interacting with all of the available adsorption
sites in a material
1 Introduction
Water is the most effective plasticizer in food matrices, decreasing glass transition
temperature (Tg) and mechanical resistance and determining a softening effect with
concentration increase.
An opposite effect (i.e., hardening, toughening), referred to as antiplasticization,
was observed in some foods in a glassy state undergoing moisture uptake within a
specific moisture or water activity (aw) range and below a specific moisture
Short dough biscuits were made by using a laboratory mixing apparatus, according
to a reference recipe that did not include added sugars, with the following compo-
sition (on weight basis): wheat flour (72 %), water (20 %), and margarine (8 %) (B0,
in the text). In the modified recipe, sucrose (B12) and glucose (G12) were added in
the formulation at a concentration of 12 % by reducing the same amount of flour,
with a final concentration of 60 %. Manufacture of all types of biscuit was carried
out according to a standardized procedure. The ingredients were weighed (on a
500 g batch basis), then mixed for an optimized time to make a dough that was
sheeted, laminated, and cut to obtain specimens of 5.0 cm length, 2 cm
width 0.3 cm height; cooking was carried out at 160 C for 20 min in a profes-
sional oven. After cooling, biscuits were packed in plastic bags and stored in a
desiccator. One series of samples of each type of biscuits was analyzed 24 h after
preparation, while the remaining samples were rehydrated upon water-glycerol
solutions at different equilibrium relative humidity (ERH, 10–90 %) until a plateau
state was reached in the weight increase due to moisture sorption.
Water Antiplasticization Effect in Biscuits as Affected by Glucose. . . 595
Biscuits just after preparation showed moisture content and aw values in the range
of 1.73 and 1.86 % and <0.1, respectively, with no significant differences due to the
presence of sucrose and glucose.
Shear test analysis evidenced a force vs. deformation graph typical of
low-moisture, amorphous matrices characterized by a jagged trend and, after the
first major brittle fracture event, a series of fractures at lower stresses occurred until
fracture was completed. In the bending test, only one main peak related to the
fracture of the specimen was evidenced (data not shown). The substitution of flour
with sugars of different types, but at the same concentration (12 %), significantly
affected the mechanical properties of the products at their lower moisture content
and aw value. In particular, a higher fracture force and modulus, as well as a lower
fracture energy, was determined for B12 and G12 by both shear and bending tests
(Table 1). Strain was not affected by the addition of sugars when shear stress was
applied, while the application of a flexural force evidenced a significant ( p < 0.05)
lower strain in both sucrose- and glucose-added products with respect to B0. The
presence of sugar controls hydration in the dough and tends to disperse protein and
starch granules, thereby impairing the formation of a continuous mass of the
product and leading to a more fragile structure. Moreover, as biscuits cool after
cooking, small saccharides also tend to crystallize, acting as hardening agents,
leading to more crispy products (Bean and Setser 1992). Because of the different
596
Table 1 Mechanical properties of differently formulated biscuits just after preparation (aw ~ 0.1) determined by shear and bending test
Mean sd
Shear test Bending test
Sample Sugar (12 % w/w) Force (N) Modulus (N/mm) Energy (mJ) Strain Force (N) Energy (mJ) Modulus (N/mm) Strain
B0a – 20.05 4.60 16.95 4.23 49.42 14.00 0.44 0.06 9.35 1.82 3.27 0.74 18.32 5.31 0.19 0.05
B12 Sucrose 24.86 2.72 19.42 6.16 36.21 7.26 0.44 0.08 15.32 4.06 2.90 0.90 49.91 7.97 0.11 0.02
G12 Glucose 29.91 6.45 26.22 5.27 36.91 10.26 0.42 0.04 19.10 4.16 4.27 0.71 57.17 14.90 0.12 0.02
a
Reference, no sugars added
P. Pittia et al.
Water Antiplasticization Effect in Biscuits as Affected by Glucose. . . 597
a 50
40
Fracture force (N) - shear
30
20
10
0
0 0.2 0.4 0.6 0.8 1
aw
b 25
B0
Fracture force (N) - bending
20
B12
G12
15
10
0
0 0.2 0.4 0.6 0.8 1
aw
Fig. 1 Fracture force of differently formulated biscuits hydrated at different aw values evaluated
by shear (a) and bending (b) test
increase of the fracture force occurred in the aw range 0.1–0.4 before a significant
decrease of strength occurred.
When flexural stress was applied, a lower fracture force was generally observed
with respect to that obtained by shear (Fig. 1b). Different patterns of change in
fracture force in response to moisture sorption were again observed in all biscuits
types, but with other trends, and several are the factors that could have caused this
result (Pittia and Sacchetti 2008).
Both B0 and B12 showed a softening and plasticizing effect of water at aw values
>0.5, as well as an increase of the strength for aw values below 0.5, which is an
index of antiplasticization. The role of water in biscuits added with glucose was, on
the other hand, strictly that of a plasticizer with a progressive decrease of the
fracture force at increasing aw.
Water Antiplasticization Effect in Biscuits as Affected by Glucose. . . 599
a 2.4
Normalised fracture modulus
1.8
(adimensional)
1.2
0.6
0
0 0.2 0.4 0.6 0.8 1
aw
b 1.8
Normalised fracture modulus
B0
1.5
B12
(adimensional)
1.2
G12
0.9
0.6
0.3
0
0 0.2 0.4 0.6 0.8 1
aw
Data of the Young’s modulus of both shear and bending tests were normalized
with respect to the modulus of the dry product (aw 0.01) to highlight the effects of
sugars on the mechanical properties of matrices due to moisture uptake (Fig. 2).
Young’s modulus can be taken as a measure of the stiffness of a product, and is
widely used to evaluate the effects of water on the textural properties of food
matrices.
The reference sample evidenced only a decrease of the shear modulus at
aw > 0.5, while in the low aw range an increased flexural modulus index of
increased toughness was observed (Fig. 2b). The presence of sucrose in the product
leads to an opposite trend, as increased shear modulus in the low aw range occurred,
while the normalized flexural modulus remained unchanged up to aw values 0.3
above, showing a decrease due to moisture sorption. In the G12 product, a signif-
icant increase of stiffness in a short aw range below 0.3 when shear was applied
occurred, while upon flexure stress, only a plasticization effect was observed.
600 P. Pittia et al.
It is evident from these results that the presence of sugars in the short dough
biscuit matrix had a significant, but not univocal, impact on textural properties due
to moisture uptake. Variations in the response to textural testing may arise due to
differences in sensitivity of different methods, which could influence the detection
of certain changes in mechanical behavior (antiplasticization–plasticization) of
food matrices at different aw (Chang et al. 2000b) as the forces that act on a material
when bending or shear/compressive stresses are applied are different. The same
authors, however, suggested that opposite effects induced by water sorption as
evaluated by similar textural tests could reflect the different macrostructural prop-
erties of the food matrices under study.
Several mechanisms have been identified to explain the moisture-induced hard-
ening or toughening effect referred to as antiplasticization. Harris and Peleg (1996)
hypothesized that partial plasticization by water would make the glassy state more
cohesive and enable a higher deformation to be sustained, corresponding to higher
stiffness. Chang et al. (2000a) proposed a “hole-filling” mechanism involving the
filling of “holes” created under high stress-strain with water molecules effectively
reducing free volume of the system.
The partial substitution of flour with either sucrose or glucose resulted in
matrices with different structural and mechanical properties, as well as different
abilities to interact with water which, in turn, modified the physical and mechanical
properties of the matrix at increasing hydration degree. In the low moisture-glassy
state, both sucrose and glucose became important components of the nonfat con-
tinuous phase (Baltsavias et al. 1999b), as shown by increased fracture force and
modulus and lower strain at fracture upon bending.
Antiplasticization occurring in starch materials could also be due to change of
density of the matrix, dependent on molecular packing and strength of the interaction
between different species, with effects on mechanical properties. It is known that
simple carbohydrate monomer-water mixtures have negative excess volumes of
mixing, arising as a result of the strength of hydrogen-bonded interactions between
water and carbohydrate. The same effect has been observed in a mixed system of
oligomeric and small saccharides in the presence of water (Lourdin et al. 2003).
Moisture uptake of biscuits could be associated with negative volume change with a
correspondent increase in material stiffness and embrittlement, although water
depresses Tg. Sucrose and glucose could have further enhanced this phenomenon
and favored the appearance of an antiplasticization effect due to moisture sorption;
this may occur despite the decrease of matrix glass transition temperature (Tgdry) due
to their addition (Peleg 1996). However, this addition may have affected the change
from a glassy-to-rubber state of G12 and B12 biscuits by making them more prone to
plasticization at increasing moisture uptake after toughening.
Levine and Slade (1993) suggested that the proportion of sucrose in the crystal-
line state and the size of sucrose crystals affect the rate and/or extent of dissolution
in water in cookies. Further, the temperature–water loss profile during baking may
influence the ratio of glassy to crystalline sucrose in the baked product. This, in
turn, would affect the partitioning of water between other hydrophilic components
because glassy sucrose has a higher affinity for water than the crystalline form.
Even if water distribution remains unchanged, the ratio of glassy sucrose to water
Water Antiplasticization Effect in Biscuits as Affected by Glucose. . . 601
B0 B12 G12
a
8
7
Toughness (sensory scale, 1.7)
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
aw
B0 B12 G12
b8
7
Crispness (sensory scale, 1.7)
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
aw
Fig. 3 Score of hardness (a) and crispness (b) evaluated by sensory analysis of differently
formulated biscuits just after preparation (aw < 0.1) and after hydration at increasing aw
may exert an influence on the mechanical properties of the product. Because they
are similar, the process conditions applied to make the differently formulated
products and the differences in the effects of water on the mechanical properties
of the biscuits have to be referred only to the different affinity for water and
physical properties of the small saccharides used in this study.
Differences in the textural properties of biscuits just after manufacturing and
after rehydration at increasing aw values were also evidenced by sensory analysis. A
higher crispness and hardness of B12 and in particular of G12 with respect to B0
were observed, which is in agreement with the results of the mechanical evaluations
(Fig. 3). When products at different hydration degrees were considered, contrary to
602 P. Pittia et al.
the mechanical results, only a decrease in toughness and crispness was perceived,
with critical aw values being different for the various biscuit types. B0 samples
showed an early softening and loss of crispness due to water sorption, as determined
at aw values exceeding 0.2 and 0.4, respectively. The presence of glucose and, in
particular, of sucrose was able to hinder the plasticizing effect of water by signif-
icantly increasing the aw at which softening and loss of crispness occurred, in
agreement with the study results of Valles Pamies et al. (2000) in extruded cereal
products added with sucrose. Differences in initial microstructural and mechanical
properties of products and different moisture sorption behavior due to the presence
of sugars may have affected the perception of textural sensory changes of the
biscuits, as well as hindered those related to antiplasticization.
4 Conclusions
Glucose and sucrose differently affected the mechanical properties of short dough
biscuits by exerting both plasticizing and antiplasticizing agents at increasing
hydration degree.
The different chemical and physical properties of two small saccharides, as well
as their different affinity for water, seem to contemporaneously affect the response
of the complex matrix to moisture increase with changes in mechanical properties
in terms of strength and stiffness, as well as the sensory perception difficult to
unravel. The use of two different mechanical tests seems to offer a good option to
explain the effects of sugars in the matrix at low moisture, as well as at increasing
hydration degrees, but these results warrant further investigation on both water state
and mobility, as well as on the physical and structural properties of these complex
food systems.
References
Baltsavias A, Jurgens A, van Vliet T (1999a) Fracture properties of short-dough biscuits: effect of
composition. J Cereal Sci 29:235–244
Baltsavias A, Jurgens A, van Vliet T (1999b) Properties of short-dough biscuits in relation to
structure. J Cereal Sci 29:245–255
Bean MM, Setser CS (1992) Polysaccharide, sugars and sweeteners. In: Bowers J (ed) Food theory
and applications. Macmililan, New York, pp 69–198
Chang YP, Cheah PB, Seow CC (2000a) Antiplasticisation-plasticisation effects of water on
physical properties of tapioca starch films in the glassy state. J Food Sci 65:445–451
Chang YP, Cheah PB, Seow CC (2000b) Variations in flexural and compressive fracture behaviour
of a brittle cellular food (dried bread) in response to moisture sorption. J Texture Stud
31:525–540
Harris M, Peleg M (1996) Patterns of textural changes in brittle cellular cereal foods caused by
moisture sorption. Cereal Chem 73:225–231
Water Antiplasticization Effect in Biscuits as Affected by Glucose. . . 603
Kalichevsky MT, Blanshard JMV (1993) The effect of fructose and water on the glass transition of
amylopectin. Carbohydr Polym 20:107–113
Katz EE, Labuza TP (1981) Effect of water activity on the sensory crispness and mechanical
deformation of snack food products. J Food Sci 46:403–409
Levine H, Slade L (1988) Water as plasticiser: physico-chemical aspects of low moisture poly-
meric systems. In: Franks F (ed) Water science reviews 3. Cambridge University Press,
Cambridge, pp 79–185
Levine H, Slade L (1993) The glassy state in applications for the food industry, with an emphasis
on cookie and cracker production. In: Blanshard JMV, Lillford PJ (eds) The Glassy State in
Foods. Nottingham University Press, Loughborough, pp 333–373
Lourdin D, Colonna P, Ring SG (2003) Volumetric behaviour of maltose/water, maltose/glycerol
and starch/sorbitol/water systems mixtures in relation to structural relaxation. Carbohydr Res
338:2883–2887
Marzec A, Lewicki PP (2006) Antiplasticization of cereal-based products by water. Part
I. Extruded flat bread. J Food Eng 73(1):1–8
Peleg M (1996) Mathematical characterisation of the plasticizing effect of fructose on amylopec-
tin. Cereal Chem 73:712–715
Pittia P, Nicoli MC, Sacchetti G (2007) Effect of moisture and water activity on textural properties
of raw and roasted coffee beans. J Texture Stud 38:116–134
Pittia P, Sacchetti G (2008) Antiplasticization effect of water in amorphous foods. A review. Food
Chem 106:1417–1427
Seow CC (2010) Antiplasticization of food polymer systems by low molecular mass diluents. In:
Reid DS, Sajjaanantakul T, Lillford PJ, Charoenrein S (eds) Water properties in food, health,
pharmaceutical and biological systems: ISOPOW 10. Wiley-Blackwell, Oxford, UK, pp
115–140
Seow CC, Cheah PB, Chang YP (1999) Antiplasticisation by water in reduced moisture systems.
J Food Sci 64:576–581
Slade L, Levine H (1995) Water and the glass transition—dependence of the glass transition on
composition and chemical structure: special implications for flour functionality in cookie
baking. J Food Eng 24:431–509
Suwonsichon T, Peleg M (1998) Instrumental and sensory detection of simultaneous brittleness
loss and moisture toughening in three puffed cereals. J Texture Stud 29:255–274
Valles Pamies B, Roudaut G, Dacremont C, Le Meste M, Mitchell JR (2000) Understanding the
texture of low moisture cereal products: mechanical and sensory measurements of crispness.
J Sci Food Agric 80:1679–1685
Van Den Berg C, Bruin S (1981) Water activity and its estimation in food systems: theoretical
aspects. In: Rockland LB, Stewart GF (eds) Water activity: influence on food quality.
Academic, New York, pp 1–61
Physicochemical Characterization
of Regional Breads Produced in the Northern
Mountains of Puebla State, Mexico
Abbreviations
Aw Water activity
TPA Texture profile analysis
1 Introduction
Water exists in foods in free and bound forms; only unbound water is available for
microbial growth and biochemical reactions (Prior 1979; Troller 1980). Measure-
ment of unbound water is termed water activity (Aw) and is defined as the ratio of
partial vapor pressure of water in the air to the vapor pressure of water vapor in
saturated air at the same temperature (Scott 1957). This measurement is numeri-
cally equal to the relative humidity at equilibrium expressed as a fraction. Tech-
niques for measurement of Aw have been reviewed by Prior (1979) and Troller and
Christian (1978). In addition, two collaborative studies have compared the various
methods (Labuza et al. 1976; Stoloff 1978). Bread is one of the oldest food products
consumed by humans. There is clear evidence of its use by different civilizations
since prehistoric times. Bread has been a popular product because of its nutritious
properties, and this product uses common ingredients as salt, sugar, and yeast
(Serna 1996). Bread is a basic food made with cereals, usually in the form of
flour, and liquid means, usually water. One of the greatest differences in bread
production is the addition of leavening. Bread can be identified by very different
names according to its form, weight, color, texture, etc., and the way in which it is
presented in different countries as well as in different localities. In Mexico, baking
product consumption is high, where approximately 750 varieties of sweet bread
exist (CANAIMPA 2007).
Proximate analyses of regional breads are conducted done with Kjeldahl method
number 955.04 (Nx5.7), dietary fiber method 962.09, fat method 920.309, ashes
method 923.03, moisture method 44-16, and following official methods (AOAC
1995).
The equipment used for water activity measurement was an Aqualab/Sins electric
hygrometer. The sample dish was modified by inserting a thermocouple through the
wall of the sample holder in order to measure the temperature of the sample
directly.
Using the same device, three samples for each region were placed in a steel tube and
samples were compressed twice in the center to 20 % of sample height at a 1.7 mm/s
rate with an acrylic probe for 15 min at a rate of 1.7 mm/s.
Color was measured in each region, and rewritten as 250 μm, and was recorded
using a Mini Scan Hunter Lab instrument (Minolta, 508d), following the method of
Cortez Escobedo et al. (2010). Color difference for each sample was calculated
Physicochemical Characterization of Regional Breads Produced in the Northern. . . 607
using the equation: ΔE ¼ (ΔL2 + Δa2 + Δb2)/2. Each assay was the average of
20 samples. The blank values were L ¼ 92.26, a ¼ 0.81, b ¼ 0.62.
Table 1 shows the proximal chemical analysis of breads. In this table it is shown
that the Concha Region 2 contained more protein in breads. Breads from Regions
1 and 3 did not contain protein, and fat content was increased in those regions.
Breads from Region 2 showed a higher content compared to other regions. Biscuits
from Region 1 showed more fat content. Conchas did not show significant differ-
ences for any of the regions.
Table 2 shows the water activity of breads for Regions 1, 2 and 3. Region 1 showed
more activity. The Aw was 0.8 in this region.
The wheat loaf bread of Region 1 showed more hardness, which affected elasticity
and cohesiveness (Table 3). This result contrasts with the results of a study of bread
dough in which addition of WPC decreased bread firmness (Güemes et al. 2009).
Means with the same letter in the same column are not significantly different (P >0.05)
a, b
608 N. Güemes-Vera et al.
Table 4 Color of the breads produced in the northern mountains of Puebla state (R1 ¼ Tenango
de Doria; R2 ¼ Necaxa; R3 ¼ Huauchinango)
Bread L a b c h
Concha R1 76.350a 3.8647a 24.239a 24.507a 81.122a
Concha R2 76.227a 4.4914a 24.670a 25.120a 80.024a
Concha R3 77.309a 3.5950a 22.968a 23.257a 81.200a
Bı́squet R1 48.610a 20.7893b 35.442a 41.117a 59.465a
Bı́squet R2 48.323a 22.3212a 37.597a 43.745a 59.259a
Bı́squet R3 33.226b 15.1270c 15.905b 21.987b 45.874b
Means with the same letter in the same row are not significantly different (P >0.05)
A, B
Means with the same letter in the same column are not significantly different (P >0.05)
a, b
Color is a vital trait in bread. Breads from Regions 1 and 2 showed more yellow
coloration with respect to other regions (Table 4), due to the different ingredients
used in bread formulation. The luminosity of breads was highest in conchas with
respect to biscuits.
Physicochemical Characterization of Regional Breads Produced in the Northern. . . 609
4 Conclusions
The best treatment was that used in Region 1; the conchas, the bisquets, and “ojo de
buey” showed the best content of protein and texture. The bread of Region 3 showed
the best content of Aw. The different ingredients used for these foods have influence
in the physicochemical characteristics of regional breads.
References
AOAC (1995) Official methods of analyses, 15th edn. Association of Official Analytical Chemists,
Washington, DC
CANAIMPA (2007) www.eleconomista.com.mx/articulos/200704-1634198. Accessed may 2008
Güemes VN, Totosaus SA, Hernández Ch JF, Soto SS, Aquino Bola~ nos EN (2009) Propiedades de
textura de masa y pan dulce tipo “concha” fortificados con proteı́nas de suero de leche. Ciencia
Tecnologı́a Alimentaria Campinas 29(1):70–75
Cortez Escobedo L, Soto-Simental S, Reyes-Santamaria MI, Hernandez-Chavez JF, Guemes Vera
N (2010) Physicochemical characterization of regional breads produced in the northern
mounting of Puebla State Mexico. Paper presented at the annual meeting for the International
of Food Technology, Chicago Illinois, July 18
Labuza TP, Acott K, Tatini SR, Lee RY, Flink J, McCall W (1976) Water activity determination: a
collaborative study of different methods. J Food Sci 41:910
Prior BA (1979) Measurement of water activity in foods: a review. J Food Protec 42:668
Scott WJ (1957) Water relations of food spoilage microorganisms. Adv Food Res 7:83
Serna S (1996) Quı́mica almacenamiento e industrializacion de los cereales. Primera edici
on AGT
editor, SA, México
Stoloff L (1978) Calibration of water activity measuring instruments and devices: collaborative
study. J Assoc Offic Anal Chem 61:1166
Troller JA (1980) Influence of water activity on microorganisms in foods. Food Technol 34(5):76
Troller JA, Christian JHB (1978) Water activity and food. Academic, New York, NY
Structural Relaxation of Salmon Gelatin
Films in the Glassy State
Abbreviations
Pro Proline
Hyp Hydroxyproline
Tg Glass transition temperature
DSC Differential scanning calorimeter
E Elastic modulus
KWW Kohlrausch–Williams–Watts model
Φ Decrease in the normalized stress
ϕ Relaxation function
t Time
β Width of the relaxation time distribution spectrum
τ0 Characteristic relaxation time
RH Relative humidity
σ Stress
γ Engineering strain
E Young’s modulus
Ta Storage temperature
R2 Coefficient of determination
SE Standard error
ΔH Enthalpy relaxation
Cp Heat capacity
ANOVA Analysis of variance
SD Standard deviations
CV Coefficient of variation
1 Introduction
The unique property of gelatin to form networks and induce plasticity and elasticity
is considered beneficial in the preparation of biopolymer-based packaging mate-
rials. However, its use in food products has been limited by some religious groups
(Gudmundsson 2002). Gelatins from marine sources have emerged as an alterna-
tive, although they are different from mammalian collagen in terms of their
biochemical constituents and therefore in their functional properties. It has been
shown that cold-water fish collagen can have a lower content of amino acids (~16–
18 %) (Gilsenan and Ross-Murphy 2000). Joly-Duhamel et al. (2002) established a
positive correlation of the concentration of amino acids, proline (Pro) and hydroxy-
proline (Hyp), with the melting temperature of native collagen (helix to coil) and
gelatin molecular weight with renaturation temperature (coil to helix), suggesting
an important effect of biochemical composition on the structure stability of the gel
network. An aspect that has not been explored in detail for marine gelatin is the
structural stability of these systems at temperatures below their glass transition
temperature (Tg). Early studies in glassy carbohydrates by Noel et al. (1999)
described the ageing kinetics of maltose in the glassy state in terms of an overshoot
in heat capacity on consecutive heating runs using a differential scanning calorim-
eter (DSC). Badii et al. (2005) evaluated the kinetics of enthalpic relaxation of
bovine gelatin state as a function of the difference between ageing temperature and
Tg. In their later work, the same authors quantified the enthalpic relaxation of the
same model system by DSC, correlating enthalpic values with an increase in the
elastic modulus (E) obtained by mechanical spectroscopy. Work by Lourdin
et al. (2002) described the mechanical relaxation of amorphous potato starch in
the glassy state, obtaining characteristic relaxation parameters by the application of
the Kohlrausch–Williams–Watts (KWW) model (Eq. 1). Anderssen et al. (2004)
discussed the KWW equation in terms of a spectrum of relaxation times describing
the ageing of a polymer. The KWW is an equation that can quantitatively describe
the kinetics of the relaxation process toward an absolute relaxed state (Anderssen
et al. 2004):
Structural Relaxation of Salmon Gelatin Films in the Glassy State 613
" #
t β
ϕðtÞ ¼ exp ð1Þ
τ0
Salmon (Salmo salar) gelatin was extracted from skin by an acid–alkaline method
giving a ~9 % w/v yield of dry gelatin. The gelatin was suspended at 7 % (w/v) in
hot water (70 C), which was then poured onto rectangular Teflon blocks
(30 12 10 cm) and cold cast at 5 C in an incubator for 5 days. The obtained
films (~0.25 mm in thickness) were then cut (100 mm ~10 mm) and stored over
P2O5 (~0 % relative humidity, RH) for 7 days at ambient temperature and later
equilibrated under a ~54 % RH (saturated solution of Mg(NO3)2). The moisture
content was determined gravimetrically in triplicates overnight at 105 C giving
~18.4 % (db.).
The stress (σ) versus engineering strain (γ) curves were obtained by a texture
analyzer (DO-FBO.5TS, Zwick, Ulm, Germany) using a uniaxial tension method.
The application of tensile force was restricted to the linear zone of the stress–strain
curve, giving Young’s modulus (E). Prior to the measurements, the thermal history
of the films was eliminated at 15 C above Tg in silicone (Down Corning 200R) oil
for 10 min, then hermetically packed and quickly cooled to the storage temperature
(Ta) (~29 C), equivalent to 5 C below Tg, and stored for 0, 4, 8, 16, and 40 h. The
stress relaxation of the films was assessed at a constant strain of 1 %. The stress
values were normalized and fitted using Eq. (1), obtaining the values β and
τ0 together with the coefficient of determination (R2), as well as the standard
error (SE) of the fits. The moisture content and temperature of the films were
kept constant throughout the measurements, using a silicone oil containing inner
cylinder in a concentric cylinder chamber previously attached to the texture
614 J. Enrione et al.
analyzer. The outer cylinder was filled with circulating water tempered at the
desired temperature (0.1 C).The rate of mechanical relaxation was predicted
using the derivative of the function KWW with respect to measuring time (dφ/dt)
using the estimated β and τ0 values. Four replicates were evaluated for each sample.
Tg and enthalpy relaxation (ΔH ) were determined by DSC (Diamond DSC, Perkin
Elmer), calibrated with indium Tm ¼ 156.6 C and ΔH ¼ 28.4 J/g. Ten milligram of
each sample was placed in 30 μL aluminum capsules and hermetically sealed. Tg
was defined as the midpoint of the change in heat capacity (Cp) on the second DSC
scan giving a value of ~34 C. The capsules were stored for the same time and
temperature as the mechanical relaxation studies. The thermal history was erased
by holding the sample at 10 C above Tg for 15 min (Lourdin et al. 2002) and then
cooled to the storage temperature (29 C). The methodology used was the follow-
ing: holding at 40 C for 2 min, heating from 40 to 90 C at 10 C/min, holding
for 2 min at 90 C, cooling from 90 to 40 C at 40 C/min, and finally heating
from 40 to 90 C at 10 C/min. The first DSC thermogram indicated denaturation
and melting like endotherms at ~55 and ~70 C, respectively (data not shown). The
ΔH was calculated from the area generated from subtracting the first and second
DSC scans. Each sample was evaluated in triplicate.
The statistical significance of the experimental data was evaluated using the
analysis of variance (ANOVA) with α ¼ 0.05 using Data Analysis Suite in Excel
(Office 2003. Microsoft Corp.). Standard deviations (SD) were calculated for
replicates and plotted as error bars. Coefficients of variation (CV) were calculated
and presented when necessary.
1400
1200
1000
800
4h
0.6
0h
0.5
0 200 400 600 800 1000 1200
Time (s)
Table 1 Parameters β and τ0, percentage of mechanical relaxation (A), and relaxation enthalpy
ΔH, for storage times 0–40 h
Ageing time (h) τ0 (s) SDτ0 β SDβ SE A (%) ΔH (J/g) SDΔH
0 6.1E + 03 3.1E + 02 0.32 0.03 6.4E-02 45.9 0 –
4 1.7E + 04 1.9E + 03 0.28 0.01 4.1E-02 40.4 1.1 0.40
8 1.3E + 05 5.3E + 03 0.24 0.03 6.4E-02 25.6 2.3 0.20
16 3.8E + 05 1.9E + 04 0.22 0.03 5.6E-03 24.5 2.3 0.25
40 9.0E + 05 2.1E + 04 0.21 0.01 3.0E-03 21.9 2.4 0.22
SE standard error from KWW equation fitting, SD standard deviation
differences ( p < 0.05) in stress between fresh aged films after 50 s of tensioning.
Indeed, the amplitude of the decrease (A) between φ at time 0 and 1200 s (Table 1)
was ~46.0 % for the fresh sample and ~21.9 % for the film stored for 40 h. Such
behavior was represented by the KWW model, whose parameters β and τ0 were
obtained individually (Table 1). β decreased from 0.30 to 0.21 ( p < 0.05) when the
616 J. Enrione et al.
1.0
0.20
0.8
(dj/dt)t/(dj/dt)to
(dj/dt)t/(dj/dt)o
0.15
0h
0.6
0.10
0.4
40h
0.05
50 100 150 200
0.2 Time (s)
0.0
0 200 400 600 800 1000 1200
Time (s)
Fig. 3 Normalized mechanical relaxation rate versus time for salmon gelatin films aged at 29 C
(Tg Ta ¼ 5 C) for 0, 4, 8, 16, and 40 h
Structural Relaxation of Salmon Gelatin Films in the Glassy State 617
2.0
1
DCp(Cp1 - Cp0)
40h
0.8 16h
1.5 0.6 8h
4h
Normalized Cp
0.4
0h
0.2
0
1.0 20 30 40 50
Temperature ( °C)
40h
16h
0.5 8h
4h
0h
0.0
0 10 20 30 40 50
Temperature (ºC)
Fig. 4 Normalized Cp versus temperature in salmon gelatin-based films aged at 29 C
(Tg T ¼ 5 C) for 0, 4, 8, 16, and 40 h. CV of Cp replicates <10 % of each ageing time
1.5
1.0
y = 2.48*(1-exp (-0.20*x)) ( R2 = 0,99)
0.5
0.0
0 10 20 30 40 50
Ageing Time (h)
618 J. Enrione et al.
after which ΔH leveled off ( p > 0.05). These results support the mechanical data in
terms of a loss in the internal energy from the system by the spontaneous densifi-
cation of a polymeric matrix approaching equilibrium. In terms of the structure
stability, these data suggest that fish gelatin can undergo similar relaxation pro-
cesses to those reported for biopolymers. In the future, comparative studies of
structure relaxation in gelatin from different biological origins would help to
understand the effect of biochemical composition on the kinetic of this
phenomenon.
4 Conclusions
Acknowledgments We thank Professor John Mitchell for useful discussions. This work was
funded by grants FONDECYT 11075053, PBCT PSD62, and Innova Chile-Corfo CT11 PUT-20.
References
Anderssen RS, Husain SA, Loy RJ (2004) The Kohlrausch function: properties and applications.
ANZIAM J 45:C800–C816
Badii F, MacNaughtan W, Farhat IA (2005) Enthalpy relaxation of gelatin in the glassy state. Int J
Biol Macromol 36(4):263–269
Badii F, Martinet C, Mitchell JR, Farhat IA (2006) Enthalpy and mechanical relaxation of glassy
gelatin films. Food Hydrocoll 20(6):879–884
Gilsenan P, Ross-Murphy S (2000) Rheological characterization of gelatins from mammalian and
marine sources. Food Hydrocoll 14:191–195
Gudmundsson M (2002) Rheological properties of fish gelatins. J Food Sci 67(6):2172–2176
Joly-Duhamel C, Hellio D, Djabourov M (2002) All gelatin network. Biodivers Phys Chem
Langmuir 18(19):7208–7217
Structural Relaxation of Salmon Gelatin Films in the Glassy State 619
Lourdin D, Colonna P, Brownsey GJ, Noel TR, Ring SG (2002) Structural relaxation and physical
ageing of starchy materials. Carbohydr Res 337(9):827–833
Ngai KL (1998) Correlation between the secondary beta relaxation time at Tg with the Kohlrausch
exponent of the primary alpha relaxation or the fragility of glass-forming materials. Phys Rev E
57(6):7346
Noel TR, Parker R, Ring SM, Ring SG (1999) A calorimetric study of structural relaxation in a
maltose glass. Carbohydr Res 319(1–4):166–171
Relationship Between Raw Material
Characteristics and Dehydration Parameters
of Vegetables Dried Using a New Kind
of Solar Dryer
Abbreviations
1 Introduction
In the last few years, an increasing demand for innovative products in response to
changing lifestyles has led to a strong increase in out-of-home food consumption.
Consumers are also becoming more interested in consuming healthy, natural, and
tasty foods because of their positive role in the treatment of a certain type of chronic
degenerative diseases (Bazzano et al. 2002). Dried food products constitute
an important sector of innovative snacks, in particular fruits and vegetables
(F&V)-based products.
G. Bertolo
ACU, Associazione Consumatori Utenti, Milan, Italy
A. Maestrelli • M. Della Campa (*)
Consiglio per la Ricerca e la Sperimentazione in Agricoltura, Unita di Ricerca per i Processi
dell’Industria Agroalimentare C.R.A.-I.A.A., via G. Venezian 26, Milan, Italy
e-mail: [email protected]
Eggplant, Solanum melongena var. Black Bell; onions, Allium cepa var. Rossa di
Tropea; and mushroom, Agaricus bisporus var. Avellanus, were purchased in the
local market. All products were hand-washed before cutting and drying. Mushroom
and onion were cut manually (slice thickness resulted 5.7 0.3 mm for mushroom
and 4.5 0.3 mm for onion). Eggplant slices were cut manually into 5.0 0.3 mm-
thick slices. The average weight of a slice resulted to be 11.8 1.4 g (eggplant),
7.4 1.0 g (onion), and 3.9 0.6 g (mushroom). The values of thickness were
obtained according to Ertekin and Yaldiz (2004). Product slices were split into two
groups, for conventional (C) and solar (S) drying, using the equipments and
according to the conditions listed below. For all samples, end of drying was
considered as completed when the product reached constant weight. The experi-
mental scheme of the drying process is presented in Fig. 1.
Solar drying was performed using the prototype named “Eolo” (Figs. 2 and 3). This
system consisted of a cabinet 54 cm high, 100 cm long, and 150 cm deep. All walls
were made of alveolar polycarbonate transparent panels. Atmospheric air enters the
heating chamber through slots on one of the cabinet’s sides, while the air outlet
ducts are in the opposite side; airflow is induced by two fans placed on the bottom of
the dryer, powered by a 12 V 50 A h battery, charged by a solar photovoltaic panel
system, located at a side of the dryer. A 5 cm layer of steel wool located under the
heating chamber roof receives the solar radiation and thus acts as an internal air
heater. The heating layer also increases the heat transfer area and decreases airflow
so increasing air temperature. The heated air enters the drying chamber, in which
there are located four trays, with a total surface of 2.4 m2 on which the fresh
Eggplant Mushroom Onion
Solanum Agaricus Allium cepa
Melongena bisporus Var. Rossa di Trope
Var. Balck Bell Var. Avellanus calabria
SOLAR CONVENTIONAL
Fig. 3 Scheme of the airflow inside the solar dryer and position of the thermal probes
material is distributed for drying. A metallic mesh is located at the air inlet to
prevent the entrance of insects and particles. During the exposure to sun radiation,
the temperature reached inside the system can be different if measured in indiffer-
ent positions inside the plant (see probe positions as shown in Fig. 3) and depends
on the instant solar radiation intensity: an example of the dryer thermal perfor-
mance during a sunny/cloudy summer day at the latitude of Milan, Italy, is shown in
Fig. 4.
Although this solar dryer system is able to provide a forced hot airflow, it can
operate completely independent from the electricity supply. Moreover, a thermal
sensor stops the fans during the night, to prevent the entrance of moist air into the
drying chamber. Structure of the cabinet also protects the product from the weather,
dirt, and dust.
c
51.07
46.07
Instant solar radiation intensity
T1
T2
T3
41.07 T4
T5
T6
T7
36.07
31.07
26.07
21.07
09:06 13:01 16:56 20:51
Hour of day
Fig. 4 Example of solar dryer performance during a sunny/cloudy summer day at latitude of
Milan, Italy
Dried was considered complete, when the materials reached constant weight.
Reported results are the mean of 3 measurements.
Water activity (aw) was evaluated by means of an electronic hygrometer
(AquaLab CX-2—Decagon Devices, Pullman, USA). Reported results are the
mean of 6 determinations.
Texture was determined by applying bending–snapping tests using a Stable
Micro Systems TA XT II plus fitted with a 5 kg cell and a three-point bend rig.
The elastic module, E, was determined as the slope of the stress–strain line before
the first fracture (largest peak) and was calculated from the force–displacement
curves and was considered as the crispness index (Konopacka et al. 2002). Dried
eggplant slides were cut into strips 15 mm width and sliced lengthwise. No further
cutting was necessary for mushroom and onion. Presented results are the mean
value of 10 determinations.
Color was evaluated on fresh and dried samples by using a D65 illuminant/10
observer refraction colorimeter (Minolta Chroma Meter CR 200—Minolta Camera
Co. Ltd., Japan). Samples were placed on a white standard plate (L ¼ 100), and CIE
La*b* coordinates were also evaluated. Presented results are the mean of
10 determinations.
626 G. Bertolo et al.
Quality characteristics of the three dried vegetables depend on tissue (and cellular)
structure and drying technique and rate. Solar drying gave place to products with
higher moisture content and water activity than those obtained by conventional
drying (Table 1) which were also softer than those obtained by hot air drying. As
expected, both drying techniques caused a significant decrease of L and increase of
a* values (Fig. 5). No significant differences in L values were detected between
solar and conventionally dried vegetables. Values for a* corresponding to solar
dried eggplant and onion slices were significantly higher than those for conven-
tional vegetables, while no differences were observed for mushrooms.
During conventional drying, a power meter was used to evaluate the processing
energy needs. Energy consumption for the conventional product was about 2 kW h
for 514 g of eggplant, 1 kW h for 220 g of mushroom, 2 kW h for 580 g of red onion,
and 0 W for the solar product.
Table 1 Moisture content, water activity, and elastic module measured on dried strips of the
different products
Moisture content (MC) Water activity Elastic module (E)
(g H2O/100 g product) (aw) (g/mm)
TQ C S C S C S
Eggplant 92.18 4.77 13.90 26.05 53.53 260.88 245.89
Mushroom 90.91 2.13 5.52 18.40 36.37 425.60 353.85
Onion 91.71 6.73 15.92 25.25 51.5 189.81 114.45
Before (TQ) and after conventional (C) and solar (S) drying up to constant weight
L a*
100,00 14,00
b b b
90,00 12,00
a a
80,00 10,00
a b
70,00 8,00
a a a
ba b b
60,00 6,00
a c
50,00 4,00
a
40,00 2,00
a b
30,00 0,00
-2,00 TQ C S TQ C S TQ C S
20,00
10,00 -4,00
0,00 -6,00
TQ C S TQ C S TQ C S -8,00
Eggplant Mushroom Onion Eggplant Mushroom Onion
Fig. 5 Comparison of initial color data L a* of eggplant, mushroom, and onion slices before
(TQ) and after conventional (C) and solar (S) drying up to constant weight. In each column,
different letters stand for significant differences (Tukey’s test, p <0.05 %)
Relationship Between Raw Material Characteristics and Dehydration Parameters. . . 627
4 Conclusions
Drying vegetables using solar drying systems (such as the prototype used in this
work) can provide significant benefits to the quality of products as compared to
those obtained by applying convection drying systems such as:
– Softer products although sticky in the mouth
– Improved product hygiene
– No additional energy to solar radiation required for drying
In addition, the low temperature reached during drying could preserve important
thermo-liable/high-nutritional-value compounds; further investigation will be
developed on this aspect.
References
Bazzano LA, He J, Ogden LG, Loria CM, Vup-puturi S, Myers L, Whelton PK (2002) Fruit and
vegetable intake and risk of cardiovascular disease in us adults: the first National Health
Examination survey Epidemiologic Follow-up study. Am J Clin Nutr 76:93
Ertekin C, Yaldiz O (2004) Drying of eggplant and selection of a suitable thin layer drying model.
J Food Eng 63(3):349–359
Konopacka D, Plocharski W, Beveridge T (2002) Water sorption and crispness of fat-free apple
chips. J Food Sci 67(1):87–92
Giri SK, Prasad S (2007) Drying kinetics and rehydration characteristics of microwave-vacuum
and convective hot-air dried mushrooms. J Food Eng 78(2):512–521
Caking Process and Microstructural Changes
of Wall Materials Used in Spray-Drying
Process
Abbreviations
A Agglomeration
C Compaction
L Liquefaction
T Storage temperature
1 Introduction
2.1 Materials
Soy protein isolate (SPI) (Cenit, México), maltodextrin 20 DE (MD) (Globe Corn
Product International, Illinois, U.S.A.), and gum arabic (GA) (E number E414)
(Morevo Quick Gum) were used as wall materials.
Suspensions of the wall materials SPI, MD, and GA were prepared according to
Table 1, keeping constant a concentration of 20 % of total solids, using distilled
water, and applying mechanical agitation during 12 h at room temperature (22 C).
2.2 Methods
Dispersions were fed into a spray dryer (Mobile Minor™ 2000, GEA Niro, Den-
mark). The equipment was operated with cocurrent flow, pneumatic nozzle, 150 C
as inlet drying-air temperature, and 70 C as outlet drying-air temperature. The
powders were stored in polyethylene bags with a hermetic seal until later analysis.
Water activity (aw) of each powder was analyzed using an AquaLab (four TE
model, Decagon Devices, USA). The moisture content was determinate gravimet-
rically following the official AOAC method (AOAC 2005).
One gram of powdered wall materials was placed in a sealed flask containing a
saturated solution of NaCl (75 % relative humidity, RH); after a week, the sample
was weighed and the difference of moisture gained was expressed as g of moisture/
100 g dry solids (Tonon et al. 2008).
Differential scanning calorimetry (DSC, Perkin Elmer Diamond, USA) was used to
measure wetted powder glass transition temperature range. Samples of powder
(15 g) were prepared in a DSC aluminum pans (40 μL), and the pans were then
hermitically sealed. All measurements were made at a scanning rate of 10 C/min,
from 20 to 120 C (Acevedo et al. 2006).
All samples of powdered wall materials (spray dried) were constituted by individ-
ual microparticles of spherical geometry. The moisture content values varied
between 2.21 and 4.38 (% dry base, d. b.) and aw between 0.10 and 0.16 (Table 1).
Cuq et al. (2010) indicate that food powders with low water activity values are
Caking Process and Microstructural Changes of Wall Materials. . . 633
stable and could have low biochemical reactivity; similar materials were obtained
in this study.
The moisture sorption by the wall materials is an important physicochemical
property that regulates processing conditions and storage stability. Values moisture
content achieved by the samples (individually, in binary, and ternary mixtures)
exposed at 75 % RH are observed in Table 2; in this case, GA was the material that
absorbed the highest % of water (20.167 % d. b.); it means five times the initial
moisture content. This could be related to the water-holding capacity of the GA,
5.7 g of water/g gum arabic, and its solubility 0.5 g GA/mL 1 mL of water
(Robertson and Eastwood 1981).
Evident changes in the microstructure of storage powders 75 % RH were
detected (Fig. 1). Bhandari and Howes (1999) reported similar results; in this
context, Rao and Labuza (2012) indicated that changes in powders properties
during storage affect its functionality and production. All samples suggested caking
process, except SPI sample (Fig. 1). This one kept microstructural integrity due to
its low water solubility, although this sample adsorbed fourfold initial moisture
content. However, blends containing SPI (SPI/GA, 13SPI/GA/MD, and SPI/GA/
13MD) may develop caking process, but its intensity depends on concentration of
highly water-soluble components. In case of GA and its blends showed caking
process with evident bridging, agglomeration, and stages of compaction. In blends
of GA with high concentration of maltodextrin the liquefaction phenomenon was
developed (Fig. 1h). Microstructure of samples containing high concentration of
maltodextrin (MD, MD/GA, APS/MD, SPI/GA/13MD) suffered collapse. This
phenomenon is a microstructural compaction that is associated with a pronounced
loss of the integrity of the system, as result of the increase of diameter of bridges
interparticles and reduction of interspaces. Microparticles prepared with the
formulations MD and APS/GA/13MD achieved a structural collapse where the
junctions between particles disappeared, due to solubilization of the low
molecular weight fractions (Aguilera et al. 1995). Similar results were observed
Fig. 1 SEM images of powders stored at 75 % relative humidity, illustrating stages of caking
process. (a) SPI, magnification 12000, (b) GA, magnification 12000, (c) MD, magnification
1500, (d) SPI/GA, magnification 24000, (e) MD/GA, magnification 6000, (f) SPI/MD,
magnification 12000, (g) 13SPI/GA/MD, magnification 12000, and (h) SPI/GA/13MD,
magnification 6000
Caking Process and Microstructural Changes of Wall Materials. . . 635
4 Conclusions
Acknowledgements The authors wish to thank the Instituto Politécnico Nacional (National
Polytechnic Institute) for the financial support provided through the SIP project: 20140253 and
20140554, BEIFI-IPN, COFAA-IPN, CONACyT, project 216044, and the scholarships for the
PhD studies of Josefina Porras-Saavedra. The authors are grateful for experimental support of
Universidad de Buenos Aires and Instituto Mexicano del Petr oleo.
636 J. Porras-Saavedra et al.
References
Abbreviations
H. Necoechea-Mondragon (*)
Direcci
on de Investigacion, Secretarı́a de Investigacion y Posgrado, Instituto Politécnico
Nacional, Miguel Bernard Av. Luis Enrique Erro S/N, Unidad Profesional Adolfo L opez
Mateos, Zacatenco, Delegacion Gustavo A. Madero, México City, DF CP 07738, Mexico
e-mail: [email protected]
D.Y. Morales-Delgado • E. Parada-Arias • D.I. Téllez-Medina
Departamento de Graduados e Investigacion en Alimentos, Escuela Nacional de Ciencias
Biol
ogicas, Instituto Politécnico Nacional, Carpio y Plan de Ayala s/n, Mexico City,
DF CP 11340, Mexico
M. Cornejo-Maz on
Departamento de Biofı́sica, Escuela Nacional de Ciencias Biol
ogicas, Instituto Politécnico
Nacional, Carpio y Plan de Ayala s/n, Mexico City, DF CP 11340, Mexico
Food product development (FPD) is a systematic task with the objective of gener-
ating new products by either modifying any existing or producing a new and
original one. Driven by consumers’ expectations, new knowledge, globalization,
and shorter product cycles, among others, companies must survive in a dynamic and
often unpredictable markets characterized by high levels of competitiveness
(De-Greef et al. 2014).
Development of new products has become more relevant to companies and, in
the food industry, holds an important and strategic role, becoming a key aspect
when trying to achieve higher levels of competitiveness. The development process
of new products is fundamentally based on the ability for better identifying ways of
meeting market needs and in the capacity of interpreting such information into
products, which in fact offers a better solution for such needs in relation to its
competition (Gonçalves-Moreira and De-Toledo 2005).
FPD is recognized as a core process that ensures the success of companies;
improving process is a key factor for the success of a company. The importance of
FPD continues to grow in response to increasingly rapid technological advances
and to globalization, which has, worldwide, opened markets to free trade and giving
place to high levels of competitiveness.
Product quality in the food sector is one of the priority criteria to analyze its
performance in the consumer market and is a variable that directly affects the
design and production process. Development time is an important factor, since it
must follow changes in customer requirements, incorporate new technologies, and
attend the increasing competitiveness of the market in the introduction of innova-
tive products. Finally, productivity of the development process must be efficient,
which means that new product development involves searching the best perfor-
mance of material, financial, and human resources available on the company, and
one of the main challenges of the food product development task is to manage the
best combination of these three factors for an efficient process. Consequently,
systematize appropriate models and tools will help to consolidate the process
management (De-Greef et al. 2014).
In the literature there are several proposals to systematize the activities of FPD
using models, according on the area of knowledge of the authors. A large number of
FPD, best manufacturing practices, tools, methods, and systems have been devel-
oped throughout the years to help companies to improve their FPD schemes
(Fig. 1).
Nonetheless, numerous companies have consistently failed in their attempts to
develop their products on time, within budget, and with the expected features and
quality (Rozenfeld et al. 2006). This is why the design of products based on market
and consumer present and future demands is an important engineering issue in
which food structuring tasks play an important role. Food matrixes are complex and
active structures; water and other components interact during and after
Food Matrix Structure Quality Preservation: Water Removal Operation. . . 639
De Greeff
Arcusin
Earle (1985) Rudolph Penso (2003) Rossetti (2014)
Divided PDP in (1995) which approaches all describes a set of tools
the food industry Model adopted the product stages
Buzzel & by Arthur D.
developed by different
Nourse (1967) into the seven from the pre- authors that can be
First techniques for the stages which Little company, development applied in the Pre-
development and integrate market and it is stage to the eventual Development of Product
processing of food research and organized into product retirement Development Process in
products technological three main food industries.
research stages.
manufacturing, and preservation processes such as those related with water removal
aim to slow the magnitude of these interactions to minimize rates of degradation.
Knowledge and systematization of structure-function relationships are important
towards development of products and processes.
The structured food systems (i.e., cellular tissues) are dissipative structures whose
functionality mainly concerns their properties and characteristics (physical-
chemical properties, chemical, and biochemical reactions), external interactions
with surroundings (interactions with microorganisms, heat, and mass transport
pathway), and especially, their interactions with consumers (nutritional value,
quality, taste and flavor, texture, appearance: size, shape, color). Dehydration or
rehydration processes involve heat and mass transport of water and solutes coupled
with phase transitions as well as with nano-, micro-, and macrostructural changes,
thus producing important effects on food functionality, and control of these phe-
nomena is a major concern in FPD which must be applied not only to the changes in
physical-chemical properties but also to those related with demands of consumers
(Fito and Chiralt 2003).
Food matrix engineering is a branch of food engineering which aims to apply the
knowledge of the food matrix composition, structure, and properties to promote and
control changes which can improve some sensorial and/or functional properties in
the food (Fito and Chiralt 2003). These changes are related to the phenomena of
640 H. Necoechea-Mondrag
on et al.
It is accepted that relevant food properties are related to their structure and that key
elements defining the architecture of same are not discernible by the eye (Ramı́rez
et al. 2009). Thus, the use of several microscopy techniques and other imaging
methods has become common in the study of food microstructures (Aguilera 2005).
It follows that images (photomicrographs) are extensively used in the scientific
literature as evidence of the structural features of food materials. These images are
usually used as support for a qualitative description of the microstructure and
complement data generated by rigorous protocols of sampling and preparation,
chemical analyses, and determination of physical properties.
Food Matrix Structure Quality Preservation: Water Removal Operation. . . 641
Fig. 2 Images of strawberry dehydrated at four different temperatures: 60, 70, 80, and 90 C
Fig. 3 Images (gray scale) of strawberry tissues; fresh and after convective drying at indicated
temperatures (20)
0 20 40 60 80 100
0.000
y(60°C) = -0.0592x + 2.0935
-0.500 R² = 0.9931
y(70°C) = -0.0946x + 3.1051
R² = 0.9755
-1.000 y (80°C)= -0.0989x + 2.3478
R² = 0.9788
LN X/X1
Fig. 4 Representation of the solution to Fick’s second law for the falling rate period of drying
samples of strawberry at 60,70, 80, and 90 C, drying air speed of 1 m/s
Effective diffusivity values of strawberry samples that were evaluated during the
drying process are shown in Table 1. It was observed that, when the temperature
increases, the samples of strawberry shrink faster, and thus, water mobility
increases.
Doymaz (2008) reported that the transport of water during drying of strawberry
was well described by Fick’s mechanisms and that the corresponding coefficients of
effective diffusivity were between 4.95 1010 and 1.42 1010 m2/s (Doymaz
2008). The coefficients of effective diffusivity of the present study agreed with
those reported by Doymaz 2008, concluding that the diffusion coefficients and
drying rates increased with temperature.
5 Final Remarks
techniques and the numerical models that describe the relationships between the
structure and properties could be used in different types of food designs for food
matrix quality preservation as in water removal in drying processes, contributing to
the systematization and integration of the food product development process in
food companies. Water removal was studied at cellular levels during drying of
strawberries; it was observed that at microlevels, the cell contours of samples dried
at higher temperatures were more disrupted than those observed at lower
temperatures.
References
Ramı́rez C, Germain JC, Aguilera JM (2009) Image analysis of representative food structures:
application of the bootstrap method. J Food Sci 74(6):65–72
Rozenfeld H, Forcellini FA, Amaral DC, Toledo JC, Silva SL, Alliprandini DH (2006) Gestão de
desenvolvimento de produtos: uma referência para a melhoria de processo. Ed. Saraiva, São
Paulo
Sánchez-Segura L, Téllez-Medina DI, Evangelista-Lozano S, Garcı́a-Armenta E, Alamilla-
Beltrán L, Hernández-Sánchez H (2015) Morpho-structural description of epidermal tissues
related to pungency of Capsicum species. J Food Eng 152:95–104
Editors’ Biographies
Politécnico Nacional, Mexico, and his Ph.D. in Food Technology from the Poly-
technic University of Valencia, Spain. He has been a Professor of Food Science and
Food Engineering at the Escuela Nacional de Ciencias Biol ogicas, of the Instituto
Politécnico Nacional, México, and Secretary General of this institute, Director
General at Instituto Mexicano del Petr oleo, and is currently Director General of
Centros de Formaci on para el Trabajo, Secretarı́a de Educacion Pública, México.
Gustavo V. Barbosa-Cánovas, a Fulbright Scholar, received his BS in Mechan-
ical Engineering at University of Uruguay and MS and Ph.D. at University of
Massachusetts. He also received a Honoris Causa Doctorate from Polytechnic
University of Cartagena, Spain. He is a Professor of Food Engineering and Director
of the Center for Nonthermal Processing of Food at Washington State University.
Index
E F
Edible/biodegradable hydrocolloids Fabela-Mor on, M.F., 345–350
microstructure of, 102, 103 Fabra, M.J., 101–110, 112
optical properties, 104 Fa-De Li, 529
phase transitions, 108–111 Farroni, A.E., 27–38, 125–133
plasticizers, 102 Fermi equation, 128
polar lipids, 103–104 Fibers submitted to modified acid-alkaline
protein-lipid ratio, 103 hydrolysis without pressure (FAAT)
WSC α,α–trehalose decomposition, 563
film plasticizers, 104–105 chemical composition, 558–561
hydrocolloid films, 105–108 DSC, 559, 561–562
Edible films (EFs) Fourier transform infrared spectroscopy, 559
active EFs fructooligosaccharide decomposition, 563
antimicrobial compounds, 454 materials, 558
deterioration, 453–454 preparation, 558
strawberry weight loss, 456–459 thermogravimetric analysis, 559, 561–562
characterization, 457 x-ray diffraction, 559, 563
functional properties, 452 Fibers submitted to modified acid-alkaline
lipids, 452 hydrolysis with pressure (FAATP)
MLC, antimicrobial activity, 455, α,α–trehalose decomposition, 563
457–458 chemical composition, 558–561
modified corn starch, 453 DSC, 559, 561–562
plasticizers, 452–453 Fourier transform infrared spectroscopy, 559
properties, 455 fructooligosaccharide decomposition, 563
starch filmogenic suspension, 454–455 materials, 558
surfactants and emulsifiers, 453 preparation, 558
Egg yolk phosphatidylcholine (EPC) thermogravimetric analysis, 559, 561–562
cooling rates, 221, 222 x-ray diffraction, 559, 563
freezing temperatures, 222–223 Fibers without treatment (FAWT). See Agave
holding temperatures, 223 tequilana leave powder
Eichner, K., 42 Fickian diffusion, 150
Electrical conductivity (EC), 528–530 Figueroa-Cárdenas, J.D., 527–530
Electron microscope by environmental mode Film barrier properties, 108
(ESEM), 35 Fine, F., 253
Electron spin resonance spectrometer Finite element method (FEM), 384, 389
(ESR), 116 Flame ionization detector (FID) method, 312
Emulsifier. See Starch, gelatinization Flores-Andrade, E., 15–23
Enrione, J., 405–410, 611–618 Flores-Morales, A., 557–564
Enterobacter sakazakii, 249–251 Flores, N.L., 568
Enthalpy relaxation Flory-Huggins interaction parameter, 151,
amorphous carbohydrates 153, 154
KWW, 162–163 Fluidized bed drying process, allspice, 568–571
MDSC, 164 Fogliano, V., 48
molecular weight dependence, Food appearance
167–171 chromatic attributes, 32
milk powder, 443–445 cornflakes production, 31
Equilibrium moisture content (Mw), 463–464 Kubelka–Munk analysis, 32–33
Equilibrium relative humidity (ERH), opacity, 32
446–447 translucence changes
Ertekin, C., 622 cereal products. (see Cereal products)
Esteban, M.A., 280 fruit products, 33–34
Ewing, S., 75–86 transparency/opacity changes
Extract of propolis (EEP), 108 dehydration or rehydration, 29–30
Eyring model, 303 light scattering particles, 30–31
656 Index
R
Rambutan S
annealing process, 416, 418 Sablani, S.S., 161–173
DSC determination, 415–417 Sacchetti, G., 289–297, 593–602
HPLC, 415 Sáez, C., 611–618
statistical analysis, 416 Sahagian, M.E., 416
sugar composition, 418–419 Saleeb, F.Z., 21
Ramos, J.J., 81 Saleki-Gerhardt, A., 302
Rana, R.K., 559 Salmon gelatin films
Rao, Q., 633 acid–alkaline method, 613
Refined hemisphaericin microencapsulated enthalpy relaxation, 614, 617–618
(RHME) uniaxial tension method
arabic gum constant strain, 613
CCD, 577–579 KWW model, 614–616
isolated enzymes, 576 modulus vs. ageing time, 614–615
preparation, 577 moisture content and temperature,
protein concentration assay, 578, 579 613–614
proteolytic activity assay, 578, 580 normalized relaxation rate vs. time, 616
RS design, 577, 579–581 parameters, 615–616
SEM, 578 stress relaxation, glass, 614–615
yield, 578, 581–582 Young’s modulus, 613
spray drying. (see Spray-drying Salt crystallization
process) DSC, 587, 589–590
Regalado, C., 451–459 mechanical properties, 587, 590
Regional breads (Puebla State, Mexico) microstructural analysis, 587
color measurement, 606–608 preparation, 586
proximal chemical analysis, 606–607 SEM micrographs, 590
TPA, 606–608 statistical analysis, 587
water activity, 606–608 water sorption isotherms, 586, 588
Relative vapor pressures (RVP), 126 Sánchez-Mundo, M.L., 533–539
Rengsutthi, K., 413–419 Sandhu, K.S., 372
Response surface (RS) design, 577, 579–581 Santagapita, P.R., 353–360
Response surface methodology (RSM) Santiago-Pineda, T., 567–573
dough composition Santos, M.V., 383–392
starch gelatinization, 371–373 Sastry, S.K., 529
water mobility, 369–371 Scanning electron microscopy (SEM), 35, 578
hydrocolloids and proteins, 368 agave powders, 347, 348
structural functions, 368 allspice, 569
surface response analysis, mixture designs, Capsicum extract, 340–343
368–369 cytosine, 179, 181
Retention index, 502–503 FEI Nova nanoSEM field emission,
Reyes, L.A., 575–583 177–178
Reyes-Santamaria, M.I., 605–609 theophylline, 179, 182
Reyes-Vega, M.L., 527–530 thymine, 179, 183
Index 663