- Email: [email protected]
Effect of glycerol on water sorption of bovine gelatin films in the glassy state
Available online at www.sciencedirect.com
Procedia Food Science 1 (2011) 267 – 274
11th International Congress of Engineering and Foods (ICEF11)
Effect of glycerol on water sorption of bovine gelatin films in the glassy state Paulo Paolo Díaz, Carol Arratia, Corina Vásquez, Fernando Osorio, Javier Enrionea* a
Departamento de Ciencias y Tecnología de Alimentos, Universidad de Santiago de Chile. Obispo Umaña 050, Estación Central, Santiago, Chile.
Abstract The addition of low molecular weight compounds to a biopolymer in the glassy state can generate changes in macrostructural properties of the matrix. This work studied the effect of glycerol on water sorption properties of gelatin films in the glassy state. Bovine gelatin-glycerol (0-10% dry basis, db) films were equilibrated to moisture contents (MC) in the range of 5 - 30% (db). Results show that the addition of glycerol generated differences in sorption behaviour by the decreasing of sorption parameters (GAB, BET and Freundlich), indicating complex glycerol-matrix interactions. Gas pycnometry showed a densification of the matrix suggesting a reduction in sorption site availability when glycerol content increased. © 2011 Published by Elsevier B.V. Open access under CC BY-NC-ND license. Selection and/or peer-review under responsibility of 11th International Congress on Engineering and Food (ICEF 11) Executive Committee. Keywords: gelatin; glycerol; water sorption; specific volume; glassy state.
1. Introduction Among hydrocolloids, gelatin is widely used in food and pharmaceutical industries due to its various functional properties. Gelatin is obtained from denaturation of collagen by acidic-alkaline treatments. Collagen is the structural unit of skin, hair and bones of animals [1]. One of its most important properties is its ability to form thermoreversible gels, however new uses are related with its capacity to form edible films or as encapsulating material for certain nutrients and drugs [1]. The addition of low molecular weight compounds, i.e. glycerol, to low moisture content gelatin matrix improves its mechanical properties generating a more flexible structure. However it can alter the mechanism of polymer-water interactions, thus changing the water sorption properties of the material. Literature suggests that this effect can be different if the matrix is in the glassy or rubbery state [2-4]. _____________ * Corresponding author. Tel.: 56-2-7184524; fax: 56-2-7764089. E-mail address: [email protected]
2211 601X © 2011 Published by Elsevier B.V. Open access under CC BY-NC-ND license. Selection and/or peer-review under responsibility of 11th International Congress on Engineering and Food (ICEF 11) Executive Committee. doi:10.1016/j.profoo.2011.09.042
268
Paulo Díaz et al. / Procedia Food Science 1 (2011) 267 – 274
Also it has been reported that the presence of plasticizer can produce changes in density of a carbohydrate based matrix [5, 6]. Little information is available looking at the changes in sorption properties and densification phenomena in protein based systems, such as gelatin. The aim of this work was to study the effect of glycerol on water sorption properties of bovine gelatin films in the glassy state and to assess variations in its macrostructure by specific volume determinations. 2. Materials & Methods 2.1. Film Preparation A 7% (w/v) bovine gelatin (type B, Bloom 220, Rousselot, Brazil) suspension at 50ºC was mixed with different glycerol (purity > 99.5%, Merck, Germany) contents (GC) of 0, 2, 6 and 10% (dry basis, db) and then deposited onto tefloned containers and maintained at 7 ± 1ºC for seven days until translucent films (~0.3 mm in thickness) were obtained. Films were maintained hermetically at ~0% Equilibrium Relative Humidity (ERH) under P2O5 for one week. The films were then equilibrated isothermally at ~20ºC under 0, 11, 22, 33, 44, 68, 75 and 85% ERH using saturated salt solutions. Thymol was used as antimicrobial agent for the high ERH values. Equilibrium was followed gravimetrically until the difference between two measurements was lower than 0.5%. The equilibrium stage generated moisture contents (MC) in the range of 5 - 30% (db). 2.2 Sorption Isotherms The water sorption isotherms were fitted by the mathematical relations of Brunauer, Emmet and Teller [7] (BET), Guggenheim, Anderson and de Boer [8] (GAB) and Freundlich [4]. BET Equation is defined as: ݉Ͳ ݓܣ ܶܧܤܥ (1) ܯൌ ሺͳ െ ݓܣሻሺͳ െ ݓܣ ݓܣ ܶܧܤܥሻ
where M the moisture content (%, db) and m0 is the moisture content at the monolayer (%, db). CBET is a temperature-dependent constant (related to the net heat of sorption associated to the monolayer) and Aw water activity [2] or relative humidity at thermodynamic equilibrium. GAB Equation is defined as: ݉Ͳ ݓܣܭ ܤܣܩܥ (2) ܯൌ ሺͳ െ ݓܣሻሺͳ െ ݓܣܭ ݓܣܭ ܤܣܩܥሻ
where M the moisture content (%, db) and m0 is the monolayer value (%, db). T CGAB is related to the energy associated with the interaction between water molecules and the matrix primary interaction sites or monolayer [2]. KGAB is also a temperature-dependent parameter related to the heat of sorption at the multilayer. BET equation is a generalization of the Langmuir model for the sorption of gas onto a solid surface through multilayers [7], but assuming that evaporation-condensation properties of the molecules of the second and higher adsorbed layers are the same as of bulk water [9]. Meanwhile GAB equation was derived from BET equation but assuming that the state of the sorbent molecule in the second and higher layers are equal but different from the liquid-like state, introducing an additional degree of freedom (parameter K), which gives to the GAB model its fitting versatility [9]. Freundlich equation has also been proposed to describe sorption in glassy materials with an energetically heterogeneous distribution of sorption sites [10]. The model has been used successfully to describe water sorption in carbohydratemaltose [10] and carbohydrate-glycerol [3] mixtures in the glassy state. Freundlich’s equation is defined as: ͳ ܯൌ ݎܨܭሺ ݓܣሻ ൗܿ (3) where M the moisture content (%, db) and Aw water activity or relative humidity at thermodynamic equilibrium. KFr and c (c 1) are temperature-dependent constants representative of the system. BET and
269
Paulo Díaz et al. / Procedia Food Science 1 (2011) 267 – 274
Freundlich equation were applied up to the data at 44%RH, as has been suggested by Ubbink et al. [4] and Roussenova et al. [3] 2.3. Glass Transition Temperature The glass transition temperature (Tg) of the samples was determined by Differential Scanning Calorimetry (Diamond DSC, Perkin Elmer, USA). Prior measurements the DSC was calibrated using indium as standard. ~20 mg of sample were hermetically sealed in aluminum pans of 30 ȝl and subjected to the following thermal profile: cooling from 20 to -40ºC at 40ºC/min, heating from -40 to 120ºC at 10ºC/min, cooling from 120 to -40ºC at 40ºC/min and heating from -40 to 120ºC at 10ºC/min. An empty pan was used as reference. Tg values were calculated from second scan and defined as the midpoint in the change in heat capacity. ܶ݃ ൌ
σ݊݅ൌͳ ܹ݅ ο݅݃ܶ ݅ܥ σ݊݅ൌͳ ܹ݅ ο݅ܥ
(4)
All measurements were performed in triplicate. Tg data were fitted by Couchman-Karasz Equation as indicated in the literature for this type of systems [11] which is defined as: where Wi is the weight fraction of component i, ǻCpi is the change in specific heat capacity of the component i between the glassy and rubbery state and Tgi is the glass transition temperature of component i. The fitting considered Tg values of glycerol (-85.6ºC) and water (-135ºC) reported in the literature [12]. The Tg value used for anhydrous bovine gelatin (194ºC) has been previously determined by the authors (unpublished data). ǻCp values were used as fitting variable, which were later compared to the values obtained experimentally by DSC. 2.4. Specific Volume The specific volume (Vs) of the films was measured by gas pycnometry (Multivolume 1303 pycnometer, Micromertics, USA). Nitrogen was used as a gas carrier at a pressure of 140-170 kPa. Vs was determined from the pressure difference between measurements. Each sample was deposited in an aluminum insert of 5 cm3 and eight gas purges were performed before each measurement. The Vs reported corresponds to the mean of six measurements. 2.5. Mathematical Fitting The fitting of all equations described above was performed by minimization of the quadratic difference between the experimental and predicted values using the Solver package in Excel (Office 2007, Microsoft Corp.). The fitting error was evaluated by the Mean Relative Error (MRE) using the following relation: ݊
หܺ݁݅ ܺ ݅ห ͳͲͲ ܧܴܯൌ ݊ ܺ݁݅
(5)
݅ൌͳ
where Xei is the experimental value, Xpi is the predicted value and n is the number of experimental data. MRE values 10% were considered as a good fit [13]. 3. Results & Discussion 3.1. Sorption Isotherms The sorption isotherms for each gelatin-glycerol sample are presented in Figure 1. All curves showed a sigmoid behaviour associated to the well known type II isotherm typical for biopolymers [14]. Interestingly up to 55% ERH an increase in glycerol content generates a decrease in moisture content in the films, while at higher ERH the sorption behaviour is completely the opposite. Moreover, for the
270
Paulo Díaz et al. / Procedia Food Science 1 (2011) 267 – 274
samples at the higher glycerol content, the characteristic sigmoidal water sorption behaviour related to the transition from the glassy to the rubbery state is observed at lower ERH, as stated previously in the literature for starch-glycerol, maltopolymer-maltose and maltopolymer-glycerol systems by Myllärinen et al. [15], Enrione et al. [2], Ubbink et al. [4] and Roussenova et al. [3] respectively. Ubbink [10] discussed that glycerol is able to position itself between polymeric chains, occupying part of the available (free) volume, preventing water molecules from the interaction with the matrix at low relative humidities. Myllärinen et al. [15] and Enrione et al. [2] have suggested that at low ERH, water would be displaced from the biopolymer active polar group by glycerol. The trihydric alcohol structure of the glycerol and the fact that at low vapour pressure hydrogen bonding is the main force involved in sorption mechanism in these materials [2], could explain these results. However Myllärinen et al. [15] and Enrione et al. [2] fount that the change in sorption behaviour in starch systems occur at ERH higher than showed in this work (~65-70% ERH). 90 15
g water/100g dry sample
80
g water/100g dry sample
70 60 50 40
10
5
0 0.0
30
0.1
0.2
0.3
0.4
0.5
0.6
Equilibrium Relative Humidity /100
20 10 0 0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Equilibrium Relative Humidity /100 0% Glycerol
2% Glycerol
6% Glycerol
10% Glycerol
Fig. 1. Sorption isotherm at 20ºC for mixtures of bovine gelatin-glycerol. The inset shows isotherm of samples in the glassy state
At ERH > 55%, glycerol facilitates the sorption of water. At these high water vapour pressures there is an increase in the molecular mobility by the plasticizing effect of the polyol and water [2]. Also, the hygroscopic behaviour of glycerol promotes the absorption of water [2, 12]. It is possible that at high ERH glycerol molecules move more freely absorbing moisture from the surroundings [9]. Table 1. GAB, BET and Freundlich parameters describing water sorption in gelatin glycerol films. Glycerol Content (%, db)
Equation Parameter 0
2
6
10
KGAB
0.67
0.85
0.84
0.89
CGAB
15.5
16.9
12.4
11.2
m0-GAB (%, db)
10.3
8.7
8.4
7.8
MRE (%)
2.5
2.0
4.4
4.9
CBET
14.9
14.4
12.7
10.7
m0-BET (%, db)
9.2
9.0
8.3
8.1
MRE (%)
3.0
2.2
1.1
1.4
KFr
0.21
0.21
0.19
0.17
CFr
1.7
1.7
1.7
1.8
*MRE (%)
1.3
1.8
2.1
1.5
*MRE: Mean Relative Error
Paulo Díaz et al. / Procedia Food Science 1 (2011) 267 – 274
The GAB parameters obtained for each curve are shown in Table 1. m0-GAB decreased with increasing glycerol content from ~10.3% (0% GC) to ~7.8% (10% GC), while CGAB parameter decreased from ~15.5 to ~11.2 indicating lower energy levels associated to water matrix interactions when the glycerol concentration increased. The KGAB values decrease when the glycerol contents increase, suggesting a shift in sorption energies to the multilayer domain. Similar behaviour was reported by Enrione et al. [2] in extruded starches from different origins (wheat, waxy maize and rice) in presence of the same plasticiser. In the case of BET parameters, mo-BET decreased from ~9.2% to ~8.1% with increasing glycerol content, suggesting a reduction in sorption sites availability, which was correlated to the reduction in CBET from ~14.9 to ~10.7. There is no clear explanation for differences in the values of m0 obtained from BET and GAB models, it is possible that the extra parameter K may affect numerically the value of m0 during the equation fitting [2, 9]. Nevertheless, literature suggest that the values given by GAB model are closer to the theoretically expected values for the monolayer [16]. This difference might be related with restrictions of these models to explain successfully the complex phenomena of water sorption. Although both models fitted well at ERH < 80% the physical meaning related their parameters should be interpreted with cautious. Indeed these equations were originally developed to describe the sorption phenomena at the surface of materials. For instance, fundamental assumptions such as all sorption sites are identical, no interaction between sorbed molecules [9] do not entirely apply to hydrophilic biopolymers. Freundlich KFr value decreased as the polyol concentration increased from ~0.21 to ~0.17 (Table I). Similar behaviour has been reported by Ubbink et al. [4] and more recently by Roussenova et al. [3]. They related this behaviour with the reduction of available sites for the binding of water by the effect of low molecular weights compounds on the microstructure (molecular packing) of the system (i.e. glycerol, maltose). The values of the constant c of the Freundlich equation was independent of the matrix composition, as in agreement with various studies [3, 4]. 3.2. Glass Transition Temperature Figure 2 shows the well known plasticizing effect of water and glycerol on the Tg in the polymer, which is explained by the ability of low molecular weight molecules (water and glycerol in this case) to disrupt hydrogen bonds between polymeric chains [17-19]. Figure 2 clearly shows that all samples under ~15%MC (68% ERH) were in the glassy state at the sorption equilibrating temperature (dotted line indicates a temperature of ~20ºC). Couchman-Karasz parameters (Equation 4) were similar to those presented in the literature [12] (Table II). Also, the fitted ǻCp values were similar to the ǻCp obtained experimentally by DSC. Experimental ǻCp gave values close to 0.36 J/g C.
Glass Transition Temperature (ºC)
230 0% Glycerol 2% Glycerol 6% Glycerol 10% Glycerol
180
130
Rubbery State 80
30
Glassy State -20 0
3
6
9
12
15
18
21
Moisture Content, wb (%)
Fig. 2. Tg of gelatin-glycerol films. Solid line corresponds to the modelling values by Couchman-Karasz equation. Dotted line represents the temperature of sorption equilibration and pycnometer measurements (20ºC)
271
272
Paulo Díaz et al. / Procedia Food Science 1 (2011) 267 – 274
3.3. Specific Volume The specific volume (Vs) of the films with different moisture contents is presented in Figure 3. Results showed that the higher amount of glycerol, lower the specific volume of the samples, which implies a densification phenomena by the addition of this polyol. This data correlates well with Table 2. Couchman-Karasz parameters for gelatin findings by Ubbink [10] and Roussenova et al. [3] glycerol mixtures. for a maltopolymer based matrix in the glassy state. Indeed, they observed a positive correlation ǻCp (J/g C) *MRE Glycerol (%) (% w/w) between the sorption of water and the specific Gelatin Glycerol Water volume of glassy carbohydrates. Thus, the higher 0 0.35 0.88 1.94 6.4 amount of a low molecular weight compound (plasticizer), lower the specific volume of 2 0.33 0.92 1.94 4.8 carbohydrate matrix and lower the equilibrium 6 0.38 1.56 1.94 5.7 moisture content of the sample. They explained this behaviour by a dependency 10 0.35 1.12 1.94 9.7 of sorption on the microstructure of the *MRE: Mean Relative Error carbohydrate matrix. The denser the matrix, fewer spaces are available to accommodate water molecules, hence less water can be absorbed at a defined water activity [3]. This behaviour is also followed by the direct relationship between KFr values and glycerol content (data not shown). Interestingly other studies at molecular level scale have shown a direct relationship between hole volume (a measured of intermolecular spaces assessed by Positron Annihilation Spectroscopy, PALS) and KFr in mixtures carbohydrate-maltose [4, 6].
Specific Volume (cm3/g)
1.8
0% Glycerol
1.6
2% Glycerol
1.4
6% Glycerol 10% Glycerol
1.2 1.0 0.8 0.6 0.4 0
5
10
15
20
Moisture Content (%, db) Fig. 3. Specific volume of bovine gelatin-glycerol films at different moisture and glycerol content
The effect of water content on specific volume is rather complex (Figure 3). Vs value is different depending on the moisture content range. For MC < ~10%, specific volume increase when MC increased but at MC > ~10% an opposite behaviour is obtained indicating a densification effect. This data shows some differences to those presented in the literature for carbohydrate systems, where an antiplastification phenomena (decrease in Vs with increasing water content) in maltopolymer systems is observed when plasticized by maltose [6] or glycerol [3] at very low moisture content. Our results are showing a complex protein-glycerol-water interaction, where interestingly a maximum Vs value was observed near m0 GAB and m0 BET moisture contents. Up to ~10%MC water plays a role of plasticizer increasing Vs, however above ~10% the decrease in Vs may be explained by local organization of the constituent molecules [3].
Paulo Díaz et al. / Procedia Food Science 1 (2011) 267 – 274
Further work is required to clarify this behavior in terms of structure organization at macro and micro level domain. 4. Conclusions Glycerol modified the sorption properties of bovine gelatin films, reducing the water content at low relative humidities and increasing it at high relative humidities suggesting complex polyol-matrix interactions. GAB, BET and Freundlich parameters shows that gelatin would interact preferably with glycerol at low ERH as indicated by reductions m0GAB, m0BET, CGAB and CBET and KFr values with increasing polyol content. The decrease in these values is in agreement with measurement of specific volume showing a densification phenomena (decrease in Vs) when the glycerol content increased, phenomenon which has also been detected in carbohydrates.
Acknowledgements The authors acknowledge the financial support from INNOVA CHILE CORFO CT11 PUT-20 and FONDECYT Grant Nº1110607.
References [1] Karim AA, Bhat R. Gelatin alternatives for the food industry: recent developments, challenges and prospects. Trends Food Sci Tech 2008, 19: 644-56. [2] Enrione JI, Hill SE, Mitchell JR. Sorption Behavior of Mixtures of Glycerol and Starch. J Agric Food Chem 2007, 55: 2956-63. [3] Roussenova M, Murith M, Alam A, Ubbink J. Plasticization, Antiplasticization, and Molecular Packing in Amorphous Carbohydrate-Glycerol Matrices. Biomacromolecules 2010, 11: 3237-47. [4] Ubbink J, Giardiello M-I, Limbach H-Jr. Sorption of Water by Bidisperse Mixtures of Carbohydrates in Glassy and Rubbery States. Biomacromolecules 2007, 8: 2862-73. [5] Townrow S, Kilburn D, Alam A, Ubbink J. Molecular Packing in Amorphous Carbohydrate Matrixes, J Phys Chem B 2007, 111: 12643-48. [6] Townrow S, Roussenova M, Giardiello M-I, Alam A, Ubbink J. Specific Volumen-Hole Volume Correlations in Amorphous Carbohydrates: Effect of Temperature, Molecular Weight, and Water Content. J Phys Chem B 2010. 114: 1568-78. [7] Brunauer S, Emmet PH, Teller E. Adsorption of Gases in Multimolecular Layers. J Am Chem Soc 1938, 60: 309-19. [8] Anderson RB. Modifications of the Brunauer, Emmett and Teller Equation. J Am Chem Soc 1946, 68: 686-91. [9] Enrione J. Mechanical Stability of Intermediate Moisture Starch-Glycerol Systems. PhD Thesis, Division of Food Sciences, School of Bioscience, The University of Nottingham, UK. 2005. [10] Ubbick J. Structural Advances in the Understanding of Carbohydrate Glasses. In: Kasapis S, Norton I, Ubbink J, editors. Modern Biopolymer Science. Bridging the Divide between Fundamental Treatise and Industrial Application. London, Academic Press, 2009, p. 277-93. [11] Couchman PR, Karasz FE. A Classical Thermodynamic Discussion of the Effect of Composition on Glass-Transition Temperatures. Macromolecules 1978, 11: 117-19. [12] Rivero S, García MA, Pinotti A. Correlations between structural, barrier, thermal and mechanical properties of plasticized gelatin films. Innov Food Sci Emerg 2009, 11: 369-75. [13] Yan Z, Sousa-Gallagher MJ, Oliveira FAR. Mathematical modeling of the kinetic of quality deterioration of intermediate moisture content banana during storage. J Food Eng 2008, 84: 359-67. [14] Basu S, Shivhare US, Mujumdar AS. Models for Sorption Isotherms for Foods: A Review. Dry Technol 2006, 24: 917 - 30. [15] Myllärinen P, Partanen R, Jukka S, Forssell P. Effect of glycerol on behaviour of amylose and amylopectin films. Carbohyd Polym 2002, 50: 355-61.
273
274
Paulo Díaz et al. / Procedia Food Science 1 (2011) 267 – 274 [16] Van den Berg C. Developement of BET-like models for sorption of water on foods, theory and relevance. In: Simatos D, Multon JL. Properties of Water in Foods in Relation to Quality and Stability (NATO Science Series E:). New York, Springer, 1985, p. 119-31. [17] Arvanitoyannis I. Formation and Properties of Collagen and Gelatin Films and Coatings. In: Gennadios A, editor. ProteinBased Films and Coatings. Boca Raton, CRC Press, 2002, p. 275-304. [18] Thomazine M, Carvalho RA, Sobral PJA. Physical Properties of Gelatin Films Plasticized by Blends of Glycerol and Sorbitol. J Food Sci 2005, 70: E172-E176. [19] Roos Y, The Glassy State. In: Aguilera JM, Lillford PJ, editors. Food Material Science, Principle and Applications. New York, Springer, 2008, pp. 67-82.
Presented at ICEF11 (May 22-26, 2011 – Athens, Greece) as paper FMS1144.
Procedia Food Science 1 (2011) 267 – 274
11th International Congress of Engineering and Foods (ICEF11)
Effect of glycerol on water sorption of bovine gelatin films in the glassy state Paulo Paolo Díaz, Carol Arratia, Corina Vásquez, Fernando Osorio, Javier Enrionea* a
Departamento de Ciencias y Tecnología de Alimentos, Universidad de Santiago de Chile. Obispo Umaña 050, Estación Central, Santiago, Chile.
Abstract The addition of low molecular weight compounds to a biopolymer in the glassy state can generate changes in macrostructural properties of the matrix. This work studied the effect of glycerol on water sorption properties of gelatin films in the glassy state. Bovine gelatin-glycerol (0-10% dry basis, db) films were equilibrated to moisture contents (MC) in the range of 5 - 30% (db). Results show that the addition of glycerol generated differences in sorption behaviour by the decreasing of sorption parameters (GAB, BET and Freundlich), indicating complex glycerol-matrix interactions. Gas pycnometry showed a densification of the matrix suggesting a reduction in sorption site availability when glycerol content increased. © 2011 Published by Elsevier B.V. Open access under CC BY-NC-ND license. Selection and/or peer-review under responsibility of 11th International Congress on Engineering and Food (ICEF 11) Executive Committee. Keywords: gelatin; glycerol; water sorption; specific volume; glassy state.
1. Introduction Among hydrocolloids, gelatin is widely used in food and pharmaceutical industries due to its various functional properties. Gelatin is obtained from denaturation of collagen by acidic-alkaline treatments. Collagen is the structural unit of skin, hair and bones of animals [1]. One of its most important properties is its ability to form thermoreversible gels, however new uses are related with its capacity to form edible films or as encapsulating material for certain nutrients and drugs [1]. The addition of low molecular weight compounds, i.e. glycerol, to low moisture content gelatin matrix improves its mechanical properties generating a more flexible structure. However it can alter the mechanism of polymer-water interactions, thus changing the water sorption properties of the material. Literature suggests that this effect can be different if the matrix is in the glassy or rubbery state [2-4]. _____________ * Corresponding author. Tel.: 56-2-7184524; fax: 56-2-7764089. E-mail address: [email protected]
2211 601X © 2011 Published by Elsevier B.V. Open access under CC BY-NC-ND license. Selection and/or peer-review under responsibility of 11th International Congress on Engineering and Food (ICEF 11) Executive Committee. doi:10.1016/j.profoo.2011.09.042
268
Paulo Díaz et al. / Procedia Food Science 1 (2011) 267 – 274
Also it has been reported that the presence of plasticizer can produce changes in density of a carbohydrate based matrix [5, 6]. Little information is available looking at the changes in sorption properties and densification phenomena in protein based systems, such as gelatin. The aim of this work was to study the effect of glycerol on water sorption properties of bovine gelatin films in the glassy state and to assess variations in its macrostructure by specific volume determinations. 2. Materials & Methods 2.1. Film Preparation A 7% (w/v) bovine gelatin (type B, Bloom 220, Rousselot, Brazil) suspension at 50ºC was mixed with different glycerol (purity > 99.5%, Merck, Germany) contents (GC) of 0, 2, 6 and 10% (dry basis, db) and then deposited onto tefloned containers and maintained at 7 ± 1ºC for seven days until translucent films (~0.3 mm in thickness) were obtained. Films were maintained hermetically at ~0% Equilibrium Relative Humidity (ERH) under P2O5 for one week. The films were then equilibrated isothermally at ~20ºC under 0, 11, 22, 33, 44, 68, 75 and 85% ERH using saturated salt solutions. Thymol was used as antimicrobial agent for the high ERH values. Equilibrium was followed gravimetrically until the difference between two measurements was lower than 0.5%. The equilibrium stage generated moisture contents (MC) in the range of 5 - 30% (db). 2.2 Sorption Isotherms The water sorption isotherms were fitted by the mathematical relations of Brunauer, Emmet and Teller [7] (BET), Guggenheim, Anderson and de Boer [8] (GAB) and Freundlich [4]. BET Equation is defined as: ݉Ͳ ݓܣ ܶܧܤܥ (1) ܯൌ ሺͳ െ ݓܣሻሺͳ െ ݓܣ ݓܣ ܶܧܤܥሻ
where M the moisture content (%, db) and m0 is the moisture content at the monolayer (%, db). CBET is a temperature-dependent constant (related to the net heat of sorption associated to the monolayer) and Aw water activity [2] or relative humidity at thermodynamic equilibrium. GAB Equation is defined as: ݉Ͳ ݓܣܭ ܤܣܩܥ (2) ܯൌ ሺͳ െ ݓܣሻሺͳ െ ݓܣܭ ݓܣܭ ܤܣܩܥሻ
where M the moisture content (%, db) and m0 is the monolayer value (%, db). T CGAB is related to the energy associated with the interaction between water molecules and the matrix primary interaction sites or monolayer [2]. KGAB is also a temperature-dependent parameter related to the heat of sorption at the multilayer. BET equation is a generalization of the Langmuir model for the sorption of gas onto a solid surface through multilayers [7], but assuming that evaporation-condensation properties of the molecules of the second and higher adsorbed layers are the same as of bulk water [9]. Meanwhile GAB equation was derived from BET equation but assuming that the state of the sorbent molecule in the second and higher layers are equal but different from the liquid-like state, introducing an additional degree of freedom (parameter K), which gives to the GAB model its fitting versatility [9]. Freundlich equation has also been proposed to describe sorption in glassy materials with an energetically heterogeneous distribution of sorption sites [10]. The model has been used successfully to describe water sorption in carbohydratemaltose [10] and carbohydrate-glycerol [3] mixtures in the glassy state. Freundlich’s equation is defined as: ͳ ܯൌ ݎܨܭሺ ݓܣሻ ൗܿ (3) where M the moisture content (%, db) and Aw water activity or relative humidity at thermodynamic equilibrium. KFr and c (c 1) are temperature-dependent constants representative of the system. BET and
269
Paulo Díaz et al. / Procedia Food Science 1 (2011) 267 – 274
Freundlich equation were applied up to the data at 44%RH, as has been suggested by Ubbink et al. [4] and Roussenova et al. [3] 2.3. Glass Transition Temperature The glass transition temperature (Tg) of the samples was determined by Differential Scanning Calorimetry (Diamond DSC, Perkin Elmer, USA). Prior measurements the DSC was calibrated using indium as standard. ~20 mg of sample were hermetically sealed in aluminum pans of 30 ȝl and subjected to the following thermal profile: cooling from 20 to -40ºC at 40ºC/min, heating from -40 to 120ºC at 10ºC/min, cooling from 120 to -40ºC at 40ºC/min and heating from -40 to 120ºC at 10ºC/min. An empty pan was used as reference. Tg values were calculated from second scan and defined as the midpoint in the change in heat capacity. ܶ݃ ൌ
σ݊݅ൌͳ ܹ݅ ο݅݃ܶ ݅ܥ σ݊݅ൌͳ ܹ݅ ο݅ܥ
(4)
All measurements were performed in triplicate. Tg data were fitted by Couchman-Karasz Equation as indicated in the literature for this type of systems [11] which is defined as: where Wi is the weight fraction of component i, ǻCpi is the change in specific heat capacity of the component i between the glassy and rubbery state and Tgi is the glass transition temperature of component i. The fitting considered Tg values of glycerol (-85.6ºC) and water (-135ºC) reported in the literature [12]. The Tg value used for anhydrous bovine gelatin (194ºC) has been previously determined by the authors (unpublished data). ǻCp values were used as fitting variable, which were later compared to the values obtained experimentally by DSC. 2.4. Specific Volume The specific volume (Vs) of the films was measured by gas pycnometry (Multivolume 1303 pycnometer, Micromertics, USA). Nitrogen was used as a gas carrier at a pressure of 140-170 kPa. Vs was determined from the pressure difference between measurements. Each sample was deposited in an aluminum insert of 5 cm3 and eight gas purges were performed before each measurement. The Vs reported corresponds to the mean of six measurements. 2.5. Mathematical Fitting The fitting of all equations described above was performed by minimization of the quadratic difference between the experimental and predicted values using the Solver package in Excel (Office 2007, Microsoft Corp.). The fitting error was evaluated by the Mean Relative Error (MRE) using the following relation: ݊
หܺ݁݅ ܺ ݅ห ͳͲͲ ܧܴܯൌ ݊ ܺ݁݅
(5)
݅ൌͳ
where Xei is the experimental value, Xpi is the predicted value and n is the number of experimental data. MRE values 10% were considered as a good fit [13]. 3. Results & Discussion 3.1. Sorption Isotherms The sorption isotherms for each gelatin-glycerol sample are presented in Figure 1. All curves showed a sigmoid behaviour associated to the well known type II isotherm typical for biopolymers [14]. Interestingly up to 55% ERH an increase in glycerol content generates a decrease in moisture content in the films, while at higher ERH the sorption behaviour is completely the opposite. Moreover, for the
270
Paulo Díaz et al. / Procedia Food Science 1 (2011) 267 – 274
samples at the higher glycerol content, the characteristic sigmoidal water sorption behaviour related to the transition from the glassy to the rubbery state is observed at lower ERH, as stated previously in the literature for starch-glycerol, maltopolymer-maltose and maltopolymer-glycerol systems by Myllärinen et al. [15], Enrione et al. [2], Ubbink et al. [4] and Roussenova et al. [3] respectively. Ubbink [10] discussed that glycerol is able to position itself between polymeric chains, occupying part of the available (free) volume, preventing water molecules from the interaction with the matrix at low relative humidities. Myllärinen et al. [15] and Enrione et al. [2] have suggested that at low ERH, water would be displaced from the biopolymer active polar group by glycerol. The trihydric alcohol structure of the glycerol and the fact that at low vapour pressure hydrogen bonding is the main force involved in sorption mechanism in these materials [2], could explain these results. However Myllärinen et al. [15] and Enrione et al. [2] fount that the change in sorption behaviour in starch systems occur at ERH higher than showed in this work (~65-70% ERH). 90 15
g water/100g dry sample
80
g water/100g dry sample
70 60 50 40
10
5
0 0.0
30
0.1
0.2
0.3
0.4
0.5
0.6
Equilibrium Relative Humidity /100
20 10 0 0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Equilibrium Relative Humidity /100 0% Glycerol
2% Glycerol
6% Glycerol
10% Glycerol
Fig. 1. Sorption isotherm at 20ºC for mixtures of bovine gelatin-glycerol. The inset shows isotherm of samples in the glassy state
At ERH > 55%, glycerol facilitates the sorption of water. At these high water vapour pressures there is an increase in the molecular mobility by the plasticizing effect of the polyol and water [2]. Also, the hygroscopic behaviour of glycerol promotes the absorption of water [2, 12]. It is possible that at high ERH glycerol molecules move more freely absorbing moisture from the surroundings [9]. Table 1. GAB, BET and Freundlich parameters describing water sorption in gelatin glycerol films. Glycerol Content (%, db)
Equation Parameter 0
2
6
10
KGAB
0.67
0.85
0.84
0.89
CGAB
15.5
16.9
12.4
11.2
m0-GAB (%, db)
10.3
8.7
8.4
7.8
MRE (%)
2.5
2.0
4.4
4.9
CBET
14.9
14.4
12.7
10.7
m0-BET (%, db)
9.2
9.0
8.3
8.1
MRE (%)
3.0
2.2
1.1
1.4
KFr
0.21
0.21
0.19
0.17
CFr
1.7
1.7
1.7
1.8
*MRE (%)
1.3
1.8
2.1
1.5
*MRE: Mean Relative Error
Paulo Díaz et al. / Procedia Food Science 1 (2011) 267 – 274
The GAB parameters obtained for each curve are shown in Table 1. m0-GAB decreased with increasing glycerol content from ~10.3% (0% GC) to ~7.8% (10% GC), while CGAB parameter decreased from ~15.5 to ~11.2 indicating lower energy levels associated to water matrix interactions when the glycerol concentration increased. The KGAB values decrease when the glycerol contents increase, suggesting a shift in sorption energies to the multilayer domain. Similar behaviour was reported by Enrione et al. [2] in extruded starches from different origins (wheat, waxy maize and rice) in presence of the same plasticiser. In the case of BET parameters, mo-BET decreased from ~9.2% to ~8.1% with increasing glycerol content, suggesting a reduction in sorption sites availability, which was correlated to the reduction in CBET from ~14.9 to ~10.7. There is no clear explanation for differences in the values of m0 obtained from BET and GAB models, it is possible that the extra parameter K may affect numerically the value of m0 during the equation fitting [2, 9]. Nevertheless, literature suggest that the values given by GAB model are closer to the theoretically expected values for the monolayer [16]. This difference might be related with restrictions of these models to explain successfully the complex phenomena of water sorption. Although both models fitted well at ERH < 80% the physical meaning related their parameters should be interpreted with cautious. Indeed these equations were originally developed to describe the sorption phenomena at the surface of materials. For instance, fundamental assumptions such as all sorption sites are identical, no interaction between sorbed molecules [9] do not entirely apply to hydrophilic biopolymers. Freundlich KFr value decreased as the polyol concentration increased from ~0.21 to ~0.17 (Table I). Similar behaviour has been reported by Ubbink et al. [4] and more recently by Roussenova et al. [3]. They related this behaviour with the reduction of available sites for the binding of water by the effect of low molecular weights compounds on the microstructure (molecular packing) of the system (i.e. glycerol, maltose). The values of the constant c of the Freundlich equation was independent of the matrix composition, as in agreement with various studies [3, 4]. 3.2. Glass Transition Temperature Figure 2 shows the well known plasticizing effect of water and glycerol on the Tg in the polymer, which is explained by the ability of low molecular weight molecules (water and glycerol in this case) to disrupt hydrogen bonds between polymeric chains [17-19]. Figure 2 clearly shows that all samples under ~15%MC (68% ERH) were in the glassy state at the sorption equilibrating temperature (dotted line indicates a temperature of ~20ºC). Couchman-Karasz parameters (Equation 4) were similar to those presented in the literature [12] (Table II). Also, the fitted ǻCp values were similar to the ǻCp obtained experimentally by DSC. Experimental ǻCp gave values close to 0.36 J/g C.
Glass Transition Temperature (ºC)
230 0% Glycerol 2% Glycerol 6% Glycerol 10% Glycerol
180
130
Rubbery State 80
30
Glassy State -20 0
3
6
9
12
15
18
21
Moisture Content, wb (%)
Fig. 2. Tg of gelatin-glycerol films. Solid line corresponds to the modelling values by Couchman-Karasz equation. Dotted line represents the temperature of sorption equilibration and pycnometer measurements (20ºC)
271
272
Paulo Díaz et al. / Procedia Food Science 1 (2011) 267 – 274
3.3. Specific Volume The specific volume (Vs) of the films with different moisture contents is presented in Figure 3. Results showed that the higher amount of glycerol, lower the specific volume of the samples, which implies a densification phenomena by the addition of this polyol. This data correlates well with Table 2. Couchman-Karasz parameters for gelatin findings by Ubbink [10] and Roussenova et al. [3] glycerol mixtures. for a maltopolymer based matrix in the glassy state. Indeed, they observed a positive correlation ǻCp (J/g C) *MRE Glycerol (%) (% w/w) between the sorption of water and the specific Gelatin Glycerol Water volume of glassy carbohydrates. Thus, the higher 0 0.35 0.88 1.94 6.4 amount of a low molecular weight compound (plasticizer), lower the specific volume of 2 0.33 0.92 1.94 4.8 carbohydrate matrix and lower the equilibrium 6 0.38 1.56 1.94 5.7 moisture content of the sample. They explained this behaviour by a dependency 10 0.35 1.12 1.94 9.7 of sorption on the microstructure of the *MRE: Mean Relative Error carbohydrate matrix. The denser the matrix, fewer spaces are available to accommodate water molecules, hence less water can be absorbed at a defined water activity [3]. This behaviour is also followed by the direct relationship between KFr values and glycerol content (data not shown). Interestingly other studies at molecular level scale have shown a direct relationship between hole volume (a measured of intermolecular spaces assessed by Positron Annihilation Spectroscopy, PALS) and KFr in mixtures carbohydrate-maltose [4, 6].
Specific Volume (cm3/g)
1.8
0% Glycerol
1.6
2% Glycerol
1.4
6% Glycerol 10% Glycerol
1.2 1.0 0.8 0.6 0.4 0
5
10
15
20
Moisture Content (%, db) Fig. 3. Specific volume of bovine gelatin-glycerol films at different moisture and glycerol content
The effect of water content on specific volume is rather complex (Figure 3). Vs value is different depending on the moisture content range. For MC < ~10%, specific volume increase when MC increased but at MC > ~10% an opposite behaviour is obtained indicating a densification effect. This data shows some differences to those presented in the literature for carbohydrate systems, where an antiplastification phenomena (decrease in Vs with increasing water content) in maltopolymer systems is observed when plasticized by maltose [6] or glycerol [3] at very low moisture content. Our results are showing a complex protein-glycerol-water interaction, where interestingly a maximum Vs value was observed near m0 GAB and m0 BET moisture contents. Up to ~10%MC water plays a role of plasticizer increasing Vs, however above ~10% the decrease in Vs may be explained by local organization of the constituent molecules [3].
Paulo Díaz et al. / Procedia Food Science 1 (2011) 267 – 274
Further work is required to clarify this behavior in terms of structure organization at macro and micro level domain. 4. Conclusions Glycerol modified the sorption properties of bovine gelatin films, reducing the water content at low relative humidities and increasing it at high relative humidities suggesting complex polyol-matrix interactions. GAB, BET and Freundlich parameters shows that gelatin would interact preferably with glycerol at low ERH as indicated by reductions m0GAB, m0BET, CGAB and CBET and KFr values with increasing polyol content. The decrease in these values is in agreement with measurement of specific volume showing a densification phenomena (decrease in Vs) when the glycerol content increased, phenomenon which has also been detected in carbohydrates.
Acknowledgements The authors acknowledge the financial support from INNOVA CHILE CORFO CT11 PUT-20 and FONDECYT Grant Nº1110607.
References [1] Karim AA, Bhat R. Gelatin alternatives for the food industry: recent developments, challenges and prospects. Trends Food Sci Tech 2008, 19: 644-56. [2] Enrione JI, Hill SE, Mitchell JR. Sorption Behavior of Mixtures of Glycerol and Starch. J Agric Food Chem 2007, 55: 2956-63. [3] Roussenova M, Murith M, Alam A, Ubbink J. Plasticization, Antiplasticization, and Molecular Packing in Amorphous Carbohydrate-Glycerol Matrices. Biomacromolecules 2010, 11: 3237-47. [4] Ubbink J, Giardiello M-I, Limbach H-Jr. Sorption of Water by Bidisperse Mixtures of Carbohydrates in Glassy and Rubbery States. Biomacromolecules 2007, 8: 2862-73. [5] Townrow S, Kilburn D, Alam A, Ubbink J. Molecular Packing in Amorphous Carbohydrate Matrixes, J Phys Chem B 2007, 111: 12643-48. [6] Townrow S, Roussenova M, Giardiello M-I, Alam A, Ubbink J. Specific Volumen-Hole Volume Correlations in Amorphous Carbohydrates: Effect of Temperature, Molecular Weight, and Water Content. J Phys Chem B 2010. 114: 1568-78. [7] Brunauer S, Emmet PH, Teller E. Adsorption of Gases in Multimolecular Layers. J Am Chem Soc 1938, 60: 309-19. [8] Anderson RB. Modifications of the Brunauer, Emmett and Teller Equation. J Am Chem Soc 1946, 68: 686-91. [9] Enrione J. Mechanical Stability of Intermediate Moisture Starch-Glycerol Systems. PhD Thesis, Division of Food Sciences, School of Bioscience, The University of Nottingham, UK. 2005. [10] Ubbick J. Structural Advances in the Understanding of Carbohydrate Glasses. In: Kasapis S, Norton I, Ubbink J, editors. Modern Biopolymer Science. Bridging the Divide between Fundamental Treatise and Industrial Application. London, Academic Press, 2009, p. 277-93. [11] Couchman PR, Karasz FE. A Classical Thermodynamic Discussion of the Effect of Composition on Glass-Transition Temperatures. Macromolecules 1978, 11: 117-19. [12] Rivero S, García MA, Pinotti A. Correlations between structural, barrier, thermal and mechanical properties of plasticized gelatin films. Innov Food Sci Emerg 2009, 11: 369-75. [13] Yan Z, Sousa-Gallagher MJ, Oliveira FAR. Mathematical modeling of the kinetic of quality deterioration of intermediate moisture content banana during storage. J Food Eng 2008, 84: 359-67. [14] Basu S, Shivhare US, Mujumdar AS. Models for Sorption Isotherms for Foods: A Review. Dry Technol 2006, 24: 917 - 30. [15] Myllärinen P, Partanen R, Jukka S, Forssell P. Effect of glycerol on behaviour of amylose and amylopectin films. Carbohyd Polym 2002, 50: 355-61.
273
274
Paulo Díaz et al. / Procedia Food Science 1 (2011) 267 – 274 [16] Van den Berg C. Developement of BET-like models for sorption of water on foods, theory and relevance. In: Simatos D, Multon JL. Properties of Water in Foods in Relation to Quality and Stability (NATO Science Series E:). New York, Springer, 1985, p. 119-31. [17] Arvanitoyannis I. Formation and Properties of Collagen and Gelatin Films and Coatings. In: Gennadios A, editor. ProteinBased Films and Coatings. Boca Raton, CRC Press, 2002, p. 275-304. [18] Thomazine M, Carvalho RA, Sobral PJA. Physical Properties of Gelatin Films Plasticized by Blends of Glycerol and Sorbitol. J Food Sci 2005, 70: E172-E176. [19] Roos Y, The Glassy State. In: Aguilera JM, Lillford PJ, editors. Food Material Science, Principle and Applications. New York, Springer, 2008, pp. 67-82.
Presented at ICEF11 (May 22-26, 2011 – Athens, Greece) as paper FMS1144.