Swelling Pressure of Tapioca Starch Gel Estimated from Distribution Coefficients of Non-electrolytes

Abstract

The distribution coefficients, K_app, of non-electrolytes of different molecular masses on a tapioca starch gel were measured at a temperature range of 25 – 60°C. The swelling pressure of the gel was estimated from the coefficients at each temperature, and was observed to increase from 1 to 3 MPa with a temperature increase from 25 to 50°C. The swelling pressure, however, decreased at 60°C. The decrease was ascribed to the gelatinization of tapioca starch, which was observed using differential scanning calorimetry. The temperature dependency of the swelling pressure suggested a significant effect of gelatinization on the distribution equilibrium of a solute on the starch gel. The experimentally observed K_app values of electrolytes (0.5 to 0.8) were lower than the values calculated using the swelling pressure and specific molecular volumes of the electrolytes. This suggested the involvement of another factor, such as an electrostatic interaction, in the distribution of electrolytes on the gel.

Introduction

Many food items contain polymers, which form a gel-like matrix. Some foods are even regarded as gels themselves. For example, rice cakes, boiled eggs, sausages, and konjac jelly are gels respectively prepared from starch, heat-induced egg protein (Hatta, 1986; Doi, 1993; Handa et al., 1998), meat proteins cross-linked by transglutaminase (Motoki and Seguro, 1998; Kuraishi et al., 2001; Gerrard, 2002; Ahhmed, 2007; 2009), and polysaccharides (Kato and Matsuda, 1969; Maekaji, 1974; Zhang et al., 2001). Different polymers express distinct chemical and electrical properties. For example, starch is an electrically neutral polymer composed of a single kind of monomer, glucose, while proteins can be regarded as amphoteric polymers constructed using 20 different amino acids. Some polysaccharides, such as alginic acid, carrageenan, or agarose, express a negative charge because of carboxylic (Gacesa, 1988), sulfonic (De Ruiter and Rudolph, 1997; Campo et al., 2009), or pyruvic (Fatin-Rouge et al., 2003) substitutes on their major chains.

Generally, low-molecular-mass compounds are used in the seasoning of food. However, their chemical or electrical properties are distinct. For example, sucrose, used in enhancing the sweetness of food, is a non-electrolyte; sodium chloride, which is used to enhance saltiness, is a strong electrolyte; while acetic acid (sour taste) is a weak electrolyte. Compounds that cause a bitter taste are generally hydrophobic in nature (Fox, 1932; Belitz and Wieser, 1985; Ishibashi et al., 1988). Therefore, many interactions, such as size exclusion and ion exchange, occur during the seasoning of food. The seasoning process is characterized by two factors: the distribution of a seasoning compound onto the food gel and the diffusion of the compound in the gel. The distribution determines the final concentration of the compound in the gel, while the diffusion regulates the time required by the gel to reach an equilibrium state.

Some studies reported diffusion of solutes in food polymer gels, such as the effective diffusion coefficient of glucose in κ-carrageenan gel (Nguyen and Luong, 1986), glucose and amino acid in alginate gel (Tanaka et al., 1984) and sodium chloride in agar gel (Odake et al., 1990). From the point of practical food processing, the diffusion of glucose and salts in beef and pork (Djelveh and Gros, 1988), and of salt in white cheese (Turhan and Kaletunç, 1992) were evaluated. Distribution of a solute onto the food matrix plays an important role in the seasoning process as well as diffusion. As abovementioned, food gel matrices and seasoning compounds are inherently different in their electrical properties and there are many combinations between food matrices and seasoning compounds. However, there has been minimal systematic study of the distribution of seasoning compounds in food matrices.

In this study, a starch gel, an elementary food gel-model, was used for analysis. Starch is a homopolymer comprised of glucose residues linked by α-1,4-glycosidic bonds (Pigman, 1970), and is therefore regarded as an electrically neutral polymer. Initially, the effects of temperature on the apparent density and porosity of the starch gel were evaluated. Subsequently, the distribution coefficients of non-electrolytes on the starch gel were measured at a temperature range of 25 – 60°C in order to estimate the swelling pressure of the gel at each temperature. The distribution coefficients of electrolytes were also measured. Finally, the distribution and diffusion coefficients of typical seasoning compounds, exhibiting different chemical properties on or within the gel, were measured.

Materials and Methods

Materials    Dried spherical tapioca beads (Youki Food, Tokyo, Japan) were purchased from a supermarket in Kobe, Japan. Ethylene glycol, glycerol, glucose, fructose, sucrose, and raffinose pentahydrate were purchased from Wako (Wako Pure Chemical Industries, Osaka, Japan). Sodium hydrogen l(+)-glutamate monohydrate, vanillin, glucose CII-test kit, and invertase (obtained from yeast) were also purchased from Wako. Other chemicals were purchased from Wako or Nacalai Tesque Inc. (Kyoto, Japan).

Preparation of starch gel    Five grams of dried tapioca beads were rehydrated for 20 min in 500 mL of boiling water in an Erlenmeyer flask, with gentle stirring using a magnetic stirrer to prevent sedimentation of the beads. The beads were then quickly removed from the boiling water and washed with 1 L of pre-cooled water. The beads were soaked in a large quantity of water for at least 3 h to attain equilibrium at 25°C. The radii of the tapioca beads before and after rehydration were ca. 1.4 and 3.0 mm, respectively. The beads are hereafter denoted as starch gel, or merely “the gel”.

Apparent density and porosity of the starch gel    The gel was equilibrated in distilled water at varying temperatures (25, 40, 50, 60, or 80°C), using a temperature-controlled water bath (SDminiN or SD; Taitec Corporation, Saitama, Japan) and a cooler (Cool Way 100; Gex Corporation, Osaka, Japan) for at least 2 h. The apparent density of the starch gel was pycnometrically determined. The gel was dried at 105°C for 5 days using a constant temperature oven (DN-400; Yamato Scientific, Tokyo, Japan). The pore volume of the gel was estimated by dividing the difference between the dry and wet weights of the gel by the density of water. Porosity, which was defined as a ratio of water volume in the gel to the whole gel volume, was calculated by dividing the pore volume by the whole gel volume.

Distribution coefficient on the starch gel    The distribution coefficient of a solute on the gel was measured by using an adsorption-desorption process. Briefly, 1 g of the gel was precisely weighed and soaked in 15 mL of a solution (containing a specific solute; concentration ranging from 0.1% (w/v) to 15% (w/v)). The gel was incubated until distribution equilibrium was achieved (6 h or longer), and was subsequently removed from the solution. The excess solution on the gel surface was blotted using a Kimtowel (Nippon Paper Crecia, Tokyo, Japan). The gel was then soaked in 2 mL of distilled water, in order to desorb the solute, for 6 h or longer. The concentrations of non-electrolytes and electrolytes were determined using a pocket refractometer (PAL-1; Atago, Tokyo, Japan) and pocket salinometer (PAL-ES1, Atago), respectively. The concentrations of acetic acid and monosodium glutamate were also analyzed using the refractometer and salinometer, respectively. The vanillin concentration was determined by measuring the absorbance at 280 nm using a UV-1200 spectrophotometer (Shimadzu Corp., Kyoto, Japan).

Mass balance for a solute before and after desorption was expressed using Eq. (1).

  

where C₀ is the solute concentration of the outer solution for adsorption, C_f is the solute concentration of the outer solution after desorption, V_L is the volume of distilled water used for desorption, and W_g is the weight of starch gel. Equation (1) can be re-written as follows:

  

When the K_app is independent of the solute concentration, the plot of C_f against [(W_g/ρ_app)/V_L] × (C₀ − C_f) produces a straight line passing through the origin, and the K_app value is obtained from the slope of this line.

Diffusion coefficient in starch gel    The gel was immersed in an excess volume of 30 mmol/L glucose, 30 mmol/L sucrose, 500 mmol/L NaCl, 500 mmol/L monosodium glutamate, 800 mmol/L acetic acid, or 10 mmol/L vanillin, in order to attain distribution equilibrium. The diffusion coefficient was then measured using a desorption method described in previous studies (Horowitz and Fenichel, 1964; Nakanishi et al., 1977; Westrin et al., 1994).

The concentrations of NaCl, monosodium glutamate, and acetic acid were determined using an electrical conductivity meter (CD-35MII; M & S Instruments, Osaka, Japan). The concentration of vanillin was determined as mentioned in the previous section. The concentration of glucose was measured with the glucose CII-test kit. Sucrose was hydrolyzed by invertase; the invertase solution was diluted in 50 mmol/L acetate buffer (pH 5.0) to a concentration of 0.016 U/L and 1 mL of the sucrose solution was mixed with 20 µL of this diluted invertase solution. Following the complete hydrolysis of sucrose, the glucose produced was determined using the glucose CII-test kit.

Differential scanning calorimetry (DSC) measurement    Distilled water (6 times the weight of tapioca beads) was added to the beads. The beads were ground into a paste with a mortar and pestle. The sample (10 mg) was precisely weighed using a thermogravimeter (TGA-50; Shimadzu Corp.) and sealed in a disposable aluminum cell. The sample-containing cell was heated to 95°C from the ambient temperature at a rate of 5°C/min using a differential scanning calorimeter (DSC-50; Shimadzu Corp.). The same quantity of distilled water was used as a reference.

Ash analysis    Dry tapioca beads were incinerated in a crucible (treated at 550°C for 2 h) using an electrical furnace (SH-OMT-II; Nitto Kagaku, Nagoya, Japan) maintained at 600°C for 2 h, and the ash content was determined.

A qualitative analysis of the ash was also performed. The ash (0.031 g) was dissolved in a 1.0 mL HCl and 0.5 mL nitric acid mixture, and diluted up to 25 mL using distilled water. The elements present in the ash were analyzed using an ICP emission spectrometer (ICP-8100; Shimadzu Corp.), at volumetric flow rates of coolant, plasma, and carrier gases of 14 L/min, 1.2 L/min, and 0.70 L/min, respectively.

Results and Discussion

Temperature dependence of apparent density and porosity of starch gel    Figure 1 displays the apparent density, ρ_app, and porosity, ɛ_p, of the starch gel at various temperatures. The ρ_app was lower and the ɛp was greater at a high temperature. These results suggested that the starch gel swells more at higher temperatures. Water regains of the starch gel were 1159 ± 3, 1125 ± 10, 1575 ± 38, 1382 ± 17 and 1986 ± 23 g-water/g-dry gel at 25, 40, 50, 60 and 80°C, respectively, also indicating greater swelling at higher temperature. However, the ɛ_p value at 60°C was slightly lower than that at 50°C.

Fig. 1.

Apparent density (○) and porosity (▵) of the starch gel at various temperatures. The values are shown as mean ± standard deviation (SD) (n = 3).

Estimation of swelling pressure of starch gel    Figure 2 shows the C_f vs. [(W_g/ρ_app)/V_L]·(C₀ − C_f) plots for ethylene glycol, glucose, and raffinose at 25°C and 50°C. The plots for each solute at a specific temperature produced a straight line passing through the origin, and the K_app value was calculated from the slope of the line. Other non-electrolytes also produced straight lines passing through the origin. Lower K_app values were observed for non-electrolytes with a larger molecular mass. The K_app values, which were lower than the ɛ_p values of the gel, indicated the effect of a factor other than the steric effect of the gel matrix on the distribution of non-electrolytes on the gel.

Fig. 2.

Relationship between C_f and [(W_g/ρ_app)/V_L]·(C₀−C_f) for ethylene glycol (○, •), glucose (▵, ▴), and raffinose (□, ■) at 25 (open symbols) and 50°C (closed symbols).

We previously reported that the K_app of a non-electrolyte on a cation-exchange resin can be expressed by Eq. (3), taking into consideration the swelling pressure of the resin (Π) (Adachi et al., 1989).

  

where R is the gas constant, T is the absolute temperature, ν_S denotes the partial molar volume of a solute, and γ₀ represents a parameter including the ratio of the activity coefficient of a solute in the outer solution phase to that in the resin phase, and the steric effect on the distribution of a solute on the resin.

Equation (3) was applied to estimate the swelling pressure of the starch gel. For this, we used the molar volume of each solute instead of the partial molar volume. Briefly, a radius of a hydrated solute, r_S, was estimated from the diffusion coefficient at 25°C (Hayduk and Laudie, 1974; Ribeiro et al., 2006) using the Stokes-Einstein equation (Eq. (4)). The molar volume of the solute assumed that the solute is spherical.

  

where D₀ is the diffusion coefficient in a dilute solution, k_B is the Boltzmann constant, and µ is the viscosity of water.

The molar volumes of ethylene glycol, glycerol, fructose, glucose, sucrose, and raffinose were 0.024, 0.046, 0.114, 0.121, 0.259, and 0.456 L/mol, respectively. The molar volumes were assumed to be constant in the tested temperature range.

The K_app values of the non-electrolytes at various temperatures were plotted on a semi-logarithmic scale against their molar volumes (Fig. 3). The plots (at all temperatures) produced a straight line, and the swelling pressure of the starch gel at each temperature was estimated from the slope of the line. The Π values of the gel at 25, 40, 50, and 60°C were evaluated to be 1.1, 1.6, 2.9, and 1.7 MPa, respectively. As shown in Fig. 4, the Π value increased with the increase in temperature from 25 to 50°C, but decreased at 60°C. The enthalpy change, ΔH, for the swelling was estimated from the swelling pressures at a temperature range of 25 to 50°C, using the van't Hoff equation (Eq. (5)), to be 29.9 kJ/mol, indicating the swelling of the gel to be endothermic.

Fig. 3.

Estimation of swelling pressure of the starch gel at 25 (○), 40 (▵), 50 (□), and 60°C (⋄) using Eq. (3). The solutes used were ethylene glycol, glycerol, fructose, glucose, sucrose, and raffinose.

Fig. 4.

Temperature dependence of the swelling pressure of starch gel.

  

DSC measurement    Figure 5 displays the DSC curve of the tapioca gel. The DSC curve shows two endothermic peaks. The former peak would be ascribed to the gelatinization of starch and the latter one might be caused by transition of the amylose-lipid complex (Eliasson, 1994). The onset, peak, and conclusion temperatures of the gelatinization process were 63, 73, and 78°C, respectively. The enthalpy of the gelatinization reaction was calculated to be −3.13 J/g. The onset of gelatinization occurred in the range of 60°C, at which a decrease in the swelling pressure was observed in previous experiments. Therefore, the lower swelling pressure at 60°C was ascribed to the gelatinization of the tapioca starch. It was suggested that gelatinization of starch gel played an important role in the distribution of seasoning compounds to the gel.

Fig. 5.

Differential scanning calorimetry curve for tapioca starch.

Starch gel incubated at 25°C for several hours was also analyzed by DSC under the same condition. The DSC curve showed no peak (data not shown), suggesting that retrogradation of the gel did not occur when soaking in distilled water at 25°C.

Distribution coefficient of electrolyte on starch gel    The K_app values of seven electrolytes were measured at 25°C, and plotted against the molar volumes of the electrolytes in Fig. 6. The molar volume of each electrolyte was estimated under the following assumptions: the larger cation and anion in the electrolyte governs the distribution of the electrolyte on the gel due to the electro-neutrality in the gel phase. The molar volumes of hydrated lithium, sodium, potassium, cesium, fluoride, chloride, bromide, and iodide ions were estimated from their diffusion coefficients in dilute aqueous solution (Li and Gregory, 1974) to be 0.034, 0.016, 0.0049, 0.0042, 0.012, 0.0044, 0.0046, and 0.0047 L/mol. For example, the molar volume of NaCl was assumed to be 0.016 L/mol because the molar volume of sodium ion (0.016 L/mol) is larger than that of chloride ion (0.0044 L/mol).

Fig. 6.

Distribution coefficients of electrolytes at 25°C. Numbers represent the electrolytes used: 1: LiCl, 2: NaCl, 3: KCl, 4: CsCl, 5: NaF, 6: NaBr, and 7: NaI. The values are represented as mean ± SD.

When there is no interaction between the solute and starch gel, the distribution of the solute on the gel at a specific temperature should be determined using the swelling pressure of the gel and the molar volume of the solute. The solid line in Fig. 6 represents the molar volume dependence calculated from Eq. (3), using the swelling pressure of the gel at 25°C (1.1 MPa). All the plots were observed to lie under the curve, indicating that any factor could participate in the distribution of electrolyte on the gel.

The ash content of the starch gel was 73 ± 23 mg/100 g-wet sample. Analysis using the ICPS-8100 led to the detection of 19 elements, with the major elements listed in Table 1. The presence of alkali and alkaline earth metals in the starch gel suggested the effect of a weak electrostatic interaction on the distribution of electrolytes on the gel.

Table 1. Ash content of tapioca starch.

	Content [% (w/w)]
Sodium, Na	9.22
Magnesium, Mg	3.45
Calcium, Ca	21.8
Phosphorous, P	8.76
Sulphur, S	2.69
Potassium, K	7.34

It is known that Sephadex, dextran gel cross-linked with epichlorohydrin, contains a few carboxylic groups originating from terminal aldehyde groups in the glucose chains (Janson, 1967). Due to the negative charge, adsorption of cationic solutes and exclusion of anionic solutes can be observed when using the gel for gel filtration chromatography (Gelotte, 1960; Janson, 1967). The presence of sodium or potassium ion in the starch gel also suggests a negative charge in the gel. Some carboxylic groups might exist in starch gel, and this might cause a weak electrostatic interaction.

Distribution and diffusion coefficients of seasoning compounds onto starch gel    Distribution and diffusion coefficients, K_app and D_eff, of typical seasoning compounds were estimated at 25°C. Six compounds with distinct chemical properties were used to represent typical seasoning materials. These were glucose and sucrose (non-electrolyte), NaCl (strong electrolyte), acetic acid (weak electrolyte), monosodium glutamate (amphoteric ion), and vanillin (hydrophobic substance).

The K_app and D_eff of the seasoning materials are listed in Table 2. The K_app values were lower than unity in all components except for vanillin. Although the molecular masses of NaCl, acetic acid, and monosodium glutamate were smaller than those of glucose and sucrose, their K_app values were lower than those of glucose and sucrose. NaCl, acetic acid, and monosodium glutamate are electrolytes, while glucose and sucrose are non-electrolytes. Therefore, any electrical interaction between NaCl, acetic acid, or monosodium glutamate and the gel would assist in their distribution onto the gel, analogous with the above-mentioned salts. The K_app value of vanillin (greater than unity) suggested the effect of an unidentified interaction between the solute and the starch gel.

Table 2. K_app and D_eff of six seasoning compounds.

	Kapp [-]	Deff × 1010 [m2/s]	D0 × 1010 [m2/s]
Glucose	0.74 ± 0.03	3.81	6.75	(Hayduk and Laudie, 1974)
Sucrose	0.66 ± 0.06	3.12	5.24	(Hayduk and Laudie, 1974)
NaCl	0.69 ± 0.02	12.0	16.0	(Handbook of chemistry, 1984)
Acetic acid	0.53 ± 0.06	9.86	11.9	(Hayduk and Laudie, 1974)
Monosodium glutamate	0.62 ± 0.09	6.82	7.0	calculated according to Wilke's method (Wilke, 1949)
Vanillin	1.36 ± 0.15	4.32	6.3

^*  K_app values are expressed as means ± standard deviation (n = 5 − 8).

Greater D_eff values were observed for smaller compounds. The D_eff of a solute in a porous material was related to its diffusion coefficient in a dilute solution, D₀, using Eq. (6) (Westrin and Axelsson, 1991).

  

where τ denotes the tortuosity factor reflecting the effective diffusion distance. Figure 7 demonstrates the relationship between D_eff and ɛ_pD₀ for the seasoning compounds. The D₀ value of each solute was also shown in Table 2. The D₀ values of glucose, sucrose, acetic acid (Hayduk and Laudie, 1974) and NaCl (Handbook of chemistry, 1984) was cited from the literature, and those of monosodium glutamate and vanillin were estimated according to Wilke's method (Wilke, 1949). Most of the plots displayed a straight line passing through the origin, and the τ value of starch gel was estimated to be 1.17 from the slope of the line. Although vanillin showed weak adsorption onto the starch gel, the adsorption might scarcely affect the estimation of the diffusion coefficient due to low solid content and weak interaction. A possible reason why Eq. (6) could be applied for all solutes is the low contribution of the gel skeleton adsorption phenomena to diffusion.

Fig. 7.

Relationship between D_eff and ɛ_pD₀. Numbers represent the seasoning compounds 1: glucose, 2: sucrose, 3: NaCl, 4: acetic acid, 5: monosodium glutamate, and 6: vanillin.

The τ values of porous materials, which were used as heterogeneous catalysts, have been previously reported to be in the range of 1.5 and 6 (Satterfield, 1970). Examples include a τ value of 2.2 for a chromatographic resin based on methacrylate polymer, TSK-HW65F (Gibbs et al., 1992), and 1.88 to 2.49 for a chitosan gel cross-linked with glutaraldehyde (Krajewska and Olech, 1996). The tortuosity factor for the starch gel was much lower than these previously mentioned values, probably because of the high gel porosity (0.95) (Fig. 1).

Conclusions

The swelling pressure of the tapioca starch gel was estimated to be 1 – 3 MPa at a temperature range of 25 – 50°C from the molar volume dependence of the apparent distribution coefficients of non-electrolytes (with different molar volumes). The swelling pressure at 60°C was lower than that at 50°C; this was ascribed to starch gelatinization. The apparent distribution coefficients of electrolytes were lower than the values calculated from their molar volume and the swelling pressure of the gel. Based on the presence of alkali and alkaline earth metals in the gel, we theorize upon the role of weak electrostatic interactions on their distribution on the gel. The apparent distribution and effective diffusion coefficients of typical seasoning compounds on the gel were measured at 25°C. The tortuosity factor of the gel was estimated to be 1.17.

References

•  Adachi,  S.,  Watanabe,  T., and  Kohashi,  M. (1989). Role of swelling pressure on distribution coefficient of maltooligosaccharide in a cation-exchange resin. Agric. Biol. Chem., 53, 3203-3208.
•  Ahhmed,  A. M.,  Kawahara,  S.,  Ohta,  K.,  Nakade,  K.,  Soeda,  T., and  Muguruma,  M. (2007). Differentiation in improvements of gel strength in chicken and beef sausages induced by transglutaminase. Meat Sci., 76, 455-462.
•  Ahhmed,  A. M.,  Nasu,  T.,  Huy,  D. Q.,  Tomisaka,  Y.,  Kawahara,  S., and  Muguruma,  M. (2009). Effect of microbial transglutaminase on the natural actomyosin cross-linking in chicken and beef. Meat Sci., 82, 170-178.
•  Belitz,  H.-D. and  Wieser,  H. (1985). Bitter compounds: occurrence and structure-activity relationship. Food Rev. Int., 1, 271-354.
•  Campo,  V. L.,  Kawano,  D. F.,  De Silva,  D. B.,  and Carvalho,  I. (2009). Carrageenans: biological properties, chemical modifications and structural analysis—a review. Carbohydr. Polym., 77, 167-180.
•  De Ruiter,  G. A. and  Rudolph,  B. (1997). Carrageenan biotechnology. Trends Food Sci. Technol., 8, 389-395.
•  Djelveh,  G. and  Gros,  J. B. (1988). Measurement of effective diffusivities of ionic and non-ionic solutes through beef and pork muscles using a diffusion cell. Meat Sci., 23, 11-20.
•  Doi,  E. (1993). Gels and gelling of globular proteins. Trends Food Sci. Technol., 4, 1-5.
•  Eliasson,  J.-C. (1994). Interactions between starch and lipids studied by DSC. Thermochim. Acta, 246, 343-356.
•  Fatin-Rouge,  N.,  Milon,  A.,  Buffle,  J.,  Goulet,  R., and  Tessier,  A. (2003). Diffusion and partitioning of solutes in agarose hydrogels: the relative influence of electrostatic and specific interactions. J. Phys. Chem. B, 107, 12126-12137.
•  Fox,  A. L. (1932). The relationship between chemical constitution and taste. Genetics, 18, 115-120.
•  Gacesa,  P. (1988). Alginates. Carbohydr. Polym., 8, 161-182.
•  Gelotte,  B. (1960). Studies on gel filtration: sorption properties of the bed material sephadex. J. Chromatogr., 3, 330-342.
•  Gerrard,  J. A. (2002). Protein-protein cross-linking in food: methods, consequences, applications. Trends Food Sci. Technol., 13, 391-399.
•  Gibbs,  S. J.,  Lightfoot,  E. N., and  Root,  T. W. (1992). Protein diffusion in porous gel filtration chromatography media studied by pulsed field gradient NMR spectroscopy. J. Phys. Chem., 96, 7458-7462.
•  Handa,  A.,  Takahashi,  K.,  Kuroda,  N., and  Froning,  G. W. (1998). Heat-induced egg white gels as affected by pH. J. Food Sci., 63, 403-407.
•  Hatta,  H.,  Kitabatake,  N., and  Doi,  E. (1986). Turbidity and hardness of a heat-induced gel of hen egg ovalbumin. Agric. Biol. Chem., 50, 2083-2089.
•  Hayduk,  W. and  Laudie.  H. (1974). Prediction of diffusion coefficients for nonelectrolytes in dilute aqueous solutions. AIChE J., 20, 611-615.
•  Horowitz,  S. B. and  Fenichel,  I. R. (1964). Solute diffusional specificity in hydrogen-bonding systems. J. Phys. Chem., 68, 3378-3385.
•  Ishibashi,  N.,  Ono,  I.,  Kato,  K.,  Shigenaga,  T.,  Shinoda,  I.  Okai,  H., and  Fukui,  S. (1988). Role of hydrophobic amino acid residue in the bitterness of peptides. Agric. Biol. Chem., 52, 91-94.
•  Janson,  J.-C. (1967). Adsorption phenomena on sephadex^®. J. Chromatogr., 28, 12-20.
•  Kato,  K. and  Matsuda,  K. (1969). Studies on the chemical structure of konjac mannan. Part I. Isolation and characterization of oligosaccharides from the partial acid hydrolyzate of the mannan. Agric. Biol. Chem., 33, 1446-1453.
•  Krajewska,  B. and  Olech,  A. (1996). Pore structure of gel chitosan membranes. I. Solute diffusion measurements. Polym. Gels Netw., 4, 33-43.
•  Kuraishi,  C.,  Yamazaki,  K., and  Susa,  Y. (2001). Transglutaminase: its utilization in the food industry. Food Rev. Int., 17, 221-246.
•  Li,  Y.-H. and  Gregory,  S. (1974). Diffusion of ions in sea water and in deep-sea sediments. Geochim. Cosmochim. Acta, 38, 703-714.
•  Maekaji,  K. (1974). The mechanism of gelation of konjac mannan. Agric. Biol. Chem., 38, 315-321.
•  Motoki,  M. and  Seguro,  K. (1998). Transglutaminase and its use for food processing. Trends Food Sci. Technol., 9, 204-210.
•  Nakanishi,  K.,  Adachi,  S.,  Yamamoto,  S.,  Matsuno,  R.,  Tanaka,  A., and  Kamikubo,  T. (1977). Diffusion of saccharides and amino acids in cross-linked polymers. Agric. Biol. Chem., 41, 2455-2462.
•  Nguyen,  A.-L. and  Luong,  J. H. T. (1986). Diffusion in κ-carrageenan gel beads. Biotechnol. Bioeng., 28, 1261-1267.
•  Odake,  S.,  Hatae,  K.,  Shimada,  A., and  Iibuchi,  S. (1990). Apparent diffusion coefficient of sodium chloride in cubical agar gel. Agric. Biol. Chem., 54, 2811-2817.
•  Pigman,  W. (Ed.) (1970). “The carbohydrates: chemistry and biochemistry, 2nd Ed.” Academic Press, New York, USA, p.483.
•  Ribeiro,  A. C. F.,  Ortona,  O.,  Simões,  S. M. N.,  Santos,  C. I. A. V.,  Prazeres,  P. M. R. A.,  Valente,  A. J. M.,  Lobo,  V. M. M., and  Burrows,  H. D. (2006). Binary mutual diffusion coefficients of aqueous solutions of sucrose, lactose, glucose, and fructose in the temperature range from (298.15 to 328.15) K. J. Chem. Eng. Data, 51, 1836-1840.
•  Satterfield,  C. N. (1970). “Mass transfer in heterogeneous catalysis,” MIT Press, Cambridge, MA, p.157.
•  Tanaka,  H.,  Matsumura,  M., and  Veliky,  I. A. (1984). Diffusion characteristics of substrates in Ca-alginate gel beads. Biotechnol. Bioeng., 26, 53-58.
• The Chemical Society of Japan (Ed.) (1984). “Handbook of chemistry,” Maruzen, Tokyo (in Japanese).
•  Turhan,  M. and  Kaletunç. (1992). Modeling of salt diffusion in white cheese during long-term brining. J. Food Sci., 57, 1082-1085.
•  Westrin,  B. A. and  Axelsson,  A. (1991). Diffusion in gels containing immobilized cells: a critical review. Biotechnol. Bioeng., 38, 439-446.
•  Westrin,  B. A.,  Axelsson,  A., and  Zacchi,  G. (1994). Diffusion measurement in gels. J. Control. Release, 30, 189-199.
•  Wilke,  C. R. (1949). Estimation of liquid diffusion coefficients. Chem. Eng. Progr., 45, 218-224.
•  Zhang,  H.,  Yoshimura,  M.,  Nishinari,  K.,  Williams,  M. A. K.,  Foster,  T. J., and  Norton,  I. T. (2001). Gelation behavior of konjac glucomannan with different molecular weights. Biopolymers, 59, 38-50.