Moisture Sorption Isotherms and Prediction Models of Carboxymethyl Chitosan Films from Different Sources with Various Plasticizers

Department of Food Science and Technology, Faculty of Agro-Industry, Kasetsart University, Bangkok 10900, $ailand Center of International Standard Food Safety, Center for Advanced Studies Agriculture and Food, Kasetsart University, Bangkok 10900, $ailand School of Agro-Industry, Mae Fah Luang University, Chiang Rai 57100, $ailand Division of Packaging Technology, Faculty of Agro-Industry, Chiang Mai University, Chiang Mai 50100, $ailand Center of Excellence in Materials Science and Technology, Faculty of Science, Chiang Mai University, Chiang Mai 50200, $ailand


Introduction
Chitosan is a deacetylated derivative of chitin that is a major component of crustacean shells such as crab, shrimp, and squid pen.Chitosan has great potential for a wide range of food applications due to its biodegradability, nontoxicity, biocompatibility, antimicrobial activity, and film-forming capacity [1].However, chitosan is insoluble in aqueous solutions with pH above 6.5, which limits the practical utilization of chitosan.erefore, the requirement of water solubility of chitosan has led to several research studies attempting to convert chitosan into chitosan derivatives such as carboxymethyl chitosan (CMCH) [2][3][4][5].Carboxymethyl chitosan (CMCH) not only is soluble in water but also has unique chemical, physical, and biological properties such as high viscosity, large hydrodynamic volume, low toxicity, biocompatibility, biodegradability, and gel-forming capabilities, all of which make it an attractive option in connection with its use in food products and cosmetics [2,3].Most biodegradable films are sensitive to moisture, and their properties change with changes in relative humidity [6].e water sorption isotherm of a material represents the equilibrium relationship between water content and water activity (a w ) of any material at a given temperature [7]. is phenomenon is a major tool to describe and predict the mobility of films in different environments pertinent to their application [8].Moreover, sorption isotherms provide significant information regarding the process steps, thermodynamics, and structure investigations [9].Numerous mathematical models for the description of the moisture sorption behavior of foods are available due to the complexity of foodstuffs [10].Some authors have studied the sorption isotherms of biodegradable films.Cervera et al. [11] determined the moisture sorption of chitosan-amylose starch film formers with glycerol or erythritol as the plasticizer.Rachtanapun and Tongdeesoontorn [12] studied the sorption isotherm of carboxymethyl cellulose from papaya peel/corn flour blended films and discovered that knowledge regarding sorption isotherms was also important for predicting the moisture sorption properties of films via moisture sorption empirical models.In 2011, Rachtanapun and Tongdeesoontorn [13] reported the effect of NaOH concentration on sorption isotherm of carboxymethyl rice starch films and prediction models.Additionally, there has been plenty of research on chitosan films in which the researchers examined the moisture sorption [14][15][16][17], but a little attention has been focused on the isotherm of CMCH.
erefore, the objective of this research was to study the characteristics of CMCH, the solubility and moisture sorption isotherms of CMCH films in the polymers and oligomers of CMCH films from three different sources (shrimp, crab, and squid), and fit the experimental data with prediction models.

Materials and Methods
2.1.Materials.Chitosans are characterized by three different sources (shrimp, crab, and squid), and each source has two molecular sizes (polymer and oligomer) and the degree of deacetylation was more than 95% as shown in Table S1, which is provided in Supplementary Materials.All six chitosans were obtained from Ta Ming Enterprises Co., Ltd., Samutsakon, ailand.All the acids and the base-glacial acetic acid and sodium hydroxide (NaOH) (Labscan, ailand) and monochloroacetic acid (MCA) (Sigma-Aldrich, Germany)-were of analytical grade.Absolute ethanol, absolute methanol, and isopropanol were of commercial grade, obtained from the Northern Chemical Co., Ltd.(Chiang Mai, ailand).

Synthesis of Carboxymethyl Chitosan (CMCH).
CMCH was synthesized as described in detail elsewhere [4,5].e chitosan flake was ground and sieved to obtain chitosan particle size under 100-mesh.e chitosan (10 g) was suspended in 50% sodium hydroxide solution (60 ml), and 160 ml of isopropanol was added at 50 °C for 1 h.Monochloroacetic acid (20 g) was dissolved in isopropanol (20 ml) and added into the reaction mixture dropwise for 30 min, and the system was allowed to continuously react at the same temperature for 4 h.e reaction was stopped by adding 70% methyl alcohol, and the pH was adjusted to neutrality.After that, the solid was filtered and rinsed in 70-90% ethyl alcohol for desalting and dried at 80 °C overnight.

Characteristics of Raw Chitosan and Its Derivatives.
e IR spectrum of raw chitosan and CMCH were recorded on an attenuated total reflectance (ATR) Fourier transform infrared (FTIR) spectrometer (Bruker IR spectrometer model tensor 27, IR microscope model Hyperion 2000, Ettlingen, Germany).Approximately two to three milligrams (2-3 mg) of chitosan and CMCH were placed on single reflection (Pike Miracle ATR).All the spectra were analyzed by OPUS 6.5.

Preparation of CMCH Films.
e casting of the CMCH films was carried out by following Rachtanapun et al. [4] and Tantala et al. [5].e CMCH (3% w/v) solution was prepared by dissolving in deionized water (100 ml), and the films forming in the solution were degasified.CMCH blended films were prepared using a concentration of plasticizers at 20% w/w (glycerol (Gly), sorbitol (Sor), and polyethylene glycol 400 (PEG 400)).e CMCH plasticizer blended solutions were stirred for 30 min, and the films were dried at room temperature for 24 h.

Water Solubility of CMCH Films.
e film solubility in water was measured as the percentage of dry matter of the film solubilized in water during a 24-h period.is method was adapted from the methods of [18,19].e initial dry matter of each film was obtained after drying film specimens at 65 °C for 24 h followed by placement in 0% RH silica gel desiccator for 2 days.e dried films (about 0.3 g) were weighed (initial dry weight) and immersed in beakers containing 50 ml of distilled water at 23 ± 2 °C, which were then sealed and periodically agitated for 24 h.e solutions containing the film residues were filtered with Whatman filter paper No. 4 (previously dried at 105 °C for 24 h and weighed before using); the filters were dried at 80 °C for 24 h, and the dried filters with film residues were weighed to determine the weight of the dry matter (final dry weight).
e tests were performed in triplicate, and the solubility was calculated using the following equation: solubility(%) � initial dried weight − final dried weight initial dried weight × 100. (1) 2.6.Color Measurement.All of the samples of the CMCH films were color measured using Mini Scan XE. e experiment results were reported based on the International Commission on Illumination (CIE) standard.e L * value specifies the luminosity or the black and white tones; the a *

2
Advances in Materials Science and Engineering value specifies the color as either a green or a magenta hue, and the b * value specifies the color as either a blue or a yellow hue.All the samples were analyzed in triplicate.

Determination of Moisture Sorption Isotherms of CMCH
Films. e moisture sorption isotherm measurement was conducted according to the method of [12].Various film specimens (20 mm × 20 mm) were predried in a hot air oven (50 °C) for 3 h and placed for 2 days in desiccators containing silica gel to eliminate the initial moisture content.Next, the films were placed in desiccators over saturated solutions having the desired relative humidity (0% RH, 23% RH, 34% RH, 43% RH, 65% RH, 77% RH, and 86% RH) at 25 ± 1 °C (Table S2).
e film specimens were weighed every 24 h.When two consecutive weights were equal, it was assumed that an equilibrium condition was reached.Under the above conditions, an equilibrium period of 7 days was sufficient to establish moisture equilibrium.
e percent equilibrium moisture content (%EMC) was calculated using the following equation: where W e is the equilibrium weight of the CMCH films (g), W i is the initial weight of the CMCH films (g), and M i is the initial moisture content of the CMCH films (g/g).

Moisture Sorption Isotherm Curve
Fitting.In the present study, the nine sorption isotherm models that were selected to be used for fitting the sorption data consist of the Oswin, Caurie, Smith, Lewicki-2, BET, Hasley, Hendeson, GAB, and Lewicki-3 models as shown in Table S3, which is provided in Supplementary Materials.e equations were rearranged in the linear form to determine the appropriate constants by regression analysis.e goodness of fit of each model was computed in terms of the coefficient of determination (r 2 ) from the plot of the experimental (M exp ) and the predicted (M pre ) values, as well as the root-mean-square percentage error (%RMS) values.Higher values of r 2 and lower values of %RMS show the goodness of fit of each of the models.

Structure of Chitosan and CMCH.
e structural changes of chitosan and its derivatives were confirmed by FTIR spectroscopy.As shown in Figure 1, as a straight line, the basic IR spectrum of chitosan shows peaks assigned to the saccharide structure at 3455 cm −1 (O-H stretch), 2867 cm −1 (C-H stretch), 1598 cm −1 (N-H bend), 1154 cm −1 (bridge-O stretch), and 1094 cm −1 (C-O stretch) [20,21] and a strong amino characteristic peak around 3420 cm −1 , 1655 cm −1 , and 1325 cm −1 that were assigned to the amide I and the amide III bands, respectively.e dotted line (Figure 1) shows the main characteristic peaks of CMCH and the intrinsic peaks of the carboxyl group (1737 cm −1 ).e bands at 1599 cm −1 and 1401 cm −1 correspond to the carboxyl groups (which overlap with the N-H bend) and the carboxymethyl groups, respectively.Compared to the peaks of chitosan, the peaks of CMCH at 1599 cm −1 and 1324 cm −1 increase, thus indicating that carboxymethylation has occurred in both the amino and the hydroxyl groups of chitosan.Ge and Lao [22] confirmed the introduction of the carboxymethyl groups.Water-soluble carboxyalkyl chitosan derivatives with modified molecular chain containing the highest possible content of N-acetyl-Dglucosamine unit, the repeat unit in chitin, were used in this research with the appropriate reaction conditions and reagents to allow the preparation of N-, O-, or N, O-carboxymethyl chitosan because the reactive sites for the carboxymethylation of chitosan are the amino and the hydroxyl groups [23].When carboxymethylation is carried out by making chitosan react with monochloroacetic acid in propanol/aqueous sodium hydroxide, O-substitution is favored if the reaction is carried out at room temperature [24]. is is attributed to the fact that the hydroxyl groups bound to the carbon atoms 3-and 6-of the glucopyranose unit and the amino group bound to the carbon atom 2 are the sites available for carboxymethylation.Indeed, an N, N-distributed derivative may be obtained depending on the reaction conditions.Additionally, it should be noted that the carboxymethylation reaction is seldom complete, and thus, some hydroxyl and amino groups remain unsubstituted.Finally, one must consider that if the chitosan is not completely deacetylated, and there are also the 2-acetamide-2-deoxy-D-glucopyranose units coming from the partial deacetylation of the parent chitin [25].

Water Solubility of CMCH Films.
e solubility of CMCH in aqueous solution is shown in Table S4 (provided in Supplementary Materials).All of the films were soluble, with more than 80% solubility, and the plasticized CMCH film did not have significantly different (p > 0.05) watersoluble properties, except for squid oligomer CMCH enriched with Gly. e water solubility of CMCH depends on the degree of substitution of the carboxymethyl group onto some of the amino and primary hydroxyl sites of the glucosamine units of the chitosan structure [3]. is phenomenon agreed with our result from the FTIR spectra of CMCH, which showed the dominant peaks of the carboxyl group at 1737 cm −1 , 1599 cm −1 , and 1401 cm −1 .Moreover, it was observed that Sor-plasticized films showed little higher film solubility than Gly-and PEG 400-plasticized films, but that they were not significantly different (p < 0.05). is might be explained by the fact that sorbitol has a ring molecular conformation that may satirically hinder insertion between the chain of polymers, resulting in an easy escape into the solution, whereas PEG and Gly have straight chains that are inserted and positioned within the tridimensional polymer network [26].

Color of Films.
e color of the CMCH films was more affected by the source and type of chitosan and the type of plasticizer.e L * , a * , and b * values of the CMCH films plasticized by Gly, Sor, and PEG 400 were not found to be significantly different (p < 0.05; Table S5   e relationship between a w and %EMC, which provided the sorption isotherm curve, is demonstrated in Figure 2. e sorption isotherm curve of the CMCH films gives the characteristic sigmoid-shaped type II isotherm curve of normal moisture adsorption isotherms [28].e sorption isotherms of the CMCH films were similar to those of the carboxymethyl cellulose films from waste product from mulberry and carboxymethyl rice starch films [12,13].e slope of all the CMCH films at lower a w values was less, although an increase in the a w value (a w > 0.65) caused the slope to increase rapidly.
e shrimp CMCH films had the highest moisture content. is result is consistent with the moisture sorption of the shrimp CMCH films, which absorbed more moisture than the crab and the squid CMCH films.

Fitting of Sorption Isotherm Models to Experimental
Data.Nine moisture sorption isotherm mathematical models were fitted to the moisture sorption data for the whole range of a w .e constant values from the mathematic models, coefficient of determination (r 2 ), and %RMS of the CMCH polymer and oligomer films are provided in Supplementary Materials (Tables S6 and S7), respectively.
e Oswin model found very suitable descriptions of the moisture isotherms throughout the entire range of water activity [28].In addition, Lomauro et al. [29] concluded that this model fitted the sorption data for a considerable number of nuts, spices, and coffee.is model showed perfect fit for all types of CMCH films with r 2 in the range from 0.9195 to 0.996 and %RMS in the range from 1.6142 to 63.0973.
e Caurie model is purely a mathematical equation valid for a w values in the range of 0.0-0.85[30].Nonplasticized CMCH films showed %RMS values from 13.8323 to 29.7123, whereas for all the other blended films, the values were in the range from 5.0586 to 46.8450.In the plasticizer blended films, squid oligomer CMCH blending with PEG 400 showed a good fit (r 2 � 0.9899 and %RMS � 5.0586), whereas all the other blends showed high %RMS values, which can assigned to the poor fitness of this model.e Smith model was found to fit the water sorption isotherms of various polymers [31]. is model was applied to study the adsorption and desorption isotherms of Virgina-type peanuts of a w values above 0.3 [32], dried peas, and dried figs [33].e model showed lower values of %RMS for nonplasticized CMCH films than for other blended films.
e Lewicki model was developed to make it applicable to the high range of a w .It fitted well the moisture sorption data at high humidity and predicted that the water content value tends to infinity when the a w value approaches 1.0 [34].For this study, the Lewicki-3 model showed goodness of fit for crab oligomer CMCH films plasticized with PEG 400 (r 2 � 0.9892 and %RMS � 6.8744), and the Lewicki-3 model showed lower %RMS than the Lewicki-2 model for other types of films.In addition, the %RMS values of other CMCH films from the Lewicki-2 model were above 10, which reflect the poor fitness of this model.e Hasley model is recommended for use for the study of the sorption isotherms of meats, milk products, and vegetables [35].Iglesias et al. [36] applied the Hasley model to describe reasonably the sorption of dried figs, apricots, and raisins.Moreover, in 1985, Lomauro et al. [29] successfully used this model to study the water sorption of several nuts and oil seeds.Application of this model to fit the sorption data of CMCH blended films demonstrated the result as excellent fit for crab polymer CMCH films blending with glycerol (r 2 � 0.9874 and %RMS � 9.0030).
As for the Henderson model, this model is one of the most widely used, relating water activity to the amount of water.e model was applied to study the sorption isotherm of potatoes [37], lentils [38], onion [39], pineapple [40], and chestnut [41].e model showed goodness of fit for squid polymer and squid oligomer CMCH films blending with Gly (r 2 � 0.9881 and 0.9894, %RMS � 7.5712 and 7.3049, respectively) and crab oligomer CMCH film blending with PEG 400 (r 2 � 0.9901 and % RMS � 7.3484).For all the other films, the r 2 values were presented to range from 0.8834 to 0.9941, whereas %RMS values were found to range from 7.3049 to 48.1156.
e GAB and the BET models are the most accepted models for foods or edible materials [42] due to the value of the monolayer moisture content in each model that indicates the amount of adsorbed water to specific sites at the food surface and used as a tool to consideration the optimum value to assure food stability.In addition, the GAB model is based on multiple layers and condensed film water with a physical meaning for each constant [43].
e M 0 is the monolayer content, C is the heat of sorption of the first layer, and K is the factor correcting the properties of multilayer molecules with respect to the bulk lipid [44]. is model is used to evaluate a monolayer moisture content, which indicated to the maximum amount of adsorbed water in a single layer of dry product water.e monolayer moisture content value not only is a measure of the number of sorbing sites but also defines the physical and chemical stability of foods [45].e acceptability of monolayer moisture content value for food product from GAB model should not be more than 10% dry basis [31].e BETmodel is used to investigate a monolayer water content of product [43] but is applicable in a narrow range of a w (a w � 0-0.5) [46].e monolayer (M 0 ) of all the CMCH films from the GAB and the BET models are in the range of 0.0255 to 4.8323 g water/g dry film and 0.0952 to 2.7992 g water/g dry film, respectively.is value indicates the maximum amount of water that could be adsorbed in a single layer per gram of dry film, and it is a measure of the number of sorption sites [47].For oligomer CMCH films including shrimp and squid CMCH plasticized with glycerol, the results showed that the GAB model gave higher M 0 than the BET model, which is a result similar to the result obtained in the case of caboxymethyl cellulose films from waste mulberry by Rachtanapun and Tongdeesoontorn [48].In contrast, for all types of shrimp polymer CMCH films, the BET model reported higher M 0 than the GAB model, which is in agreement with the results obtained in the case of cassava starch-based films blended with gelatin and carboxymethyl cellulose [49]           Advances in Materials Science and Engineering GAB model is related to the di erence in the magnitude between the upper layers and the monolayer [50].e shrimp oligomer had higher C values than the plasticized lms.e shrimp polymer CMCH lms had higher C values than the shrimp oligomer CMCH lms, whereas the C values of the CMCH lms from the crab oligomer were higher than    Advances in Materials Science and Engineering and squid oligomer films can absorb moisture in the monolayer and in the upper layers.However, several models in this study can predict sorption isotherms of various CMCH films and fitted as well as or better than the GAB model, which reflected from higher r 2 values and lower % RMS than GAB model (Tables S6 and S7, provided in Supplementary Materials).Unfortunately, their constants from other models that are used in this case had no physical meaning.us, the GAB model was found to have better capability of estimation in predicting the CMCH films than other models.Similar was the case with Srinivasa et al. [30] who found that the GAB model was the best-fitting model for chitosan films.Figures 3-11 demonstrate the experimental versus predicted moisture content values by all the models used of all types of CMCH films, which obtained diagonal lines for low and intermediate a w levels, indicating low interaction between the components in accordance with their separation in the independent phases, as observed during the film drying [51].It can also be observed that at high levels of a w , the point increased rapidly and fall on the diagonal as a result of the interaction between the water molecules and the polar groups of the film [52].ese results indicate that all the models can be used to predict the moisture content of the CMCH films at a w values in the range of 0.1-0.8.

Conclusions
In summary, the synthesis of CMCH from various types and source of chitosan via carboxymethylation method showed that the carboxymethyl groups were substituted on the -OH and the -NH 2 groups of chitosan.e water solubility of the CMCH control films and the films blended with the plasticizer were not significantly different.e moisture sorption isotherms of CMCH from different chitosan sources (shrimp, crab, and squid) and molecular sizes (polymer and oligomer) were evaluated at several values of relative humidity (0% RH, 23% RH, 34% RH, 43% RH, 65% RH, 77% RH, and 86% RH) at 25 ± 1 °C.e equilibrium moisture content values of all the CMCH films were low at lower values of a w but increased considerably for values above a w � 0.65.e sorption analysis of the different models showed good fit, as determined by the r 2 and the %RMS values.e r 2 and the %RMS values of the different models were dependent on the source of CMCH and the type of the plasticizer and predicted as well as or better than the GAB model.Unfortunately, their constants from other models that are used in this case had no physical meaning.It can, therefore, be safely concluded that the GAB model is the better estimation model for predicting the CMCH films in comparison with other models.Overall, it can be stated that the constants derived from the different sorption models were useful in the gathering of information and evaluation of the stability of CMCH packaging films for specific end uses.

Figure 8 :
Figure 8: e comparison between the experimental moisture content values and those predicted for all the CMCH films from the Hasley model: shrimp polymer CMCH films (a); shrimp oligomer CMCH films (b); crab polymer CMCH films (c); crab oligomer CMCH films (d); squid polymer CMCH films (e); squid oligomer CMCH films (f ).
. e C parameter in the Advances in Materials Science and Engineering