Water sorption isotherms of different sodium alginates : thermodynamic evaluation and influence of mannuronate-guluronate copolymers

J Food Process Preserv. 2022;46:e17179.   | 1 of 8 https://doi.org/10.1111/jfpp.17179 wileyonlinelibrary.com/journal/jfpp 1 | INTRODUCTION Alginate extraction from brown seaweeds is based on the conversion of the alginic acid from the cell walls into alginate salt forms, followed by its precipitation and purification. Milled seaweeds are soaked in dilute mineral acid to remove fucoidans, laminarins, proteins, and polyphenols that could modify alginate features. Then, alginic acid is transformed into sodium alginate (SA) by employing alkaline solutions, meanwhile solid residues are removed by centrifugation and filtration. Finally, alginate is precipitated with ethanol Received:27July2022 | Revised:8September2022 | Accepted:14September2022 DOI: 10.1111/jfpp.17179 ORIGINAL ARTICLE Water sorption isotherms of different sodium alginates: Thermodynamic evaluation and influence of mannuronateguluronate copolymers Leticia Montes | Mauro Gisbert | Ramón Moreira This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. © 2022 The Authors. Journal of Food Processing and Preservation published by Wiley Periodicals LLC. Chemical Engineering Department, Universidade de Santiago de Compostela, Santiago de Compostela, Spain Correspondence Ramón Moreira, Department of Chemical Engineering, Universidade de Santiago de Compostela, rúa Lope Gómez de Marzoa, SantiagodeCompostelaE-15782,Spain. Email: ramon.moreir[email protected] Funding information Consellería de Cultura, Educación e Ordenación Universitaria, Xunta de Galicia; European Regional Development Fund; Ministerio de Ciencia e Innovación Abstract Water desorption isotherms of three alginates with different structural features were determined at 25, 37, and 50°C. The Halsey model was selected to fit the equilibrium water sorption data. Differential and integral enthalpy and entropy were estimated for tested alginates. Optimal storage conditions of tested alginates (moisture content from0.15to0.20 kgwater/kgdrysolidandrelativehumidityfrom35%to50%)were determined from the maximum and minimum integral enthalpy and entropy values, respectively. A model was proposed to estimate the water sorption isotherms of alginates based on the alginate monomers (mannuronate, M, and guluronate, G) at low water activity (<0.4). M fraction was mainly responsible for the hygroscopicity of alginates. Alginates with similar G fraction showed different hygroscopic features by the presence of more homopolymeric G blocks that could form helical structures at low moisture content, decreasing the water affinity. Practical applications Determination of water sorption isotherms is fundamental to determine the optimal storage conditions at different temperatures. Their knowledge is essential for designing drying equipment, selecting adequately drying conditions (temperature and relative humidity of air) and drying time. Mathematical models are useful to estimate equilibrium moisture content in wide ranges of water activity and temperature. The thermodynamic study also provides valuable information about energy consumption and consequently operational costs. In this case, high content of mannuronate, M, increases the hygroscopic character of alginates, but the optimal moisture content of dried alginates to achieve maximum stability during storing varies in a narrow interval (0.15–0.20 kgwater/kgdrysolid). 2 of 8 | MONTES et al. to obtain SA (Chee et al., 2011). Recent studies have showed that SA can also be obtained from wasted solids after polyphenols extraction (Montes et al., 2021). SA is a linear polysaccharide composed of βDmannuronic acid (M) and αLguluronic acid (G) linked by 1– 4 glycosidic bonds, and the M/G ratio gives relevant information about the polymer structure (Abkakhajouei et al., 2022). The composition of alginates depends on their natural source, geographical location, and seasonal variations (Fernando et al., 2020). SA is used in cosmetic, pharmaceutical, medical, and textile industries. Particularly, it is widely used in food industry due to its thickening, emulsifying, gelling, stabilizing, and filmforming properties, being one of the most important food additives (Qin et al., 2018). SA is a hygroscopic material and tends to modify its moisture content as a function of environmental air conditions; hence, adequate storage conditions are very relevant for its correct conservation (Lee & Mooney, 2012). Most biopolymers are sensitive to moisture content, so their properties change with relative humidity and temperature (Kurek et al., 2014). By means of water sorption isotherms, which relate the equilibrium moisture content (X) and water activity (aw), the optimally hygroscopic conditions can be obtained for its preservation (Shivhare et al., 2004). There are many (empirical, semitheoretical, and theoretical) equations to model the water sorption isotherms. Halsey model can be used to study the multilayer water adsorption (Halsey, 1948). In this model, temperature can be introduced as variable within the model, and one equation is useful to reproduce simultaneously the equilibrium moisture content of a sample over a broad water activity and temperature ranges. Hygroscopic properties of a food material depend on its chemical composition (Moreira et al., 2009). In this way, the presence of additives such as SA can noticeably modify the equilibrium moisture content of the final product under the same storage conditions. At these circumstances, it is crucial to knowledge the hygroscopic behavior of compounds present in food and nonfood formulations. To understand the SA water sorption features and to estimate its optimal storage conditions, some thermodynamic properties can be evaluated such as differential heat of sorption and differential entropy, as well as the integral enthalpy and entropy (Zhang et al., 2016). The differential enthalpy of sorption is an indicator of the water binding strength to the solid, meanwhile the differential entropy is proportional to the number of available sorption site corresponding to a specific energy level (Koksharov et al., 2021). On the other hand, the integral enthalpy provides an indication of the total energy available to do work, and the integral entropy describes the degree of disorder and randomness of motion of water molecules (Moreira et al., 2008). There are some studies concerning the water sorption isotherms of alginates. For instance, Galus and Lenart (2013) studied the water adsorption isotherms of SA at 25°C, and the experimental data were successfully fitted by Peleg's model. Adamczak et al. (2017) determined water adsorption and desorption isotherms of SA at 25°C, with slightly higher moisture content for desorption in comparison to adsorption process. Xiao and Tong (2013) employed GAB model to fit experimental data of water adsorption of lowviscosity SA in the temperature range from 25 to 45°C. Previous results indicate discrepancies in the literature regarding hygroscopic properties of SA due to probably the different sources and extraction procedures employed to obtain commercial SA. To the best of our knowledge, no studies relating the sorption isotherms features with the structural features of alginates were found. Therefore, the goals of this study were to: 1. determine water sorption characteristics of SA with different M/G ratios and their modeling using the Halsey model; 2. evaluate some thermodynamic properties to understand sorption phenomena in depth and to assess the optimal storage conditions for dried SA; and 3. establish the relationship between M/G ratio of the different SA and the corresponding water sorption isotherms. 2 | MATERIALS AND METHODS 2.1 | Materials Processed and commercial sodium alginates (SA) were used. Specifically, processed SA from Ascophyllum nodosum brown seaweeds was obtained using a previously reported methodology (Montes et al., 2021), seaweeds pellets dried at 50°C (50D). Commercialsodiumalginates(CASNo.9005-38-3)werepurchased from Sigma-Aldrich Chemical Company (S) (Lot MKCJ1280, St. Louis, MO, USA) and PanReac (P) (Lot 0F009964, Barcelona, Spain). Average viscosimetric molecular weights (Mv, kg/mol), P (459 ± 8), S(156 ± 2),and50D(257 ± 1)andthecorrespondingaverageblock length, mannuronateguluronate ratio (M/G), P (1.15), S (0.91) and 50D (1.21), diads (FGG), P(0.23), S (0.35) and 50D (0.20) and triads (FGGG) P(0.13), S (0.30) and 50D (0.10), were previously determined by Montes et al. (2021). 2.2 | Determination of water desorption isotherms A gravimetric technique was carried out to determine the equilibrium moisture content of SA samples. Firstly, samples were hydrated for3 weeksuntilconstantweight.Then,severaljarswereprepared with different saturated salt solutions to obtain atmospheres with constant water activity. The salt solutions used were LiCl, MgCl2, Mg(NO3)2, NaCl, KCl, and BaCl2, which were prepared according to Moreira et al. (2008). The range of relative humidity of air achieved with these salts was within the interval 11%–90%. The samples (0.5 g), previously weighted, were placed in glass jars and introduced in the flasks at three temperatures: 20, 37, and 50°C (±0.1°C). Thymol was introduced in the jars to inhibit microbial growth at relative humidity higher than 0.6. An analytical balance (SI234, Denver Instrument, ±0.0005 g) was used to weight the samples at regular intervals until constant weight. The time required to achieve the equilibrium was about 12 weeks. The moisture content was determined using a vacuum oven(VacuthermVT650,HeraeusHanau)at70°Cand13 kPauntil achieve constant weight. The equilibrium moisture content (X) was 17454549, 2022, 12, Downloaded from https://ifst.onlinelibrary.wiley.com/doi/10.1111/jfpp.17179 by Universidade de Santiago de Compostela, Wiley Online Library on [15/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | 3 of 8 MONTES et al. measured for all samples, and the water sorption isotherms curves were plotted as X versus aw. All experiments were done at least in duplicate. 2.3 | Data analysis 2.3.1 | Sorptionisothermsmodels The experimental results were fitted by Halsey model, Equation (1): where X is the equilibrium moisture content (kg water/kg dry solid, d.b.), aw is the water activity, T (K) is the absolute temperature, and A and r are the fitting parameters of Halsey model. 2.3.2 | Differentialandintegral enthalpy and entropy The complete and detailed procedure is explained in Moreira et al. (2008). Briefly, the isosteric heat of sorption, Qst (kJ/mol), (or differential enthalpy) is an indicator of the state of water absorbed by the solid material and is defined by Equation (2): where qst (kJ/mol) is the net isosteric heat of sorption, and HL (kJ/mol) is the heat of vaporization of water at the sorption temperature. Using the Clausius– Clapeyron relationship, qst at constant X can be evaluated from the experimental data by Equation (3): where R(kJ/mol K)istheuniversalconstantofgases. The differential entropy, Sd(kJ/mol K),ofwateradsorptioncan be calculated from Gibbs– Helmholtz equation and substituting the Gibbs energy by its definition, a linear equation, Equation (4), involving Qst, Sd, and aw, is obtained: where the Sd value can be calculated from the intercept (Sd/R). The net integral enthalpy, qeq (kJ/mol), must be evaluated at constant spreading pressure, ∅ (J/m2), according to Equation (5): The spreading pressure represents the surface excess free energy and provides an indication of the increase in surface tension of bare sorption sites due to adsorbed molecules (Fasina et al., 1999). This property cannot be experimentally measured but can be estimated by Equation (6). where KBistheBoltzmann'sconstant(1.38 × 10−23 J/K), Am represents theareaofawatermolecule(1.06 × 10−19 m2), a is A/T. Finally, the net integral entropy, Seq(kJ/mol K),iscalculatedby Equation (7): where aw is obtained at constant ∅ at different T. 2.4 | Statistical analysis Experimental data were analyzed through onefactor analysis of variance (ANOVA), followed by the Duncan test, and considering significant p values <0.05 (IBM SPSS Statistics 27, SPSS Inc). All experimentalresultswereexpressedasmean ± standarddeviation from at least duplicate experiments. The goodness of fitting of Halsey model was estimated by two statistical indices previously proposed by Moreira et al. (2017), φ (Equation [8]), which is a lumped measure and involves the coefficient of determination (R2), the root mean squared error (RMSE) and the mean relative deviation (MRD), and χ2. If φ shows low values, the model shows a poor adequacy to describe the experimental behavior, and if the value of χ2 ≥ 5.99,themodelshouldberejectedwith p > 0.95. 3 | RESULTS AND DISCUSSION 3.1 | Experimental water desorption isotherms and modeling Experimental data of equilibrium water desorption isotherms of different sodium alginates (SA) at 20, 37, and 50°C are depicted in Figure 1. Water sorption isotherms can be classified as type III according to BET classification (Brunauer et al., 1940). All tested SA showed the same temperature trend at constant water activity, X decreased with increasing temperature. In all cases, at a constant temperature, the equilibrium moisture content (X) increased with increasing water activity (aw). However, two different regions in (1) X = ( −A Tln ( a w))1 r (2) Qst =qst +HL (3) q st =−R [dlna w d ( 1 ∕ T )]X (4) ln ( aw )|||X = Q st RT − Sd R (5) q eq =−R [ dln ( aw ) d(1∕T) ]� (6) �= KBT Am a1∕r ⎡⎢⎢⎢⎣ 1 � 1 r−1 �� −ln � aw �� 1 r−1 ⎤⎥⎥⎥⎦ a w 0.05 (7) S eq = −q eq T −Rln ( aw ) (8) 𝜑 =R 2 (RMSE)(MRD) 17454549, 2022, 12, Downloaded from https://ifst.onlinelibrary.wiley.com/doi/10.1111/jfpp.17179 by Universidade de Santiago de Compostela, Wiley Online Library on [15/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 4 of 8 | MONTES et al. the isotherm curves can be observed. Firstly, a linear trend was observed at low and intermediate aw values (from 0.1 to 0.5), this region is called multilayer sorption region. Secondly, at higher aw values, the capillary condensation region can be observed by the pronounced increase of X with increasing aw. Significant differences were found between tested SA in this last region (aw > 0.5).Particularly,X value at the highest aw (0.9) was approximately 0.6 (kg/kg d.b.) for S and Palginates(withoutsignificantdifferencesbetweenthem)and0.8 (kg/kg d.b.) for 50D. However, below 0.5 of aw, no significant differences (p ≤ 0.05)werefoundbetweentestedSA.Theseresultspartly agree with those reported previously by Galus and Lenart (2013) at 25°C, that is, at aw of 0.5 X was 0.25 (kg/kg d.b.); however, at aw of 0.9 these authors found higher moisture content value (1.3 kg/kg d.b.) for adsorption process. Also, Adamczak et al. (2017) reported higher X values (up to 1.16 kg/kg d.b. at aw of 0.9), for the water adsorption stage at 25°C. However, Shimanuki et al. (2020) determined, at the same high aw value, a moisture content approximately 0.5 kg/kg d.b that agreed with results obtained for S and P alginates tested in the present study. Finally, Xiao and Tong (2013) found for water adsorption of lowviscosity SA at 25°C an equilibrium moisture content of 0.4 kg/kg d.b. at aw of 0.90. The Halsey model was employed to fit experimental data, desorption isotherms are plotted in Figure 1 by continuous lines, and the corresponding fitting parameters are presented in Table 1. No significant differences (p ≤ 0.05)betweenHalseyparameters(A and r) for S and P alginates at constant temperature. For these alginates, Avaluesdecreased(from12.08to8.73andfrom12.46to8.42,respectively) with increasing temperature. However, for 50D, A values increased(from23.21to25.28)withtemperature.Interestingly,a (A/T) parameter value was invariant with temperature in the case of 50D alginate. In all cases, the r values were invariant with temperatureforeachalginate,and50Dshowedthelowestvalue(1.187), meaning the highest slope of the isotherm curve, against S and P alginates(around1.820). 3.2 | Thermodynamic properties 3.2.1 | Differentialenthalpyandentropy Figure 2a shows the variation of the differential enthalpy (qst), evaluated by means of Equation (3), with the moisture content of tested alginates at arithmetic mean temperature (35.7°C). At low moisture content, there are high attractive intermolecular forces between alginate surface and adsorbed water. Afterward, a sharp fall in the qst values is observedwhenmoisturecontentincreasesfrom0.08to0.2kg/kgd.b., because water molecules are adsorbed in other available sites with lower specific energy (Polachini et al., 2016). Subsequently qst values dramatically decreased and at moisture content above 0.3 kg/kg d.b., water molecules multilayers were formed and progressively approached zero meaning that the adsorption of a water molecule involved an energy equivalent to the heat of vaporization of Xiao and Tong (2013) employing lowviscosity SA observed this effect, but the values obtained (10 kJ/mol at X of 0.1 kg/kg d.b.) were lower than the valuesofthiswork(60 kJ/molatX of 0.1 kg/kg d.b.). The highest qst values found at low X (<0.25 kg/kgd.b.)wereforPalginate. Figure 2b shows the trend of differential entropy (Sd), Equation (4), with moisture content of tested alginates at 35°C. Sd is proportional to the number of available sorption sites at a specific energy level. Negative values are related to the loss of mobility of water molecules during sorption. The Sd values increased continuously atlowmoisture content(below 0.20 kg/kgd.b.)and above this content remained practically constant. At low moisture content, water molecules were strongly retained (high activation energies) in many available sorption sites, but these sites were progressively occupied with increasing moisture content (Madamba et al., 1996). FIGURE 1 Waterdesorptionisothermsofdifferentalginates:(a) from sigma (S), (b) from PanReac (P), and (c) processed (50D) at 20, 37, and 50°C. lines correspond to Halsey model (Equation [1]). 0 0.2 0.4 0.6 0.8 00.2 0.40.6 0. 81 X(kg water/kg d.b.) a w (a) 20ºC37ºC 50ºC 0 0.2 0.4 0.6 0.8 00.2 0.40.6 0. 81 X(kg water/kg d.b.) a w (b) 20ºC 37ºC50ºC 0 0.2 0.4 0.6 0.8 1 00.2 0.40.6 0. 81 X(kg water/kg d.b.) a w (c) 20ºC 37ºC 50ºC 17454549, 2022, 12, Downloaded from https://ifst.onlinelibrary.wiley.com/doi/10.1111/jfpp.17179 by Universidade de Santiago de Compostela, Wiley Online Library on [15/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | 5 of 8 MONTES et al. The theory of compensation needs to be applicable that isokinetic (TI) and harmonic mean (Th) temperatures are different and, in the range of moisture content studied, the existence of a linear relationship between differential enthalpy and differential entropy (Moreira et al., 2008). The linear relationship was verified by the plotting of qst versus Sd and from the slope TI values were obtained (340, 337 and 329 K for S, P and 50D, respectively), and Th was TABLE 1 ValuesoftheparametersofHalseymodel(Equation 1) and goodness of fitting for water sorption isotherms of commercial S (sigma) and P (PanReac) and processed (50D) alginates Sample SP50D T (K) 293.1 310.1 323.1 293.1 310.1 323.1 293.1 310.1 323.1 A12.08 ± 0.84a,B 10.18 ± 0.77ab,B 8.73 ± 0.71b,B 12.46 ± 1.03a,B 10.17 ± 0.96ab,B 8.42 ± 0.91b,B 23.21 ± 1.02a,A 24.38 ± 1.48ab,A 25.28 ± 0.84b,A a = A/T0.041 ± 0.003a,B 0.033 ± 0.003b,B 0.027 ± 0.002b,B 0.043 ± 0.004a,B 0.033 ± 0.003ab,B 0.026 ± 0.003b,B 0.079 ± 0.002a,A 0.079 ± 0.004a,A 0.078 ± 0.005a,A r1.815 ± 0.076A1.847 ± 0.027A1.187 ± 0.012B φ*347.5 308.1 1287.5 293.8 183.6 261.3 65.7 319.8 152.8 χ2** 4.2 3.3 5.4 3.9 5.6 1.1 1.0 0.6 0.1 Note:Dataarepresentedasmean ± standarddeviation.Datavalueofeachparameterwithdifferentsuperscriptlowercaselettersaresignificantlydifferentbytemperatureandcapitallettersbyalginateat constant temperature (p < 0.05). *Low values of φ indicate a poor adequacy of the model.; **With value χ2≥ 5.99,themodelshouldberejected(p > 0.95). FIGURE 2 Effectofmoisturecontent,X, on the net isosteric heat of sorption, qst, (a) and differential entropy, Sd, (b) for sigma (S), PanReac (P), and processed (50D) alginates. 0 20 40 60 80 100 00.1 0.20.3 0. 40 .5 q st (kJ/mol) X(kg water/kg d.b.) (a)S P 50D -300 -200 -100 0 00.1 0.20.3 0. 40 .5 Sd(J/mol K) X (kg water/kg d.b.) (b) S P 50D FIGURE 3 Spreadingpressure, ∅ , versus water activity, aw, at different temperatures for alginates (a) from sigma (S), (b) from PanReac (P), and (c) processed alginate (50D). 0.0 0.5 1.0 1.5 2.0 2.5 00.3 0. 60 .9 Ø(J/m 2 ) a w S_20 ºC S_37 ºC S_50 ºC P_20 ºC P_37 ºC P_50 ºC 50D_20 ºC 50D_37 ºC 50D_50ºC 17454549, 2022, 12, Downloaded from https://ifst.onlinelibrary.wiley.com/doi/10.1111/jfpp.17179 by Universidade de Santiago de Compostela, Wiley Online Library on [15/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 6 of 8 | MONTES et al. 308 K.AsTI > Th in all cases, the water sorption of tested alginates can be characterized as enthalpydriven (Moreira et al., 2016). 3.2.2 | Integralenthalpyandentropy Figure 3 shows the spreading pressure, calculated by means of Equation (6), as a function of aw for tested alginates at studied temperatures.Thespreadingpressureincreased(upto1.50 J/m2) with increasing aw and decreased with increasing temperature in all cases. Slight differences were found between spreading pressure values of S and P alginates at constant temperature, and values of 50D alginate were systematically lower. In fact, the ∅ values for S and P at 50°C were similar to 50D values at 37°C. These trends agreed with results reported for alginates by Xiao and Tong (2013), but with lower values (<0.1 J/m2). Figure 4a shows the variation of the net integral enthalpy, qeq, calculated with Equation (5) at constant spreading pressure, with the moisture content. At low X, qeq increased up to achieve maximum values(9.8,11.0and15.2kJ/molforS,P,and50D,respectively)and then continuously decreased with increasing X. On the other hand, the net integral entropy, Seq evaluated with Equation (7), decreased at low Xreachingaminimumvalue(−24.9,−28.7and−42.2J/mol K for S, P and 50D, respectively), Figure 4b. Both enthalpy and entropy trends with moisture content reflect the transition from the water molecules occupation of easily accessible sites to localized binding followed by the formation of multilayers (Moreira et al., 2008). Regarding the integral enthalpy and entropy curves, the maximum enthalpy and minimum entropy values occurred at the same narrow moisture content range from 0.15 to 0.2 kg/kg d.b., for tested alginates. The minimum integral entropy determines the water activity at which the food product has the highest stability, and additionally, other authors indicated that maximum enthalpy is achieved with the formation of a monolayer of adsorbed water (Kaya & Kahyaoglu, 2007). In this case, it means that maximum stability (optimal water activity) is achieved in the interval from 0.35 to 0.45 for S and P alginates and from 0.40 to 0.50 for 50D alginate. 3.3 | Relationship between equilibrium moisture content and M/G ratio of alginates To find a relationship between structural features of alginates and water molecules sorption on the sample surface, the water activity range must be restricted to the water activity range in which the water monolayer is formed. In this case, this aw interval was below 0.4. Parameters of Halsey model (Table 1) were employed to find correlations with M/G ratio of tested alginates. Firstly, a linear regression (R2 > 0.9) was established between guluronate, G, fraction (G/[M + G]) values, and the r parameter from Halsey model. Extrapolating this linear regression to guluronate fraction values of zero and one (G = 0 and M = 0), it could be possible to obtain r values corresponding to hypothetical alginate formed exclusively by M (rM = 2.074) and by G (rG = 0.560), respectively. The water content of alginate can be assumed as the sum of water adsorbed on guluronate (XG) and mannuronate (XM) surface (Equations [9] and [10]). where aG and aM are the corresponding parameters for G and M of Halsey model. The aG and aM values were obtained by means of a multivariable optimization procedure with the following objective, Equation (11): where X were the moisture content values given by Halsey model. This procedure was applied to each alginate at constant temperature. As chemical characteristics of M and G are independent of the type of alginate, average aG and aM values were obtained after individual optimization of each alginate. The aG and aM values were 0.30, 0.26, and 0.24 and 0.052, 0.035, and 0.019 at 20, 37, and 50°C, respectively. (9) XG=G ( −aG lnaw)1 rG (10) X M=(1−G) ( −aM lnaw)1 rM (11) min[ aw=0.4 ∑ aw=0.1 X−G ( −aG lnaw )1 rG−(1−G) ( −aM lnaw )1 rM ] FIGURE 4 Effectofmoisturecontent,X, on the integral enthalpy, qin, (a) and on the integral entropy, Sin, (b) for alginates: Alginate from sigma (S), alginate from PanReac (P), and processed alginate (50D). 0 4 8 12 16 00.2 0.40.6 0.8 q in (kJ/mol) X(kg water/kg d.b.) (a) S P 50D -60 -40 -20 0 20 40 00.2 0.40.6 0.8 S in (J/mol K) X (kg water/kg d.b.) (b) S P 50D 17454549, 2022, 12, Downloaded from https://ifst.onlinelibrary.wiley.com/doi/10.1111/jfpp.17179 by Universidade de Santiago de Compostela, Wiley Online Library on [15/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | 7 of 8 MONTES et al. These values were linearly correlated (R2 > 0.99) with temperature. Therefore, an equation useful to estimate the equilibrium moisture content of alginate with its guluronate and mannuronate content at different temperatures is given by Equation (12): Analyzing Equation (12), it was observed that the X value was mainly due to the contribution of XM meaning that M was more hygroscopic than G. For instance, in S alginate at 20°C and at aw = 0.2, the contribution of XMtothetotaladsorbedwaterwas69.7%and atthesameconditionsachieved73.7%and84.2%inthecaseofP and 50D alginates. The XM contribution decreased with increasing aw (i.e., at aw =0.4,59.1,63.5and76.4%forS,Pand50Dalginates,respectively). As example, Figure 5 shows the XM and XG contributions to the total equilibrium moisture content of 50D alginate at 20 and 50°C. An acceptable agreement can be observed between the water sorption isotherm fitted by Halsey model and the proposed model given by Equation (12). Shimanuki et al. (2020) demonstrated that, at very low moisture content, helical structures formed by short molecular chains present in alginates are composed exclusively of G blocks. Therefore, under these conditions, the surface of G units was not completely available to adsorb water so easily as M units. The 50D alginate was the tested sample with the highest amount of M units (M/G = 1.21) and S the lowest (M/G = 0.91), but P showed a M/G (1.15) closer 50D. It seems that exclusively the relative amount of G is not enough to explain these differences. Nevertheless, S alginate contained a greater number of diads FGG (0.35) and triads FGGG (0.30) than P (0.23 and 0.13) and 50D (0.20 and 0.10) alginates. This fact could explain the existence of more helical structures formed by G blocks in S alginate and the contribution of XG to total adsorbed water consequently decreased. Helical structures were progressively disappearing with water adsorption (higher aw values) and the contribution of XG increased with the creation of new available surfaces. 4 | CONCLUSIONS No significant differences were found between tested commercial alginates, and processed alginate was the most hygroscopic one. Halsey model was chosen to fit adequately for describing experimental desorption isotherm data for several alginates in the temperature range from 25 to 50°C. The differential enthalpy and entropy of sorption for all samples decreased and increased, respectively, exponentially with increasing moisture content to 0.15–0.20 kg/kgd.b.,thendecreasedslowlytonearzeroathigher moisture content, due to a decrease of binding energies between water molecules and sorption sites with increasing moisture content. Integral enthalpy and entropy showed respective maximum andminimumvaluesintheintervalfrom0.15to0.20 kg/kgd.b. Alginates must be dried up to this moisture content to achieve optimum stability during storage. A model based on the structural features of alginate was proposed and satisfactorily tested to predict the equilibrium moisture content of alginates at low water activity values (<0.4). Mannuronate, M, was more hygroscopic than guluronate, G, and consequently, a higher amount of M in the alginate increases its hygroscopicity. Nevertheless, the presence of helical structures formed by G blocks must be also considered in a more complex structural model. AUTHOR CONTRIBUTIONS Leticia Montes: Conceptualization; Data curation; Formal analysis; Investigation; Validation; Methodology; Writing— original draft. Mauro Gisbert: Formal analysis; Validation; Writing— review & editing. Ramón Moreira: Conceptualization; Methodology; Writing— review & editing; Project administration; Supervision; Funding acquisition. FUNDING INFORMATION Authors acknowledge the financial support of the Spanish Ministry ofScienceandInnovation(ProjectRTI2018-095919-B-C2)andthe European Regional Development Fund (FEDER) and Xunta de Galicia (Consolidation Project ED431B 2019/01). CONFLICT OF INTEREST The authors declared no conflicts of interest for this article. DATA AVAILABILITY STATEMENT Data available on request due to privacy/ethical restrictions. ORCID Leticia Montes https://orcid.org/0000-0001-9758-4142 Mauro Gisbert https://orcid.org/0000-0001-5923-4651 Ramón Moreira https://orcid.org/0000-0002-6388-0063 (12) X=XG+XM=G(−0.9616+0.002246 T lnaw) 1 0.5604 +(1−G) ( −0.3772+0.001107 T lna w) 1 2.0742 FIGURE 5 Watersorptionisothermsof50Dalginateat20and 50°C estimated by Halsey model and Equation (12) (XM + XG) at low water activity, aw, with the contributions of M (XM) and G (XG) fractions. 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.05 0.15 0.25 0.35 0.45 X(kg water/ kg d. b.) a w 20 ºC Halsey 20 ºC XG 20 ºC XM 20 ºC XM + XG 50 ºC Halsey 50 ºC XG 50 ºC XM 50 ºC XM + XG 17454549, 2022, 12, Downloaded from https://ifst.onlinelibrary.wiley.com/doi/10.1111/jfpp.17179 by Universidade de Santiago de Compostela, Wiley Online Library on [15/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 8 of 8 | MONTES et al. REFERENCES AbkaKhajouei, R., Tounsi, L., Shahabi, N., Patel, A. K., Abdelkafi, S., & Michaud, P. (2022). Structures, properties and applications of alginates. Marine Drugs, 20, 364. https://doi.org/10.3390/md200 60364 Adamczak, M. I., Martinsen, Ø. G., Smistad, G., & Hiorth, M. (2017). Polymer coated mucoadhesive liposomes intended for the management of xerostomia. International Journal of Pharmaceutics, 527, 72–78.https://doi.org/10.1016/j.ijpha rm.2017.05.032 Brunauer, S., Deming, L. S., Deming, W. E., & Teller, E. (1940). On theory of the van der Waals adsorption of glass. Journal of the American Chemistry Society, 62, 1723– 1732. https://doi.org/10.1021/ja018 64a025 Chee, S. Y., Wong, P. K., & Wong, C. L. (2011). Extraction and characterisation of alginate from brown seaweeds (Fucales, Phaeophyceae) collected from Port Dickson, peninsular Malaysia. Journal of Applied Phycology, 23, 191– 196. https://doi.org/10.1007/s1081 101095337 Fasina, O. O., Ajibola, O. O., & Tyler, R. T. (1999). Thermodynamics of moisture sorption in winged bean seed and gari. Journal of Food Process Engineering, 22, 405–418. https://doi.org/10.1111/ j.17454530.1999.tb004 96.x Fernando, I. P. S., Lee, W., Han, E. J., & Ahn, G. (2020). Alginatebased nanomaterials: Fabrication techniques, properties, and applications. Chemical Engineering Journal, 391, 123823. https://doi. org/10.1016/j.cej.2019.123823 Galus, S., & Lenart, A. (2013). Development and characterization of composite edible films based on sodium alginate and pectin. Journal of Food Engineering, 115, 459– 465. https://doi.org/10.1016/j.jfood eng.2012.03.006 Halsey, G. (1948). Physical adsorption on non-uniform surfaces. The Journal of Chemical Physics, 16, 931– 937. https://doi. org/10.1063/1.1746689 Kaya, K., & Kahyaoglu, T. (2007). Moisture sorption and thermodynamic properties of safflower petals and tarragon. Journal of Food Engineering, 78, 413– 421. https://doi.org/10.1016/j.jfood eng.2005.10.009 Koksharov, S. A., Aleeva, S. V., Lepilova, O. V., Krichevskii, G. E., & Fidorovskaya, Y. S. (2021). The properties of sodium alginate hydrocolloids upon sorption binding of papain. Colloid Journal, 83, 722– 736. https://doi.org/10.1134/S1061 933X2 1060077 Kurek, M., Guinault, A., Voilley, A., Galic, K., & Debeaufort, F. (2014). Effect of relative humidity on carvacrol release and permeation properties of chitosan based films and coating. Food Chemistry, 144, 9– 17. https://doi.org/10.1016/j.foodc hem.2012.11.132 Lee, K. Y., & Mooney, D. J. (2012). Alginate: Properties and biomedical applications. Progress in Polymer Science, 1, 106– 126. https://doi. org/10.1016/j.progp olyms ci.2011.06.003 Madamba, P. S., Driscoll, R. H., & Buckle, K. A. (1996). Enthalpyentropy compensation models for sorption and browning of garlic. Journal of Food Engineering, 28, 109– 119. https://doi. o r g / 1 0 . 1 0 1 6 / 0 2 6 0 - 8 7 7 4 ( 9 4 ) 0 0 0 7 2 - 7 Montes, L., Gisbert, M., Hinojosa, I., Sineiro, J., & Moreira, R. (2021). Impact of drying on the sodium alginate obtained after polyphenols ultrasoundassisted extraction from Ascophyllum nodosum seaweeds. Carbohydrate Polymers, 272, 118455.https:// doi.org/10.1016/j.carbpol.2021.118455 Moreira, R., Chenlo, F., Sineiro, J., Sánchez, M., & Arufe, S. (2016). Water sorption isotherms and air drying kinetics modelling of brown seaweed Bifurcaria bifurcata. Journal of Applied Phycology, 28,609–618. h t t p s : / / d o i . o r g / 1 0 . 1 0 0 7 / s 1 0 8 1 1 - 0 1 5 - 0 5 5 3 - 1 Moreira, R., Chenlo, F., & Torres, M. D. (2009). Simplified algorithm for the prediction of water sorption isotherms of fruits, vegetables and legumes based on chemical composition. Journal of Food Engineering, 94, 334– 343. https://doi.org/10.1016/j.jfood eng.2009.03.026 Moreira, R., Chenlo, F., Torres, M. D., & Prieto, D. M. (2017). Statistical criteria for modelling of water desorption isotherms of sugars. Estimation of sucrose hygroscopic properties from glucose and fructose data. Advances in Food Science and Engineering, 1,18–27. https://doi.org/10.22606/ afse.2017.11003 Moreira,R.,Chenlo,F.,Torres,M.D.,&Vallejo,N.(2008).Thermodynamic analysis of experimental sorption isotherms of loquat and quince fruits. Journal of Food Engineering, 88, 514– 521. https://doi. org/10.1016/j.jfoodeng.2008.03.011 Polachini, T. C., Betiol, L. F. L., LopesFilho, J. F., & TelisRomero, J. (2016). Water adsorption isotherms and thermodynamic properties of cassava bagasse. Thermochimica Acta, 632, 79–85. https://doi. org/10.1016/j.tca.2016.03.032 Qin,Y.,Jiang,J.,Zhao,L.,Zhang,J.,&Wang,F.(2018).Applicationsofalginate as a functional food ingredient. In A. M. Grumezescu & A. M. Holban (Eds.), Biopolymers for food design (pp. 409– 429). Academic Press. h t t p s : / / d o i . o r g / 1 0 . 1 0 1 6 / B 9 7 8 - 0 - 1 2 - 8 1 1 4 4 9 - 0 . 0 0 0 1 3 - X Shimanuki, N., Imai, M., & Nagai, K. (2020). Effects of counter cations on the water vapor sorption properties of alginic acid and alginates. Journal of Applied Polymer Science, 137, 49326. https://doi. org/10.1002/app.49326 Shivhare, U. S., Arora, S., Ahmed, J., & Raghavan, G. S. V. (2004). Moisture adsorption isotherms for mushroom. LWTFood Science and Technology, 37, 133– 137. https://doi.org/10.1016/S0023 - 6 4 3 8 ( 0 3 ) 0 0 1 3 5 - X Xiao, Q., & Tong, Q. (2013). Thermodynamic properties of moisture sorption in pullulansodium alginate based edible films. Food Research International, 54, 1605– 1612. https://doi.org/10.1016/j.foodr es.2013.09.019 Zhang, H., Bai, Y., Zhao, X., & Duan, R. (2016). Water desorption isotherm and its thermodynamic analysis of glutinous rice flour. American Journal of Food Technology, 11, 115– 124. https://doi.org/10.3923/ ajft.2016.115.124 How to cite this article: Montes, L., Gisbert, M., & Moreira, R. (2022). Water sorption isotherms of different sodium alginates: Thermodynamic evaluation and influence of mannuronateguluronate copolymers. Journal of Food Processing and Preservation, 46, e17179. https://doi. org/10.1111/jfpp.17179 17454549, 2022, 12, Downloaded from https://ifst.onlinelibrary.wiley.com/doi/10.1111/jfpp.17179 by Universidade de Santiago de Compostela, Wiley Online Library on [15/06/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License