ASSESSMENT of A SEMI-EMPIRICAL FOUR PARAMETER GENERAL MODEL FOR SIGMOID MOISTURE SORPTION ISOTHERMS1

Published sigmoid moisture sorption isotherms (0 < a_w < ～ 0.9) were fitted by the four parameter models m = k₁a_wⁿ¹+ k₂a_wⁿ² where m is the moisture contents (dry basis), a_w the water activity and the k and n values are constants (n₁ < 1 and n₂ > 1). Not surprisingly, the model had the same or better fit than the GAB model. In contrast with both the BET and GAB models, the proposed model is not based on the assumption that there exists a well-defined monolayer of absorbed water. At a_w < ～ 0.4 and n₁ > ～ 0.55, however, the model produces a practically linear a_w/[m(1 – a_w)] vs a_w plot, of the kind used to calculate the monolayer moisture with the BET model. The proposed model can be a convenient means to catalog both sigmoid and nonsigmoid isotherms, and used to calculate the equilibrium water activity of dry mixtures with equations solving software.     

THERMAL PROPERTIES and MOISTURE SORPTION ISOTHERMS of SPRAY-DRIED ENCAPSULATED MILKFAT

Stable emulsions containing 40 or 60% anhydrous butter oil and carbohydrate encapsulants (sucrose, modified starch or all-purpose flour) were spray dried to produce free-flowing shelf-stable powders, according to a 2 x 3 factorial design, replicated three times. Differential scanning calorimetry profiles showed well-defined melting ranges that were related to encapsulant used. Butter oil was almost completely encapsulated when fat content was 40% and sucrose was used as the wall material. Moisture sorption isotherms of powders with sucrose showed characteristic breaks caused by sugar crystallization followed by moisture desorption, whereas powders with modified starch or all-purpose flour continuously absorbed moisture with increasing relative humidity. the solvent-extractable fat fraction increased with increasing relative humidity in all cases. Scanning electron microscopy showed that sucrose-containing powder particles partially dissolved and fused together as a result of moisture uptake, whereas powders with modified starch or all-purpose flour maintained particle identity, even at 80% relative humidity.     

MOISTURE SORPTION CHARACTERISTICS OF TARHANA, A FERMENTED TURKISH CEREAL FOOD

The adsorption and desorption isotherms of three types of commercial tarhana, a popular Turkish cereal food used in soup-making, were determined as 10, 20 and 30C, and water activities (a_a) ranging from 0.08 to 0.92 using the gravimetric static method. The results showed that commercial tarhana has a type HI BET classification. Sieve size affected hygroscopicity. Hysteresis effects were not detected among the three formulations examined, but because of the limited number of formulations, that conclusion is only tentative. Moisture up-take accelerated around a_w 0.6.     

WATER ACTIVITY AND MOISTURE SORPTION ISOTHERMS OF GAZİANTEP CHEESE

Water activity of Gaziantep cheese prepared in the laboratory was determined as 0.95 by using the Proximity Equilibration Cell method. Adsorption and desorption isotherms were obtained at 12, 20 and 30C and 5, 20 and 27C, respectively. Halsey, Chung-Pfost, Guggenheim-Anderson-deBoer, and cubic and quadratic polynomial equations were used to evaluate the adsorption and desorption data. Parameter estimation was carried out. GAB and quadratic polynomial equations were the best equations for defining the adsorption and desorption data, respectively.     

MOISTURE SORPTION ISOTHERMS FOR KARINGDA (CITRULLUS LANATUS (THUMB) MANSF) SEED, KERNEL AND HULL

Adsorption-desorption behaviors of karingda (Citrullus lanatus (Thumb) Mansf) seed, kernel and hull for nine equilibrium relative humidities (ERH) ranging between 11 and 96% at temperatures of 10, 20, 30, 40 and 50C were studied following a static equilibriation technique using saturated solutions of various salts. Under both the adsorption and desorption processes, the equilibrium moisture contents (EMC) of the hull were found to be highest followed by those of the seed and the kernel at all the corresponding temperatures and ERH values. Analysis of these data using four sorption models (modified Henderson, modified Halsey, modified Chung-Pfost and Guggenheim-Anderson-de Boer) and taking the temperature dependence of the respective coefficient into consideration, it revealed that both the Chung-Pfost and the GAB models were acceptable in describing EMC-ERH relationships for karingda seed, kernel and hull over the entire range of temperatures. The excess heat of sorption of all the samples, estimated from the Clausius-Clapeyron equation, decreased exponentially with the increase in moisture content of the same.     

ESTIMATION of HEAT of MOISTURE SORPTION and IMPROVED CRITERIA FOR EVALUATING MOISTURE SORPTION ISOTHERM EQUATIONS FOR FOODS

Heat of sorption of moisture on foods can be estimated from sorption isotherm equations having both moisture and temperature as independent variables. Sorption thermodynamics dictate that the heat of moisture sorption should decrease with increased moisture concentration, or increased temperature of food, provided no biological, chemical or other changes occur during sorption. This fact, and the Clausius-Clapeyron equation were used to derive two analytical formulae to evaluate the reliability of moisture sorption equations in the calculation of heat of sorption. As examples, Chung and Pfost, modified Harkins and Jura, modified Henderson, and Guggenheim-Anderson-De Boer equations at specified conditions were analyzed.     

MOISTURE ADSORPTION ISOTHERMS AND ISOSTERIC ENERGY FOR ALMOND

The equilibrium moisture contents of almond were determined using the gravimetric‐static method at 15, 30, 55 and 75C for powder and 15, 55 and 75C for nut state of almond for water activity (a_w) ranging from 0.11 to 0.87. At a given a_w, the results show that the moisture content decreases with increasing temperature. The experimental sorption curves are then described by the BET, GAB, Henderson, Oswin, Smith and Halsey models. A nonlinear regression‐analysis method was used to evaluate the constants of equations. The GAB model was found to be the most suitable for describing the sorption curves; the monolayer‐content values for the sorption at different temperatures are calculated. Also, the isosteric heats of adsorption of water were determined as a function of moisture content from the equilibrium data at different temperatures using the Clasius‐Clapeyron equation.     

SORPTION ISOTHERMS OF TURKISH DELIGHT

Moisture sorption isotherms of Turkish delight were determined using the gravimetric static method of saturated salt solutions at 10, 20 and 30C. Isotherms were found to be of type III, typical of high sugar foods. The effect of temperature on moisture content was not significant (P>0.05). The sorption isotherms exhibited hysteresis at low water activities (a_w < 0.5). At higher water activities the moisture content increased sharply as the temperature was increased, resulting in crossing of the isotherm curves at 0.65 water activity. Six models namely the BET, the GAB, the Halsey, the Henderson, the Chung & Pfost and the Iglesias & Chirife were evaluated to determine the best fit for the experimental data. The GAB and the Iglesias & Chirife models fitted well the data of Turkish delight in the temperature and water activity range investigated. However, the GAB model was not appropriate for the estimation of monolayer value. The Clausius-Clapeyron equation was used to examine the isosteric heats of sorption.     

A PREDICTIVE MODEL FOR THERMAL CONDUCTIVITY OF AN INTERMEDIATE MOISTURE GRANULAR FOOD

A mathematical model for prediction of thermal conductivity of granular foods was developed to consider the influence of moisture content and bulk density. The range of moisture contents considered was from approximately 20% (db) to 150% (db). The model considers the influence of thermal conductivity and volume fraction of product solids and water as a first step in predicting thermal conductivity of product particles. The second step in the predictive model considers thermal conductivity and volume fraction of the product particles and air when predicting effective thermal conductivity of the food system. The model was evaluated by comparison of model outputs with experimental measurements for casein and corn meal starch particles. Experimental measurements of thermal conductivity were obtained using a thermal conductivity probe. The model provided excellent prediction of effective thermal conductivity over the entire moisture content range. Statistical agreement was described by standard errors of 0.0143 W/mK (7.2%) for starch at 25C, 0.01835 W/mK (12.1%) for casein at 25C, and 0.02584 W/mK (16.4%) for casein at 40C, with most of the variability associated with experimental measurements.     

VARIATIONS IN FLEXURAL AND COMPRESSIVE FRACTURE BEHAVIOR OF A BRITTLE CELLULAR FOOD (DRIED BREAD) IN RESPONSE TO MOISTURE SORPTION

Mechanical properties of dried bread (a cellular baked product) equilibrated at different water activities (A_w) were measured using three-point bending and uniaxial compression to compare their responses to moisture sorption. The glass-to-rubber transition was clearly demarcated in all the mechanical property-A_w curves, although there were differences in critical A_w (0.32–0.56) at which the dramatic changes in mechanical properties occurred. The role of water appeared to be strictly that of a plasticizer where flexural mechanical parameters were concerned, leading to reduced modulus and fracture stress but increased fracture strain with increasing A_w. Uniaxial compression tests revealed moisture-induced mechanical antiplasticization effects on the material in the glassy state which resulted in maximum compressive fracture stress but minimum fracture strain over the A_w range from 0–0.56. Compressive modulus apparently was not affected much by moisture sorption up to an A_w of 0.43, above which it decreased sharply.     

ELECTRIC SENSOR FOR PROMPT MEASUREMENT OF MOISTURE CONTENT IN SOLID FOOD PRODUCTS

Originally designed sensor for electrical measurement of moisture content in comminuted food was presented. Equations were formulated to demonstrate the dependence of tested sample electrical resistance on resistivity (specific resistance) of food as well as the dependence of investigated sample electrical capacitance on permittivity of food. Experiments were carried out to determine moisture content in wheat flour, potato starch, and raw pork loin serving as examples. The results of these measurements were compared with the results obtained using conventional gravimetry. The measurement of moisture content lasted for between 3 to 5 s, and the time used for a sample preparation was not longer than 60 s. An advantage of the designed sensor of moisture content is that the density of the investigated food sample does not affect the measurement results.