Application of the Guggenheim, Anderson and De Boer model to correlate water activity and moisture content during osmotic dehydration of apples

Abstract

Water activity and moisture content are the most valuable characteristics for assessing the stability of dried foods. The aim of this work was to evaluate if a desorption isotherm model could be used to describe the relationship between the water activity and moisture content in the product during osmotic dehydration. This model can lead to a better understanding of how water activity of a product can be reduced during osmotic dehydration. The Guggenheim, Anderson and De Boer (GAB) model was chosen because it has been proven to best fit the desorption and adsorption of foods over a large range of water activities. The studied material was apple pieces, and the osmotic medium with a water activity of 0.939 was prepared with sucrose. The osmotic dehydration process was studied by collecting moisture content and water activity data at different temperatures and times. The model was shown to fit well the experimental points at temperatures 25, 45 and 55 °C. At 65 °C, however, the model is satisfactory if a_w is below 0.970 but not above. The effect of temperature can be expressed on the C-parameter of the GAB model.

Introduction

In the context of minimal processing, the measurement and prediction of water activity provide the best available tool for evaluating the stability of foods. The sorption isotherm, consisting of a graphic representation of water activity (a_w) against moisture content at constant temperature, is a common way of presenting the relationship between these two parameters (McLaughlin & Magee, 1998; Rahman & Labuza, 1999). Desorption isotherms are of particular importance in the design of a food dehydration process, especially in the determination of a drying end point. The end point of drying is the residual moisture content of the final product which ensures economic viability and microbiological safety, i.e. a water activity value lower than 0.60. Desorption isotherms focus therefore on low moisture or intermediate moisture zones, because this is the range of moisture content of the food product after drying methods such as air-drying and freeze-drying. The effect of temperature on moisture sorption isotherms has also been reported in many studies (Velásquez de la Cruz, Torres, & Martı́n-Polo, 2001; Wang & Brennan, 1991).

Numerous mathematical models have been proposed for the study of both adsorption and desorption of foods, such as the GAB, Iglesias and Chirife, BET, Oswin, and Ferro Fontán models (Barbosa-Canovas & Vega-Mercado, 1996; Crapiste & Rotstein, 1982; Gekas, 1992; Wang & Brennan, 1991). There is a lack of knowledge on how the overall water activity of cellular material is evaluated near the full turgor state (Crapiste & Rotstein, 1982). The Guggenheim–Anderson–de Boer (GAB) model is reported to be the best for fitting sorption isotherm data for the majority of food products up to a_w levels of approximately 0.9 (Barbosa-Canovas & Vega-Mercado, 1996; Timmermann, Chirife, & Iglesias, 2001; Tsami, Krokida, & Drouzas, 1999). At a_w>0.9, which is the region where cellular material is still in a quasi-turgor state, the water activity was seldom accurately measured or studied.

Osmotic dehydration is characterized by a more complex mass transfer process than air-drying. The transport of water out of the tissue is completed by a counter-current diffusion of the osmotic agent from the solution toward the tissue. These two simultaneous transports have a depressing effect on the a_w of the samples. Osmotic processes are often evaluated in terms of water loss (WL) and solid gain (SG) (Bolin, Huxsoll, Jackson, & Ng, 1983; Erba, Forni, Colonello, & Giangiacomo, 1994; Garrote, Silva, & Bertone, 1992; Hawkes & Flink, 1978). Mathematic models have been reported in the literature to describe the water transport during osmotic drying (Azuara, Beristain, & Garcia, 1992; Fito & Chiralt, 1997; Panagiotou, Karathanos, & Maroulis, 1998; Toupin, Marcotte, & Le Maguer, 1989). However, only few studies aim at representing kinetics of water activity changes (Hough, Chirife, & Marini, 1993; Monsalve-Gonzalez, Barbosa-Cánovas, & Cavalieri, 1993). Marcotte and Le Maguer (1991) developed a model based on the equality of a_w between the osmotic solution and potato material.

A mathematical model, based on the principles of desorption isotherms, can be a suitable way to predict the a_w of a product during osmotic dehydration using the moisture content data. A reduction of the experimental work can therefore be reached. In addition, this model can be included in the mathematical models for description of water transport during dehydration allowing the direct prediction of the water activity in the product.

The aim of this study was to evaluate if the GAB model can be used (1) to describe the relationship between a_w and moisture content in the particular case of osmotic dehydration and (2) to study the effect of temperature on the parameters of this model.