Moisture profile analysis of food models undergoing glass transition during air-drying

Highlights

• Application of Fick's law was judged by fitting parabolas to MRI moisture profiles.

• Water profiles were well characterised by Fick's law up to middle stages of drying.

• Dry layer development provoked non-ideal moisture profiles at the end of drying.

• Glass transition effect on dry layer formation during drying was found negligible.

Abstract

Moisture profiles evolution of a cellular food product (potato) and two non-cellular food models made with maltodextrins (DE19 and DE36) were investigated during air-drying at two different temperatures, 25 °C and 55 °C. MRI results of suggested a non-uniform moisture distribution within potato, whilst the non-cellular food models had a more uniform profile. As expected, temperature had an impact on the drying kinetics and on the pore formation. While temperature had no impact on dry layer development for potato, it had a major impact on dry layer development for non-cellular food models. The fact that the two non-cellular food models used in this study have different glass transition temperatures but presented the same dry layer development curve does not support the hypothesis that glass transition is the driver mechanism for dry layer formation. The applicability of Fick's law was assessed by fitting experimental moisture profile data to quadratic polynomials. Fick's law provided a fairly accurate estimation of the water profiles of materials having different glass transition temperatures only when the dry layer at the surface was not taken into account, and more specifically during the middle stages of drying.

Introduction

Air-drying is frequently used to reduce food moisture content for preservation. However, it is a complicated phenomenon of simultaneous mass and heat transports (Krokida et al., 2003, Ratti, 2001). Understanding the basic relationships prevailing during moisture transfer, such as moisture profiles, may lead to better process control (Ruan et al., 1991), which is essential to improve the quality of the product, including physical, chemical and/or biological characteristics, and also the process economics (Ratti, 2001, Ruan et al., 1991).

Glass transition temperature, Tg, is defined as the temperature at which an amorphous system changes from glassy to rubbery states (Roos and Karel, 1991). It depends on water content, molecular weight, crystallinity and composition of material (Thiewes and Steeneken, 1997). Many foods undergo glass transition in the range of operation conditions used for air-drying, such as potato and cassava starches (Farahnaky et al., 2009), banana (Katekawa and Silva, 2007) and papaya (Kurozawa et al., 2012). Some authors have suggested, based on experimental observations or mathematical simulations, that glass transition is responsible for the formation of a dry shell at the surface of the product leading to case-hardening phenomena, frequently observed during air-drying, particularly with fast drying rates. In other words, it was suggested that shell formation cannot occur if drying conditions do not permit a glass transition (Mayor and Sereno, 2004). Case-hardening is the phenomenon in which the external surface of the material dries out much faster than the center, leading to an extremely dry layer on the outer zone (Achanta et al., 1997, Gulati and Datta, 2015). This phenomenon can occur during the last minutes of the drying process (Lozano et al., 1983, Wang and Brennan, 1995) or throughout the drying process (Fang et al., 2009, Gulati and Datta, 2015). According to Gulati and Datta (2015), case-hardening increased residual stresses in dehydrated food, causing numerous phenomena including the deviation behaviors of shrinkage and porosity and cracking of dry material surface. Nguyen et al. (2018b) found that the impact of glass transition on shrinkage of carrot and potato samples is negligible and could be considered a minor phenomenon affecting non-ideal shrinkage during air-drying of cellular food systems.

It is generally accepted that diffusion is the overall mass transport mechanism during air-drying and thus, Fick's law representing the random movement of molecules from higher to lower concentration zones is commonly applied. When this constitutive relationship is employed in a mass balance, such as when it is applied to cylindrical coordinates, the following equation is obtained (Crank, 1975):

$$\frac{\partial C}{\partial t}=\frac{1}{r}\frac{\partial }{\partial r}\left(rD\frac{\partial C}{\partial r}\right)$$

where C is concentration (kg m⁻³); D, diffusion coefficient (m² s⁻¹); r, radial coordinate (m); and t, time (s). The analytical and graphical solutions of Eq. (1) for uniform initial concentration and constant surface concentration can also be found in Crank (1975). The analytical solution is a rather complicated relationship of concentration with time and radial position, in terms of an infinite series of exponential and Bessel functions. The graphical solution, however, is simple, resulting in a round profile (Rizvi, 2005) that could be flawlessly represented by a second order polynomial since the concentration profiles are parabolic for constant Fourier numbers.

Numerous authors suggest that moisture diffusion during air-drying follows Fick's law only when the material is in rubbery state, but not through the transition or in glassy state. Fluid transport becomes non-Fickian when material undergoes glass transition (Kim et al., 1996, Thomas and Windle, 1982, Xing et al., 2007). According to Kim et al. (1996) when glass transition occurs, discontinuous stress profiles are formed. The polymer stress gradient provides negative driving forces to the solvent transport against the chemical potential gradient, causing non-Fickian transport behavior (Kim et al., 1996). Xing et al. (2007) suggested that a sharp shape of the moisture profile of material indicated non-Fickian transport during air-drying at the glass transition regime. It is likely that the shell layer at the surface of the product during air-drying forms a barrier, suggesting an explanation for the anomalous moisture transport (Xing et al., 2007), which could be related to glass transition, high drying rates, structure or composition (Jin et al., 2012, Mayor and Sereno, 2004, Xing et al., 2007).

Most studies of moisture transport in foods use experimental data based on average moisture content, such as the characterization of food drying from the average drying rate only. Average drying data are often insufficient to allow investigation of the underlying physics of moisture transport, particularly in heterogeneous systems such as foods (McCarthy et al., 1994). To study the effect of glass transition on drying kinetics, it is important to determine the moisture profiles, i.e. water distribution across the cross-section in the food during drying (Xing et al., 2007). According to Achanta et al. (1997), moisture profiles could be used to study dry shell development at the surface of the material where sudden moisture gradients are observed during drying (Achanta et al., 1997). The formation of a dry shell at the surface of microcrystalline cellulose-agar-water gels during drying was confirmed experimentally through MRI by Schrader and Litchfield (1992).

The purpose of the present work was to investigate the effect of glass transition on the development of moisture profiles in gel food systems (non-cellular) and potato (cellular) for different drying conditions. Determination of sample shell formation during air-drying will elucidate phenomena occurring in food materials undergoing glass transition. Applicability of Fick's law to explain mass transport phenomena during air-drying will be analyzed through simple second polynomial fitting of moisture profiles.