Glassy State Transition and Rice Drying: Development of a Brown Rice State Diagram

Cereal Chem. 77(6):708–713 The effect of moisture content (MC) on the glass transition temperature (Tg) of individual brown rice kernels of Bengal, a medium-grain cultivar, and Cypress, a long-grain cultivar, was studied. Three methods were investigated for measuring Tg: differential scanning calorimetry (DSC), thermomechanical analysis (TMA), and dynamic mechanical analysis (DMA). Among these methods, TMA was chosen, because it can also measure changes in the thermal volumetric coefficient (β) of the kernel during glass transition. TMA-measured Tg at similar MC levels for both cultivars were not significantly different and were combined to generate a brown rice state diagram. Individual kernel Tg for both cultivars increased from 22 to 58°C as MC decreased from 27 to 3% wb. Linear and sigmoid models were derived to relate Tg to MC. The linear model was sufficient to describe the property changes in the MC range encountered during rice drying. Mean β values across both cultivars in the rubbery state was 4.62 × 10/°C and was higher than the mean β value of 0.87 × 10/°C in the glassy state. A hypothetical rice drying process was mapped onto the combined state diagram generated for Bengal and Cypress. In the United States, rice is usually harvested at ≈16–22% moisture content (MC) (wb) and must be dried to ≈12% MC for safe storage. Drastic drying conditions and subsequent exposure of the dried kernels to air at high relative humidity can increase the number of fissured kernels (Kunze and Prasad 1978, Kunze 1979, Sharma and Kunze 1982). Proposed theories on fissure formation are based on the response of rice when subjected to tensile and compressive stresses due to the existence of a moisture gradient within the kernel (Kunze and Choudhury 1972, Kobayashi et al 1976, Yamaguchi et al 1980). The effect of water on rice thermal and physical properties is a key in understanding the drying process (Wratten et al 1969, Arora et al 1973, Morita and Singh 1979, Steffe and Singh 1980, Muthukumarappan et al 1992). Wratten et al (1969) showed that the physical properties (length, width, thickness, volume, density, and specific gravity) and thermal properties (specific heat and thermal conductivity) of longand medium-grain rice were linear functions of MC. Morita and Singh (1979) also found that short-grain rough rice kernel dimensions, bulk density, and specific gravity varied linearly with MC at 26°C. Short-grain rice shrunk by an average of 12.3% when dried from 30 to 15% MC (Steffe and Singh 1980). Rough rice showed lower coefficients of volumetric expansion than brown and milled rice at room temperature. The volumetric reduction of the rice hull was significantly lower than that of the other parts of the rice kernel. Rice properties are also affected by kernel temperature. Arora et al (1973) reported an increase in milled rice thermal expansion at >53°C and showed that more rice kernels fissure above this transition temperature. Muthukumarappan et al (1992) measured volumetric changes in long-grain rice during moisture desorption and adsorption. In their measurements, the coefficients of linear hygroscopic expansion were greater across grain thickness than with either kernel length or width during adsorption and desorption. During desorption, brown rice had a greater coefficient of linear expansion compared with rough and milled rice. However, milled rice had a higher coefficient of cubical thermal expansion than rough and brown rice. The rate of thermal expansion of rice was uniform at ≤58°C. However, Muthukumarappan et al (1992) assumed that thermal expansion at >60°C was not important, and they used a single rate in their model for rice drying. Integrating the effects of water and temperature on material property changes is important in understanding fissure formation during grain drying. Polymer science has been applied in studying the effect of temperature and MC changes during processing on food components such as starch and protein (Slade and Levine 1991a,b, 1993, 1995; Levine and Slade 1992; Kokini et al 1994; Roos 1995; Madeka and Kokini 1996; Strahm 1998). In a review of rice milling, Rhind (1962) suggested that starch, the main constituent of rice, is brittle at <15% MC but is plastic at higher MC levels. In food polymer science terms, starch is considered a partially crystalline, partially amorphous polymer of glucose, whose properties change depending on temperature and MC (Slade and Levine 1991a,b, 1995). Within the starch granule, amylose and the branching points of amylopectin contribute to the amorphous phase, while the outer chains of amylopectin contribute to the crystalline phase. Changes in the crystalline region occur during processing at high MC (>25–28%) or high temperatures (>70°C). At lower MC and temperatures, conditions applicable during rice drying, most of the changes occur in the amorphous region. Physical and thermal properties of amorphous solids such as specific heat, specific volume, expansion coefficients, and elastic modulus change as the solid goes through a glass transition temperature (Tg) (White and Cakebread 1966). At temperatures below Tg, amorphous materials are glassy, with high viscosity and modulus of elasticity but low specific heat, specific volume, and expansion coefficient. At temperatures above Tg, these materials are rubbery, with much higher specific heat, specific volume, and expansion coefficient. Moisture content changes during processing affect the thermal properties of a food system. Water is a very effective plasticizer and will reduce Tg (as MC increases, Tg decreases) (Slade and Levine 1991a). Plotting Tg against corresponding MC will generate a state diagram that can be used to predict the mechanical properties of an amorphous solid at a particular temperature and MC (Levine and Slade 1992). The temperature of the material relative to Tg, which is governed by MC, will determine whether the material will be in the glassy solid or rubbery liquid state (Slade and Levine 1991b). State diagrams and the concept of glass transition have been applied to predict changes in food properties and reaction kinetics below and above Tg (Roos and Karel 1991; Slade and Levine 1991a, 1995; Nelson and Labuza 1994; Roos 1995; Peleg 1992, 1996). Mathematical models have been developed to predict drying and MC curves for grain and pasta (Fortes and Okos 1981a,b; Litchfield and Okos 1992; Achanta et al 1995). The diffusion model of Litchfield and Okos (1992) for pasta successfully predicted the 1 Published with the approval of the director, University of Arkansas Agricultural Experimental Station. 2 Former research assistant; professor and head, Food Science Department; and associate professor, respectively, Agricultural Statistics Laboratory, University of Arkansas, Fayetteville, AR. 3 Corresponding author. E-mail: [email protected] Publication no. C-2000-1010-03R. © 2000 American Association of Cereal Chemists, Inc.