Composition‐Based Prediction of Temperature‐Dependent Thermophysical Food Properties: Reevaluating Component Groups and Prediction Models

• Autores: David Martin Phinney, John C. Frelka, Dennis Ray Heldman
• Localización: Journal of food science, ISSN 0022-1147, Vol. 82, Nº 1, 2017, págs. 6-15
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • Prediction of temperature‐dependent thermophysical properties (thermal conductivity, density, specific heat, and thermal diffusivity) is an important component of process design for food manufacturing. Current models for prediction of thermophysical properties of foods are based on the composition, specifically, fat, carbohydrate, protein, fiber, water, and ash contents, all of which change with temperature. The objectives of this investigation were to reevaluate and improve the prediction expressions for thermophysical properties. Previously published data were analyzed over the temperature range from 10 to 150 °C. These data were analyzed to create a series of relationships between the thermophysical properties and temperature for each food component, as well as to identify the dependence of the thermophysical properties on more specific structural properties of the fats, carbohydrates, and proteins. Results from this investigation revealed that the relationships between the thermophysical properties of the major constituents of foods and temperature can be statistically described by linear expressions, in contrast to the current polynomial models. Links between variability in thermophysical properties and structural properties were observed. Relationships for several thermophysical properties based on more specific constituents have been identified. Distinctions between simple sugars (fructose, glucose, and lactose) and complex carbohydrates (starch, pectin, and cellulose) have been proposed. The relationships between the thermophysical properties and proteins revealed a potential correlation with the molecular weight of the protein. The significance of relating variability in constituent thermophysical properties with structural properties––such as molecular mass––could significantly improve composition‐based prediction models and, consequently, the effectiveness of process design.