Effective thermal conductivity prediction of foods using composition and temperature data

• Methods for predicting thermal conductivity of foods from composition and temperature were compared against measured data.
• Multi-step procedure involving sequential application of Parallel, Levy and Maxwell-Eucken model was most accurate method.
• Other than for frozen, porous foods, the Co-continuous models also provided predictions within ±20% on average.

Thermal conductivity data are important for food process modelling and design. Where reliable thermal conductivity data are not available, they need to be predicted. The most accurate ‘first approximation’ methodology for predicting the isotropic thermal conductivity of foods based only on data for composition, initial freezing temperature and temperature dependent thermal conductivity of the major food components was sought. A key feature of the methodology was that no experimental measurements were to be required. A multi-step prediction procedure employing the Parallel, Levy and Effective Medium Theory models sequentially for the components other than ice and air, ice and then air respectively is recommended. It was found to provide the most accurate predictions over the range of foods considered (both frozen and unfrozen, porous and non-porous). The Co-Continuous model applied in a single step also provided prediction accuracy within ±20% (on average), except for the porous frozen foods considered. For greater accuracy more rigorous modelling approaches based on knowledge of the foods structure would be required.