Characterisation of physical changes, fluidisation properties and drying behaviour of large particulate food materials during fluidised bed drying

Abstract

This study extended the knowledge and understanding of the physical changes, fluidisation properties and drying behaviour of large food particulate during fluidised bed drying. The heat pump dehumidifier system was used to supply low temperature drying air suitable for the drying of food particulates.Food particles come in a variety of shapes and sizes, and are often comprised of larger particles of irregular geometry. In this study natural food materials were selected to make regular particles needed for experimentation. This provided sufficient control of shape and size to enable results to be interpreted, while being able to observe to changes that take place in real food materials. Three geometrical shapes cylindrical, parallelepiped and spherical were taken into consideration. Green beans, potato and peas were chosen as the representative food materials respectively to make required shapes. In the case of beans three length:diameter ratios 3:1, 2:1 and 1:1 were considered. For potato three aspect ratios 3:1, 2:1 and 1:1 were taken. In the case of peas only one size spherical particles were used.For the fluidisation study at various levels of moisture contents, the drying experiments were conducted at 50 p 2o C, with 15 p 5 % relative humidity and a hot air velocity of 3 m/s using a heat pump dehumidifier system. Fluidisation experiments were undertaken for the bed heights of 100, 80, 60 and 40 mm and at 10 moisture levels. Minimum fluidisation velocity for each height and moisture content was observed.Minimum fluidisation velocity decreased with the decreasing moisture content as well as with decreasing bed height. Empirical relationships were developed for the change of minimum fluidisation velocity with moisture content during drying. The Generalized equation was also used to calculate minimum fluidisation velocity.The minimum fluidisation velocity (Umf) of green beans with change in moisture content was predicted with an empirical model Umf = A + B e -Cm with a satisfactory fit for L:D = 1:1. This model could not be applied to L:D = 2:1 and 3:1 due to irregularities in the variation of minimum fluidisation velocity with the reduction in moisture. The calculated Umf using the Generalized equation based on the dimensional changes of the beans during drying can be applied to predict minimum fluidisation velocity for all L:D ratios with a reasonable accuracy. The minimum fluidisation velocity of potato could not be fitted to any available empirical model satisfactorily. This was due to the irregular behaviour of minimum fluidisation velocity with change in moisture content. However, the general trend was the reduction in Umf as the moisture content decreased. The Umf predicted using the Generalized equation was valid only for aspect ratios of 1:1 and 2:1. Minimum fluidisation velocity behaviour of peas during drying was best fitted with linear equation of the form Umf = A + B m for all the bed heights. It was found that, Umf predictions for peas were success with both the Generalized equation and the conventional Ergun equation.The terminal velocity of the beans, potato and peas were calculated and were found to be well above the minimum fluidisation velocity (nearly eight times) at every moisture content during drying. Hence, operational velocity of the fluidised bed could be kept at a constant value for whole drying period.To undertake study on the physical changes during drying, the drying experiments were conducted at 50o C, 40o C and 30o C with 15 p 5% RH in a batch fluidised bed dryer. Hot air was supplied by a heat pump dehumidifier system coupled to the fluidised bed dryer. Empirical models were developed for shrinkage, bulk density, particle density and bed porosity during fluidised bed drying.Shrinkage was examined during drying relating volume ratio to moisture ratio using a linear model of the form VR = A + B MR for potato and peas. In contrast to the peas and potato, beans shrinkage was modelled using a non-linear exponential of the form VR = 1- B e -KMR. This exponential behaviour was attributed to the open heterogenous structure of beans. The shrinkage rate constant K in beans decreased with the increased drying temperature. The effect of L:D ratio was not significant. The constant B was similar for all drying temperatures and for all L: D ratios. In the case of potato none of A and B parameters in the shrinkage equation were significantly different with change in temperature or size.n n n