Self-heating in spray dried detergents

• Main Author: Maxfield, Lewis
• Other Authors: Bayly, Andrew E. ; Hassanpour, Ali
• Published: University of Leeds 2018
• Online Access: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.758335

During the spray drying of detergent formulations, powder commonly accumulates on the inner walls of the spray drying tower. Under certain conditions, when these accumulations are large enough, self-heating can occur, whereby exothermic reactions within the accumulations cause an increase in the powder temperature within these layers. This can lead to unwanted charring and in severe cases to thermal runaway. This study aims to evaluate the methods for characterising the self-heating behaviour of these detergent powders. Firstly, two basket heating methods, namely the steady-state approach and cross-point temperature (CPT) method, were used to estimate the zero-order kinetics of the self-heating reaction of a typical detergent formulation. The resulting kinetics of these methods were not in agreement, with this being attributed to the CPT method’s sensitivity to errors. The kinetics estimated from these methods were used in a developed 2D-axismmetric transient model of heat and mass transfer within an oven heated basket of detergent powder. This was used to make temperature-time profile and critical ambient temperature (temperature above which thermal runaway occurs) predictions, and to simulate aspects of the basket heating experiments. A novel approach was developed using a combination of the basket experiments and the numerical model. This “parameter estimation approach” uses maximum likelihood estimations to estimate the required parameters. The model was fitted to experimentally measured temperature data, allowing the values of the powder thermal conductivity, specific heat capacity, and self-heating reaction kinetics could be estimated. Determining the specific heat capacity prior to fitting greatly improved the results. This approach was found to be considerably faster than the existing oven based methods (5 times faster than steady-state approach), determine more parameters, and improve greatly on the error of the estimated parameters (over 20 times more accurate than CPT method). An approach was presented showing how the findings of this research are applied in predicting self-heating in spray dryers for different tower temperatures, heat transfer coefficients, and build-up thicknesses.