Simulated emission spectra of samples at non-uniform temperature based on numerical solutions of the general radiative transfer equation

Infrared spectroscopic methods are increasingly being used for analysis in the food industry. A potentially attractive approach is that of infrared emission spectroscopy: in many food processes, the raw materials are subjected to heating, and once hot, act as sources of infrared radiation. If the emitted radiation is steered directly into a spectrometer with the conventional source removed, an emission spectrum may be obtained of the sample without any further light guiding requirement. However, for bulk samples, the spectra obtained are not very useful, due to the mechanism of self-absorption, whereby chemically different samples give rise to the same, featureless spectral profile. Promising strategies for overcoming this problem are Transient Infrared Emission and Transient Infrared Transmission Spectroscopies (TIRES and TIRTS), in which thermal gradients are induced at the sample surface, enabling composition-specific features to be seen in the emission spectrum. In this paper mathematical models of the effect of thermal gradients on emission spectra are presented, based on numerical solutions of the Radiative Transfer Equation (RTE) for the non-uniform temperature case, using parameters determined experimentally. This provides a method of relating the nature of thermal gradient within a sample to the information content expected in its emission spectrum. Thus, the conditions required to obtain useful emission spectra from bulk samples can be determined, and the potential of emission spectroscopy as an analysis tool in food production processes assessed.