The saturation of the fluorescence and its consequences for laser-induced fluorescence thermometry in liquid flows

The temperature dependence of the fluorescence emission of certain organic dyes such as rhodamine B has been widely utilized for measuring the temperature in liquid flows. Measurements are generally based on two assumptions: The fluorescence signal is proportional to the intensity of the laser excitation, and the temperature sensitivity of the dye is not affected by the laser irradiance. In the ratiometric methods, these assumptions allow justifying that the influence of the laser intensity can be totally eliminated in the intensity ratio of two spectral bands of the fluorescence emission and thus that measurements can be taken with no biases under experimental conditions, where the laser propagation is disturbed by the flow. However, when pulsed lasers are used (mainly in planar LIF measurements), the peak irradiance usually compares or exceeds the saturation intensity of the dyes. The present study assesses the consequences of a saturation of the dye emission on temperature measurements. Tests among fluoresceins and rhodamines reveal that the saturation can be accompanied by a significant loss of temperature sensitivity. The dyes, for which this loss of sensitivity is observed, mainly owe their temperature dependence to the fluorescence quantum yield and have a fluorescence signal decreasing with the temperature. The couple fluorescein/sulforhodamine 640 is finally proposed for an implementation of the ratiometric method, since its relatively high temperature dependence (+ 3 %/degrees C) is not altered at high laser irradiances. The possibility of measuring instantaneous temperature fields with this pair of dyes using a single laser shot is finally demonstrated on a turbulent heated jet injected into quiescent water.