Estimation of infrared detector bandwidths and effect on the Pulsed Periodic Photothermal Radiometry performance

The Pulsed Periodic Photothermal Radiometry (PPTR) setup in front face configuration (FF-PPTR) allows the investigation of multi-layered samples with layer thicknesses of a few hundred nanometers. The characteristic conduction time in such thin layers may be in the order of magnitude of 10 nanoseconds. In such conditions, the estimation of the thermal properties and/or thermal contact resistances requires a measurement system with a minimum cutoff frequency of about 100 MHz. The bandwidth of the FF-PPTR method depends on several critical components: infrared detector bandwidth, amplifier bandwidth and acquisition system bandwidth. Moreover, the laser pulse is another key element since it must be short enough to ensure that high frequencies actually exist in the thermal excitation. In this work, a spectral analysis of these different elements is carried out. Since photovoltaic infrared (IR) detectors exhibit variable response times depending on their size and technology, a comparison of the response time of several detectors is done. They are measured using a reference sample submitted to laser pulses (Coherent Nd:YAG diode-pumped laser) and by using a new high-speed amplifier (1.4E6 V/A@40 MHz) developed by Alt-RD company. The Kolmar detector KMPV11 with its own integrated amplifier is included in the benchmark. An electrical model of the Alt-RD amplifier is introduced and combined to the heat transfer model of a multilayer sample. An experiment design analysis is then performed to identify the optimum laser pulse width and the detector cut-off frequency that minimize the uncertainty of the estimated thermal diffusivity. Finally, the frequency content of the laser pulse is estimated based on the laser pulse-shape measured by a fast photodiode (THORLABS DET10) with 1 nanosecond rise time. These results are presented and discussed to derive their effect on the performance of the FF-PPTR method.