Fig. 2. Schematic of one-dimensional photoacoustic signal generation showing the temperature changes that occur in the sample and adjacent gas.

The photoacoustic signal is generated by thermal expansion of the gas caused by heat associated with the sum of all the contributions. Contributions come from each of the sample layers in which energy of the infrared beam is absorbed and which is close enough to the surface so that the thermal-wave amplitude has not decayed to a vanishing contribution after crossing the sample-gas interface.
Both the infrared and thermal-wave decay coefficients and , respectively, play a key role in the photoacoustic signal generation. The term in the temperature oscillation expression leads to the linear photoacooustic signal dependence on infrared absorption when <<. In this situation a layer of sample extending a distance beneath the surface contributes 63% of the signal with the other 37% coming from deeper layers of the sample. The thermal-wave decay length L is referred to as the sampling depth of a FTIR-PAS measurement.
The possibility of varying the sample depth of FTIR-PAS measurements via the mirror velocity is apparent when L is written as where the substitution has been made. note also that the sampling depth increases with decreasing wavenember by more than a factor of three across the spectrum of a sample going from 4000 cm^-1 to 400 cm^-1. A constant sampling depth, , versus wavenumber is obtained with FTIR systems operating in the step-scan mode that allows a constant modulation frequency to be selected and detected independent