The fingerprint method for analysis of thermal desorption spectra

The analysis of thermal desorption spectra (TDS) and the calculation of hydrogen detrapping activation energies rely on Gaussian peak deconvolution and Choo-Lee plot regression since 1982. However, this method imposes important assumptions about the number and shape of the TDS peaks used for fitting. In this study, we propose the <i>fingerprint</i> method, an alternative approach that eliminates these long-standing constraints. By applying the <i>fingerprint</i> analysis to eight TDS spectra from three different Fe-C model alloys, we demonstrate its exceptional sensitivity and ability to resolve activation energy distributions - the material <i>fingerprint</i> - unattainable with traditional methods. We further showcase by manual and automated analysis how the such obtained <i>fingerprints</i> can be used to uniquely distinguish the TDS spectra of each alloy independent of the heating rate. Thus the <i>fingerprint</i> method also increases experimental efficiency by reducing the amount of necessary heating rates for TDS down to one.