The effect of cryogenic sample cooling on X-ray absorption spectra

Abstract

It is common for samples to be cooled to near liquid He temperature (4 K) during measurement of their X-ray absorption spectra (XAS). This procedure is believed to improve spectrum quality either by minimizing radiation damage, or by decreasing thermal motions of atoms. The actual benefits realized by cooling are rarely assessed, since that would require duplicate measurements at multiple temperatures, followed by duplicate data analyses. With a cryostat installed, it is difficult to measure room temperature or hotter spectra, which often requires removing the cryostat from the beam path. Here we investigate the effect of cooling and show that it is not globally useful in X-ray absorption spectroscopy. Photolysis does not occur, or its consequences are not controlled by cooling. Secondary photochemical damage is delayed and is remote from the absorption site. Thermal motions do not usually contribute significantly to disorder and consequent damping of EXAFS, either because vibrational amplitudes are small, or because static disorder and structural complexity affect EXAFS in the same way but more profoundly. The low probability of photochemical effects on XAS is in contrast to the situation with crystallography of biological specimens, where photochemical damage anywhere in the sample degrades the data set regardless of proximity to atoms of one specific element. Thermal disorder becomes important in certain types of samples and when the Debye–Waller factor is itself used as a tool. In most cases, it is more efficient to first measure room temperature spectra, and then repeat measurement using a cryostat of only those spectra where the objectives of the analysis justify an attempt to gain more information.