A close look at dose: Toward L-edge XAS spectral uniformity, dose quantification and prediction of metal ion photoreduction

This paper represents a call to the broader XAS community to more closely assess dose, as well as the other experimental factors (detection method, temperature, pressure and sample conditions) that affect the rate of X-ray induced chemistry during metal L-edge (as well as other) XAS measurements. While there have been detailed studies on photoreduction within the X-ray community, which have aimed to establish dose-photoreduction relations, it is our view that further standardization of data collection and reporting procedures are required. For molecules as simple as KMnO4 and K3[Fe(CN)6], the literature reported Mn and Fe L-edges, respectively, vary greatly in both overall shapes and spectral distributions. In addition, the reported experimental procedures are often not sufficient to establish the dose-photoreduction relationships. In an effort to more rigorously and quantitatively address this issue, we have measured Mn and Fe L-edge data on KMnO4 and K3[Fe(CN)6] at seven beam lines around the world. We demonstrate that while at a given beam line under identical experimental conditions, there is a dose-photoreduction correlation, this correlation does not hold across different experimental setups or upon variation of the experimental conditions at the same beam line. We introduce a Matlab-script XASskindose to calculate the absorbed dose related to a spectrum. By quantifying the rate of reduction (k), half-dose (D1/2) and the, here introduced, spectroscopically tolerable dose (Dst) values, we find that these values can vary by up to two orders of magnitude for the same compound. Hence, the results presented herein, indicate that a simple dose-photoreduction correlation is insufficient for predicting the rate of X-ray induced changes within a sample. The need to also consider detection method, temperature, pressure and sample conditions are discussed, and correlated to experimental data. A generalized reporting procedure for the broader XAS community is proposed in order to streamline the assessment of reported spectra. In addition, G-values are proposed as a means to predict the rate of metal ion photoreduction prior to an XAS measurement at any beamline by considering the metal ion type, its oxidation state and its stoichiometry in a compound, together with the dose ratio of the compound.