Monte Carlo bottom-up evaluation of global instrumental quantification uncertainty: Flexible and user-friendly computational tool.

Many instrumental methods of analysis require the daily collection of calibrator signals to calibrate their response. The quality of quantifications based on these calibrations depends on calibrators quality, instrumental signal performance and regression model fitness. Linear Ordinary Least Squares (LOLS), Linear Weighted Least Squares (LWLS) or Linear Bivariate Least Squares (LBLS) regression models can be used to calibrate and evaluate the uncertainty from instrumental quantifications, but require the fulfilment of some assumptions, namely, constant signal variance (LOLS), high calibrators quality (LOLS and LWLS) and linear variation of instrumental signal with calibrator values. The LBLS is flexible regarding calibrator values uncertainty and correlation but requires the determination of calibrator values and signals covariances. This work developed a computational tool for the bottom-up evaluation of global instrumental quantifications uncertainty which simulates calibrator values correlations from entered calibrators preparation procedure and simulates calibrators and samples signals precision from prior precision data, allowing accurate uncertainty evaluation from a few replicate signals of the daily calibration. The used signal precision models were built from previously observed repeatability variation throughout the calibration interval adjusted to daily precision condition from a residual standard deviation adjustment factor. This approach was implemented in a user-friendly MS-Excel file and was successfully applied to the analysis of As, Cd, Ni and Pb in marine sediment extracts by Absorption Spectroscopy. Evaluations were tested by the metrological compatibility of estimated and reference values of control standards for confidence levels of 95% and 99%. The success rates of the compatibility tests were statistically equivalent to the confidence level (p-value>0.01).

A Deep Neural Network for the Rapid Prediction of X-ray Absorption Spectra.

X-ray spectroscopy delivers strong impact across the physical and biological sciences by providing end users with highly detailed information about the electronic and geometric structure of matter. To decode this information in challenging cases, e.g., in operando catalysts, batteries, and temporally evolving systems, advanced theoretical calculations are necessary. The complexity and resource requirements often render these out of reach for end users, and therefore, the data are often not interpreted exhaustively, leaving a wealth of valuable information unexploited. In this paper, we introduce supervised machine learning of X-ray absorption spectra through the development of a deep neural network (DNN) that is able to estimate Fe K-edge X-ray absorption near-edge structure spectra in less than a second with no input beyond geometric information about the local environment of the absorption site. We predict peak positions with sub-eV accuracy and peak intensities with errors over an order of magnitude smaller than the spectral variations that the model is engineered to capture. The performance of the DNN is promising, as illustrated by its application to the structural refinement of tris(bipyridine)iron(II) and nitrosylmyoglobin, but also highlights areas on which future developments should focus.