PowdR: An R package for quantitative mineralogy using full pattern summation of X-ray powder diffraction data

Abstract

X-ray powder diffraction (XRPD) is consistently found to be the most accurate analytical technique for qualitative and quantitative mineralogy. For environmental samples such as rocks, sediments and soils that can contain a range of crystalline, disordered and amorphous components, quantitative XRPD methods implemented in Excel-based programmes such as FULLPAT and RockJock have proved particularly appropriate. Such programmes quantify phase concentrations in complex mixtures by treating the diffractogram of a given mixture as the sum of contributions from individual components. Thus, using this full pattern summation (FPS) approach, an observed diffractogram of a sample can be modelled by scaling the contributions of the diffractograms for all pure phases within it and then phase concentrations computed using reference intensity ratios.Here the principles implemented in FULLPAT, RockJock and an in-house application for FPS have together been used to create the powdR package for R. The powdR package advances on previous implementations of FPS in terms of functionality, speed (parallel computation) and adaptability (via use of bespoke R code), and includes a Shiny web application for user-friendly analysis. powdR can be used to quantify samples prepared either with or without an internal standard, and includes the public domain RockJock library of 169 pure diffractograms that cover a wide range of phases that may be encountered in soil and rock samples. Furthermore, powdR contains functionality for automated phase quantification, which can select appropriate phases from a comprehensive reference library (e.g., RockJock). Here the FPS functions implemented in powdR and their use are described in detail.