Improved multivariate analysis for fast and selective monitoring of structural dynamics by in situ X-ray powder diffraction12. Issue 4 (6th November 2017)

• Record Type: Journal Article
• Authors: Guccione, Pietro; Palin, Luca; Milanesio, Marco; Belviso, Benny Danilo; Caliandro, Rocco
• Is Part Of: Physical chemistry chemical physics. Volume 20:Issue 4(2017)
• Page Start: 2175
• Page End: 2187
• Publication Date: 2017-11-06
• Subjects: Chemistry, Physical and theoretical -- Periodicals; 541.3
• DOI: 10.1039/c7cp06326a
• Languages: English
• ISSNs: 1463-9076

Abstract

Abstract : New algorithms for peak alignment of diffraction profiles and advanced multivariate analysis allow improving structural characterization by in situ experiments. Abstract : The development of two solid-state reactions, Xe absorption into MFI and molecular complex formation, where samples are affected by changes of crystal lattice due to temperature or pressure variation was structurally monitored through in situ or in operando X-ray powder diffraction experiments. Consequent variations of the peak positions prevent collective analysis of measured patterns, aiming at investigating structural changes occurring within the crystal cell. Moreover, an intrinsic and variable error in peak position is unavoidable when using the Bragg–Brentano geometry and, in some cases (sticky, bulky, aggregate samples) the sample mounting can increase the error within a dataset. Here we present a general multivariate analysis method to process in a fast and automatic way in situ XRPD data collected on charge transfer complexes and porous materials, with the capacity of disentangling peak shifts from intensity and shape variations in diffraction signals, thus allowing an efficient separation of the contribution of crystal lattice changes from structural changes. The peak shift correction allowed an improved PCA analysis that turned out to be more sensible than the traditional single pattern Rietveld analysis. The developed algorithms allowed, with respect to the traditional approach, the location of two new Xe positions into MFI with a better interpretation of the experimental data, while a much faster and more efficient recovery of the reaction coordinate was achieved in the molecular complex formation reaction.