CRYSTAL: a multi-agent AI system for automated mapping of materials’ crystal structures

Abstract

Abstract

Random forest machine learning models for interpretable X-ray absorption near-edge structure spectrum-property relationships

“…First, we demonstrate successful machine-learned analysis of XANES data for coordination, in agreement with prior work 46,47 , and to our knowledge, the first models for mean nearest-neighbor distance and Bader charge (the partial charge distribution on atoms, which is correlated with the oxidation state-see "Methods" section and Supplementary Figs. [4][5][6][7][8][9][10][11]. Secondly, we represent XANES spectra using two types of featurization and two choices of normalization, and find that models using post-edge normalized data and multiscale polynomial featurization are readily interpretable and conform with multiple known trends.…”

“…One avenue to accelerate this process involves the use of machine learning (ML) models, which are becoming more reliable with the availability of libraries generated by high-throughput materials experiments and calculations [1][2][3][4][5][6][7][8] . Using these libraries, data-driven techniques are now powerful enough that bulk structure-property relationships can be extracted from experimental X-ray diffraction data using automated agents 9,10 . Data-driven probes of relevant local properties (such as those descriptive of electrochemical behavior 11 ) could further help to accelerate scientific discovery, with the ultimate promise of in operando characterization and automated planning of experiments 2,12,13 .…”

Progress and prospects for accelerating materials science with automated and autonomous workflows

“…composition, structure, processing, morphology) via combinatorial materials science has yielded a wide variety of discoveries and advancements in fundamental knowledge 14,[18][19][20] and has additionally produced experiment databases with unprecedented breadth of materials and measured properties, as exemplied by the recent publication of the High Throughput Experimental Materials database (HTEM) 21 based on photovoltaics materials and the Materials Experiments and Analysis Database (MEAD) 22 based on solar fuels materials. These compilations of raw and analyzed 23 data from individual combinatorial materials science laboratories complement the suite of computational materials databases 60,61 as well as a rapidly growing number of materials data repositories including the Citrination platform, 24 the Materials Data Facility (MDF), 25 and text mining of the literature. 26 For the purposes of the present analysis of automating 12,16,27 materials science workows, these databases serve as successful examples of experiment automation and as resources that can be used to accelerate experiment planning, for example by training machine learning models to identify promising materials.…”

ESAMP: Event-Sourced Architecture for Materials Provenance Management and Application to Accelerated Materials Discovery

“…This has limited application of machine learning for acceleration of experimental materials discovery to specific datasets such as microstructure data, x-ray diffraction spectra, x-ray absorption spectra, or Raman spectra. [8][9][10][11] Recent application of high-throughput experimental techniques have resulted in two large, diverse experimental datasets: a) High Throughput Experimental Materials (HTEM) dataset, which contains synthesis conditions, chemical composition, crystal structure, and optoelectronic property measurements (> 150,000 entries), and b) Materials Experiment and Analysis Database (MEAD) that contains raw data and metadata from millions of materials synthesis and characterization experiments, as well as the corresponding property and performance metrics. 12,13 These datasets contain thousands to millions of data entries for a given type of experimental process, but the experimental conditions or prior processing of the materials leading up to the process of interest can vary substantially.…”