Recent advances and applications of deep learning methods in materials science

Deep learning(DL)is one of the fastest-growing topics in materials data science,with rapidly emerging applications spanning atomistic,image-based,spectral,and textual data modalities.DL allows analysis of unstructured data and automated identification of features.The recent development of large materials databases has fueled the application of DL methods in atomistic prediction in particular.In contrast,advances in image and spectral data have largely leveraged synthetic data enabled by high-quality forward models as well as by generative unsupervised DL methods.In this article,we present a high-level overview of deep learning methods followed by a detailed discussion of recent developments of deep learning in atomistic simulation,materials imaging,spectral analysis,and natural language processing.For each modality we discuss applications involving both theoretical and experimental data,typical modeling approaches with their strengths and limitations,and relevant publicly available software and datasets.We conclude the review with a discussion of recent cross-cutting work related to uncertainty quantification in this field and a brief perspective on limitations,challenges,and potential growth areas for DL methods in materials science.

Functionalized, carbon nanotube material for the catalytic degradation of organophosphate nerve agents

Recent world events have emphasized the need to develop innovative, functional materials that will safely neutralize chemical warfare （CW） agents in situ to protect military personnel and civilians from dermal exposure. Here, we demonstrate the efficacy of a novel, proof-of-concept design for a Cu-containing catalyst, chemically bonded to a single-wall carbon nanotube （SWCNT） structural support, to effectively degrade an organophosphate simulant. SWCNTs have high tensile strength and are flexible and light-weight, which make them a desirable structural component for unique, fabric-like materials. This study aims to develop a self-decontaminating, carbon nanotube-derived material that can ultimately be incorporated into a wearable fabric or protective material to minimize dermal exposure to organophosphate nerve agents and to prevent accidental exposure during decontamination procedures. Carboxylated SWCNTs were functionalized with a polymer, which contained Cu-chelating bipyridine groups, and their catalytic activity against an organophosphate simulant was measured over time. The catalytically active, functionalized nanomaterial was characterized using X-ray fluorescence and Raman spectroscopy. Assuming zeroth-order reaction kinetics, the hydrolysis rate of the organophosphate simulant, as monitored by UV-vis absorption in the presence of the catalytically active nanomaterial, was 63 times faster than the uncatalyzed hydrolysis rate for a sample containing only carboxylated SWCNTs or a control sample containing no added nanotube materials.

Novel optical properties and induced magnetic moments in Ru-doped hybrid improper ferroelectric Ca3Ti2O7

Hybrid improper ferroelectric Ca3Ti2O7 and Ca3Tii 9RuO.iO7 ceramics were successfully synthesized by conventional solid-state reaction method.Two strongest diffraction peaks located around 2θ=33°shifted towards the lower angle region with Ru substitution,reflecting structure variation.Grain growth and higher oxygen vacancy concentration after doping resulted in a reduction in the coercive field about 20 kV/cm.Optical bandgap estimated by UV-vis diffuse reflectance(DR)spectrum and X-ray photoelectron spectroscopy(XPS)valence band spectra showed a decreasing trend due to the existence of impurity energy level upon Ru doping,which was consistent with the results of first-principles calculations.The origin of the unexpected induced magnetic moments in Ru-dope Ca3Ti2O7 is also discussed.

Predicting strength distributions of MEMS structures using flaw size and spatial density

The populations of flaws in individual layers of microelectromechanical systems(MEMS)structures are determined and verified using a combination of specialized specimen geometry,recent probabilistic analysis,and topographic mapping.Strength distributions of notched and tensile bar specimens are analyzed assuming a single flaw population set by fabrication and common to both specimen geometries.Both the average spatial density of flaws and the flaw size distribution are determined and used to generate quantitative visualizations of specimens.Scanning probe-based topographic measurements are used to verify the flaw spacings determined from strength tests and support the idea that grain boundary grooves on sidewalls control MEMS failure.The findings here suggest that strength controlling features in MEMS devices increase in separation,i.e.,become less spatially dense,and decrease in size,i.e.,become less potent flaws,as processing proceeds up through the layer stack.The method demonstrated for flaw population determination is directly applicable to strength prediction for MEMS reliability and design.