Machine learning (ML) is becoming an effective tool for studying 2D materials. Taking as input computed or experimental materials data, ML algorithms predict the structural, electronic, mechanical, ...and chemical properties of 2D materials that have yet to be discovered. Such predictions expand investigations on how to synthesize 2D materials and use them in various applications, as well as greatly reduce the time and cost to discover and understand 2D materials. This tutorial review focuses on the understanding, discovery, and synthesis of 2D materials enabled by or benefiting from various ML techniques. We introduce the most recent efforts to adopt ML in various fields of study regarding 2D materials and provide an outlook for future research opportunities. The adoption of ML is anticipated to accelerate and transform the study of 2D materials and their heterostructures.
Layered molybdenum disulfide has demonstrated great promise as a low-cost alternative to platinum-based catalysts for electrochemical hydrogen production from water. Research effort on this material ...has focused mainly on synthesizing highly nanostructured molybdenum disulfide that allows the exposure of a large fraction of active edge sites. Here we report a promising microwave-assisted strategy for the synthesis of narrow molybdenum disulfide nanosheets with edge-terminated structure and a significantly expanded interlayer spacing, which exhibit striking kinetic metrics with onset potential of -103 mV, Tafel slope of 49 mV per decade and exchange current density of 9.62 × 10(-3) mA cm(-2), performing among the best of current molybdenum disulfide catalysts. Besides benefits from the edge-terminated structure, the expanded interlayer distance with modified electronic structure is also responsible for the observed catalytic improvement, which suggests a potential way to design newly advanced molybdenum disulfide catalysts through modulating the interlayer distance.
The pressing need for novel materials that can serve rising demands in solar cell and optoelectronic technologies makes the nexus of halide perovskites, high-throughput computations, and machine ...learning, very promising. Ever increasing amounts of data on the structure, fundamental properties, and device performance of halide perovskites provide opportunities for learning chemical rules and design principles that make these materials attractive, and applying them across wide chemical spaces. In this work, we show that impurity properties of halide perovskites computed using density functional theory (DFT) can be combined with machine learning (ML) to deliver predictive models and quick identification of optoelectronically active impurity atoms. Our computation lead to the largest reported dataset of the formation energies and charge transition levels of Pb-site impurities in methylammonium lead halide (
MAPbX
3
) perovskites. Descriptors are defined to uniquely represent any impurity atom in any
MAPbX
3
compound and mapped to the computed impurity properties using regression techniques such as Gaussian process regression, neural networks, and random forests. We use the best optimized predictive models to make predictions for hundreds of impurities across 9
MAPbX
3
compounds and create lists of dominating impurities, that is, impurities that can shift the equilibrium Fermi level in the perovskite as determined by native point defects. This accelerated screening powered by computations and machine learning can guide the identification of problematic impurities that may cause undesired recombination of charge carriers, as well as impurities that can be deliberately introduced to tune the perovskite conductivity and resulting photovoltaic absorption.
Nasopharyngeal carcinoma (NPC) is an aggressive head and neck cancer characterized by Epstein-Barr virus (EBV) infection and dense lymphocyte infiltration. The scarcity of NPC genomic data hinders ...the understanding of NPC biology, disease progression and rational therapy design. Here we performed whole-exome sequencing (WES) on 111 micro-dissected EBV-positive NPCs, with 15 cases subjected to further whole-genome sequencing (WGS), to determine its mutational landscape. We identified enrichment for genomic aberrations of multiple negative regulators of the NF-κB pathway, including CYLD, TRAF3, NFKBIA and NLRC5, in a total of 41% of cases. Functional analysis confirmed inactivating CYLD mutations as drivers for NPC cell growth. The EBV oncoprotein latent membrane protein 1 (LMP1) functions to constitutively activate NF-κB signalling, and we observed mutual exclusivity among tumours with somatic NF-κB pathway aberrations and LMP1-overexpression, suggesting that NF-κB activation is selected for by both somatic and viral events during NPC pathogenesis.