Weike’s paper on “Deep Neural Networks for Accurate Predictions of Crystal Stability” is now out in Nature Communications. Predicting the stability of crystals is one of the central problems in materials science. Here, we show that deep neural networks, i.e., algorithms that mimic the animal brain, utilizing just the electronegativity and ionic radii as inputs can predict formation energies of crystals with extremely high accuracy. We also demonstrate how these models can be generalized for mixed crystals using a binary encoding scheme, and use it to identify thousands of potentially stable new compositions. We have published a web app (http://crystals.ai) that enables anyone to use these models. News: UCSD News: Scientists use artificial neural networks to predict new stable materials Paper: W. Ye, C. Chen, Z. Wang, I.-H. Chu, S.P. Ong, Deep neural networks for accurate predictions of crystal stability, Nat. Commun. 9 (2018) 3800. doi:10.1038/s41467-018-06322-x.
Xiangguo’s article on “Quantum-accurate spectral neighbor analysis potential models for Ni-Mo binary alloys and fcc metals” has just been published in Physical Review B! In this work, we extend the spectral neighbor analysis potential, or SNAP, approach to fcc Ni-bcc Mo binary alloy systems. These new potentials are a substantial improvement over previous potentials based on the embedded atom method in terms of both energy and property (elastic constants, phonons, surface energies, etc.) predictions. In particular, we show that we can reproduce the Ni-Mo finite temperature phase diagram with high accuracy using the Ni-Mo SNAP model. Such high-accuracy, low-computational-cost SNAP models are an exciting enabler to studies of microstructural properties of alloys. Article: Li, X.-G.; Hu, C.; Chen, C.; Deng, Z.; Luo, J.; Ong, S. P. Quantum-Accurate Spectral Neighbor Analysis Potential Models for Ni-Mo Binary Alloys and Fcc Metals. Phys. Rev. B 2018, 98 (9), 094104, doi:10.1103/PhysRevB.98.094104.
We have recently published two articles in Nature Scientific Data and MRS Bulletin! The first article is a follow-up from our npj Computational Materials article that specifically deals with the scope of data present in the X-ray absorption spectroscopy database (XASDb). The second article is a review on harnessing Materials Project data for machine learning and accelerated discovery. Check out both articles in our publications page!
Our first PhD graduate, Zhenbin Wang, has just been awarded the Chancellor’s Dissertation Medal! The Chancellor’s Dissertation Medal recognizes outstanding Ph.D. dissertations with impact and originality. Zhenbin’s dissertation on the “Design and Optimization of Phosphors for Solid-State Lighting using First-Principles Calculations” has provide immense insights into the structure-composition-property relationships in phosphors (materials which emit light), culminating in the discovery of the novel Sr2LiAlO4 phosphor – the first known crystal in the Sr-Li-Al-O chemistry. Congratulations to Zhenbin on the well-deserved honor!
Shyue Ping recently co-authored a Future Energy article on “The Promise and Challenges of Quantum Computing for Energy Storage” with Alan Ho and Jarrod McClean of Google. This article frames and explores the opportunity of applying quantum computing to energy storage, with a focus on computational materials design of batteries.
Yuh-chieh Lin and Iek-Heng Chu are proud co-authors of a recent article published in Advanced Energy Materials on KVOPO4, a novel, high capacity multi-electron cathode for Na-ion batteries. This highly collaborative work, which is part of the NorthEast Center for Chemical Energy Storage (NECCESS) demonstrates fully activated Na+ intercalation over the V3+/4+/5+ couple in a vanadyl phosphate phase for the first time, with a high practical energy density of over 600 Wh/kg, the highest yet reported for any sodium cathode material. DFT calculations (contribution from MAVRL) shows that KVOPO4 is a 3D ionic conductor with low Na+ migration energy barrier of
Hanmei’s co-author paper with the Chen and Lipomi groups on “Understanding the Electrochemical Properties of Naphthalene Diimide: Implication for Stable and High-Rate Lithium-Ion Battery Electrodes” has just been published in Chemistry of Materials. In this work, we investigate the redox-active organic molecule, 1,4,5,8-naphthalenediimide (NDI), as a low-cost, high-abundance alternative to transition metal-based electrodes for lithium-ion batteries. Hanmei’s contribution is in using the latest SCAN functional combined with the HSE functional to identify the stable Li intercalation sites and compute the voltage profile of NDI, which are in excellent agreement with the experiments from the Chen group.
We are proud to be part of a collaborative publication with the Laboratory of Energy Storage and Conversion on “New Insights into the Interphase between the Na Metal Anode and Sulfide Solid-State Electrolytes: A Joint Experimental and Computational Study” published in ACS Applied Materials & Interfaces. This combined experimental and computational study shows that capacity fade is primarily brought about by the reaction between the Na anode and Na solid electrolytes such as Na3SbS4, Na3PS4, and Cl-doped Na3PS4, and demonstrates techniques that can be used to identify the interfacial products. Read the article here!
Chen’s paper on “Automated generation and ensemble-learned matching of X-ray absorption spectra” has been published in npj Computational Materials. In this work, we developed XASdb, a large database of computed reference X-ray absorption spectra (XAS), and a novel Ensemble-Learned Spectra IdEntification (ELSIE) algorithm for the matching of spectra. XASdb currently hosts more than 800,000 K-edge X-ray absorption near-edge spectra (XANES) for over 40,000 materials from the open-science Materials Project database. We will demonstrate that the ELSIE algorithm, which combines 33 weak “learners” comprising a set of preprocessing steps and a similarity metric, can achieve up to 84.2% accuracy in identifying the correct oxidation state and coordination environment. The XASdb with the ELSIE algorithm has been integrated into a web application in the Materials Project, providing an important new public resource for the analysis of XAS to all materials researchers. Finally, the ELSIE algorithm itself has been made available as part of Veidt, an open source machine learning library for materials science.
Prof Ong has been appointed to the Editorial Board of iScience. iScience is a new interdisciplinary, open-access journal by Cell Press that publishes basic and applied research that advances a specific field across life, physical, and earth sciences. Check out their latest articles at http://www.cell.com/iscience/home.