skip to main content

Title: Physics guided deep learning for generative design of crystal materials with symmetry constraints

Discovering new materials is a challenging task in materials science crucial to the progress of human society. Conventional approaches based on experiments and simulations are labor-intensive or costly with success heavily depending on experts’ heuristic knowledge. Here, we propose a deep learning based Physics Guided Crystal Generative Model (PGCGM) for efficient crystal material design with high structural diversity and symmetry. Our model increases the generation validity by more than 700% compared to FTCP, one of the latest structure generators and by more than 45% compared to our previous CubicGAN model. Density Functional Theory (DFT) calculations are used to validate the generated structures with 1869 materials out of 2000 are successfully optimized and deposited into the Carolina Materials, of which 39.6% have negative formation energy and 5.3% have energy-above-hull less than 0.25 eV/atom, indicating their thermodynamic stability and potential synthesizability.

more » « less
Award ID(s):
1940099 2110033 1905775
Author(s) / Creator(s):
; ; ; ; ; ;
Publisher / Repository:
Nature Publishing Group
Date Published:
Journal Name:
npj Computational Materials
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Abstract

    High‐throughput screening has become one of the major strategies for the discovery of novel functional materials. However, its effectiveness is severely limited by the lack of sufficient and diverse materials in current materials repositories such as the open quantum materials database (OQMD). Recent progress in deep learning have enabled generative strategies that learn implicit chemical rules for creating hypothetical materials with new compositions and structures. However, current materials generative models have difficulty in generating structurally diverse, chemically valid, and stable materials. Here we propose CubicGAN, a generative adversarial network (GAN) based deep neural network model for large scale generative design of novel cubic materials. When trained on 375 749 ternary materials from the OQMD database, the authors show that the model is able to not only rediscover most of the currently known cubic materials but also generate hypothetical materials of new structure prototypes. A total of 506 such materials have been verified by phonon dispersion calculation. Considering the importance of cubic materials in wide applications such as solar panels, the GAN model provides a promising approach to significantly expand existing materials repositories, enabling the discovery of new functional materials via screening. The new crystal structures discovered are freely accessible

    more » « less
  2. Abstract

    Elevated seismic noise for moderate‐size earthquakes recorded at teleseismic distances has limited our ability to see their complexity. We develop a machine‐learning‐based algorithm to separate noise and earthquake signals that overlap in frequency. The multi‐task encoder‐decoder model is built around a kernel pre‐trained on local (e.g., short distances) earthquake data (Yin et al., 2022, and is modified by continued learning with high‐quality teleseismic data. We denoise teleseismic P waves of deep Mw5.0+ earthquakes and use the clean P waves to estimate source characteristics with reduced uncertainties of these understudied earthquakes. We find a scaling of moment and duration to beM0 ≃ τ4, and a resulting strong scaling of stress drop and radiated energy with magnitude ( and ). The median radiation efficiency is 5%, a low value compared to crustal earthquakes. Overall, we show that deep earthquakes have weak rupture directivity and few subevents, suggesting a simple model of a circular crack with radial rupture propagation is appropriate. When accounting for their respective scaling with earthquake size, we find no systematic depth variations of duration, stress drop, or radiated energy within the 100–700 km depth range. Our study supports the findings of Poli and Prieto (2016, with a doubled amount of earthquakes investigated and with earthquakes of lower magnitudes.

    more » « less
  3. Abstract

    A new class of core–shell adsorbents has been created by electrospun metal–organic framework (MOF) particles embedded in polymer nanofibers, which have provided many unique properties compared to the existing MOF coating technologies. For the first time, we demonstrate the improved adsorption selectivity of CO2over N2using electrospun polymer/ZIF‐8 adsorbents in experiments. Furthermore, an analytical model based on the assumption that the diffusivity in core is 10 times higher than that in shell is developed to describe the theory of improved selectivity for core–shell adsorbents that is validated against a more accurate finite element model developed in COMSOL. Our model shows three regimes including exclusive shell uptake, linear core uptake, and asymptotic core uptake. These regimes are related to material properties and uptake times, which could be used as design criteria to balance core stability, maximum selectivity, and maximum uptake. An advanced HAADF STEM tomography (MovieS1) shows that the shell thickness in the case of polymer/ZIF‐8 is on the order of 10 nm, allowing the regime of maximum selectivity to be realized. Kinetically limited adsorption tests at 45°C demonstrate that these composite fibers can perform in a regime of selectivity and uptake for the separation of CO2and N2that is unobtainable by either the MOF or fiber independently, showing a great potential for postcombustion CO2capture.

    more » « less
  4. Abstract Background

    No versatile web app exists that allows epidemiologists and managers around the world to comprehensively analyze the impacts of COVID-19 mitigation. The app presented here fills this gap.


    Our web app uses a model that explicitly identifies susceptible, contact, latent, asymptomatic, symptomatic and recovered classes of individuals, and a parallel set of response classes, subject to lower pathogen-contact rates. The user inputs a CSV file of incidence and, if of interest, mortality rate data. A default set of parameters is available that can be overwritten through input or online entry, and a user-selected subset of these can be fitted to the model using maximum-likelihood estimation (MLE). Model fitting and forecasting intervals are specifiable and changes to parameters allow counterfactual and forecasting scenarios. Confidence or credible intervals can be generated using stochastic simulations, based on MLE values, or on an inputted CSV file containing Markov chain Monte Carlo (MCMC) estimates of one or more parameters.


    We illustrate the use of our web app in extracting social distancing, social relaxation, surveillance or virulence switching functions (i.e., time varying drivers) from the incidence and mortality rates of COVID-19 epidemics in Israel, South Africa, and England. The Israeli outbreak exhibits four distinct phases: initial outbreak, social distancing, social relaxation, and a second wave mitigation phase. An MCMC projection of this latter phase suggests the Israeli epidemic will continue to produce into late November an average of around 1500 new case per day, unless the population practices social-relaxation measures at least 5-fold below the level in August, which itself is 4-fold below the level at the start of July. Our analysis of the relatively late South African outbreak that became the world’s fifth largest COVID-19 epidemic in July revealed that the decline through late July and early August was characterised by a social distancing driver operating at more than twice the per-capita applicable-disease-class (pc-adc) rate of the social relaxation driver. Our analysis of the relatively early English outbreak, identified a more than 2-fold improvement in surveillance over the course of the epidemic. It also identified a pc-adc social distancing rate in early August that, though nearly four times the pc-adc social relaxation rate, appeared to barely contain a second wave that would break out if social distancing was further relaxed.


    Our web app provides policy makers and health officers who have no epidemiological modelling or computer coding expertise with an invaluable tool for assessing the impacts of different outbreak mitigation policies and measures. This includes an ability to generate an epidemic-suppression or curve-flattening index that measures the intensity with which behavioural responses suppress or flatten the epidemic curve in the region under consideration.

    more » « less
  5. Abstract

    We analyze three substorms that occur on (1) 9 March 2008 05:14 UT, (2) 26 February 2008 04:05 UT, and (3) 26 February 2008 04:55 UT. Using ACE solar wind velocity and interplanetary magnetic fieldBzvalues, we calculate the rectified (southwardBz) solar wind voltage propagated to the magnetosphere. The solar wind conditions for the two events were vastly different, 300 kV for 9 March 2008 substorm, compared to 50 kV for 26 February 2008. The voltage is input to a nonlinear physics‐based model of the magnetosphere called WINDMI. The output is the westward auroral electrojet current which is proportional to the auroral electrojet (AL) index from World Data Center for Geomagnetism Kyoto and the SuperMAG auroral electrojet index (SML). Substorm onset times are obtained from the superMAG substorm database, Pu et al. (2010,, Lui (2011, and synchronized to Time History of Events and Macroscale Interactions during Substorms satellite data. The timing of onset, model parameters, and intermediate state space variables are analyzed. The model onsets occurred about 5 to 10 min earlier than the reported onsets. Onsets occurred when the geotail current in the WINDMI model reached a critical threshold of 6.2 MA for the 9 March 2008 event, while, in contrast, a critical threshold of 2.1 MA was obtained for the two 26 February 2008 events. The model estimates 1.99 PJ of total energy transfer during the 9 March 2008 event, with 0.95 PJ deposited in the ionosphere. The smaller events on 26 February 2008 resulted in a total energy transfer of 0.37 PJ according to the model, with 0.095 PJ deposited in the ionosphere.

    more » « less