publications (catalysis/ml)

2024

1. Catalysis/ML

   Pourbaix Machine Learning Framework Identifies Acidic Water Oxidation Catalysts Exhibiting Suppressed Ruthenium Dissolution

   Jehad Abed, Javier Heras-Domingo, Rohan Yuri Sanspeur , and 8 more authors

   Journal of the American Chemical Society, 2024

2023

1. Catalysis/ML

   Identifying limitations in screening high-throughput photocatalytic bimetallic nanoparticles with machine-learned hydrogen adsorptions

   Kirby Broderick, Eric Lopato, Brook Wander , and 3 more authors

   Applied Catalysis B: Environmental, Jan 2023
2. Catalysis/ML

   The Open Catalyst 2022 (OC22) Dataset and Challenges for Oxide Electrocatalysts

   Richard Tran, Janice Lan, Muhammed Shuaibi , and 14 more authors

   ACS Catalysis, Jan 2023
3. Catalysis/ML

   AdsorbML: a leap in efficiency for adsorption energy calculations using generalizable machine learning potentials

   Janice Lan, Aini Palizhati, Muhammed Shuaibi , and 6 more authors

   npj Computational Materials, Jan 2023
4. Catalysis/ML

   WhereWulff: A Semiautonomous Workflow for Systematic Catalyst Surface Reactivity under Reaction Conditions

   Rohan Yuri Sanspeur, Javier Heras-Domingo, John R. Kitchin , and 1 more author

   Journal of Chemical Information and Modeling, Jan 2023

   PMID: 37017312
5. Catalysis/ML

   Applying Large Graph Neural Networks to Predict Transition Metal Complex Energies Using the tmQM_wB97MV Data Set

   Aaron G. Garrison, Javier Heras-Domingo, John R. Kitchin , and 3 more authors

   Journal of Chemical Information and Modeling, Jan 2023

   PMID: 38049389

2022

1. Catalysis/ML

   FINETUNA: Fine-tuning Accelerated Molecular Simulations

   Joseph Musielewicz, Xiaoxiao Wang, Tian Tian , and 1 more author

   Machine Learning: Science and Technology, Sep 2022
2. Catalysis/ML

   Screening of bimetallic electrocatalysts for water purification with machine learning

   Richard Tran, Duo Wang, Ryan Kingsbury , and 4 more authors

   The Journal of Chemical Physics, Sep 2022
3. Catalysis/ML

   The Open Catalyst Challenge 2021: Competition Report

   Abhishek Das, Muhammed Shuaibi, Aini Palizhati , and 22 more authors

   Dec 2022

   PDF
4. Catalysis/ML

   Catlas: an automated framework for catalyst discovery demonstrated for direct syngas conversion

   Brook Wander, Kirby Broderick, and Zachary W. Ulissi

   Catal. Sci. Technol., Jul 2022

   Abs

   Catalyst discovery is paramount to support access to energy and key chemical feedstocks in a post fossil fuel era. Exhaustive computational searches of large material design spaces using ab initio methods like density functional theory (DFT) are infeasible. We seek to explore large design spaces at relatively low computational cost by leveraging large, generalized, graph-based machine learning (ML) models, which are pretrained and therefore require no upfront data collection or training. We present Catlas, a framework that distributes and automates the generation of adsorbate-surface configurations and ML inference of DFT energies to achieve this goal. Catlas is open source, making ML assisted catalyst screenings easy and available to all. To demonstrate its efficacy, we use Catlas to explore catalyst candidates for the direct conversion of syngas to multi-carbon oxygenates. For this case study, we explore 947 stable/metastable binary, transition metal intermetallics as possible catalyst candidates. On this subset of materials, we are able to predict the adsorption energy of key descriptors, *CO and *OH, with good accuracy (0.16, 0.14 eV MAE, respectively). Using the projected selectivity towards C2+ oxygenates from an existing microkinetic model, we identified 144 candidate materials. For 10 promising candidates, DFT calculations reveal a good correlation with our assessment using ML. Among the top elemental combinations were Pt–Ti, Pd–V, Ni–Nb, and Ti–Zn, all of which appear unexplored experimentally.

2021

1. Catalysis/ML

   Open Catalyst 2020 (OC20) Dataset and Community Challenges

   Lowik Chanussot, Abhishek Das, Siddharth Goyal , and 14 more authors

   ACS Catalysis, Apr 2021
2. Catalysis/ML

   Computational catalyst discovery: Active classification through myopic multiscale sampling

   Kevin Tran, Willie Neiswanger, Kirby Broderick , and 3 more authors

   The Journal of Chemical Physics, Apr 2021
3. Catalysis/ML

   Deep reinforcement learning for predicting kinetic pathways to surface reconstruction in a ternary alloy

   Junwoong Yoon, Zhonglin Cao, Rajesh K Raju , and 5 more authors

   Machine Learning: Science and Technology, Apr 2021

2020

1. Catalysis/ML

   Accelerated discovery of CO2 electrocatalysts using active machine learning

   Miao Zhong, Kevin Tran, Yimeng Min , and 19 more authors

   Nature, May 2020

   Abs

   The rapid increase in global energy demand and the need to replace carbon dioxide (CO2)-emitting fossil fuels with renewable sources have driven interest in chemical storage of intermittent solar and wind energy1,2. Particularly attractive is the electrochemical reduction of CO2 to chemical feedstocks, which uses both CO2 and renewable energy3-8. Copper has been the predominant electrocatalyst for this reaction when aiming for more valuable multi-carbon products9-16, and process improvements have been particularly notable when targeting ethylene. However, the energy efficiency and productivity (current density) achieved so far still fall below the values required to produce ethylene at cost-competitive prices. Here we describe Cu-Al electrocatalysts, identified using density functional theory calculations in combination with active machine learning, that efficiently reduce CO2 to ethylene with the highest Faradaic efficiency reported so far. This Faradaic efficiency of over 80 per cent (compared to about 66 per cent for pure Cu) is achieved at a current density of 400 milliamperes per square centimetre (at 1.5 volts versus a reversible hydrogen electrode) and a cathodic-side (half-cell) ethylene power conversion efficiency of 55 ± 2 per cent at 150 milliamperes per square centimetre. We perform computational studies that suggest that the Cu-Al alloys provide multiple sites and surface orientations with near-optimal CO binding for both efficient and selective CO2 reduction17. Furthermore, in situ X-ray absorption measurements reveal that Cu and Al enable a favourable Cu coordination environment that enhances C-C dimerization. These findings illustrate the value of computation and machine learning in guiding the experimental exploration of multi-metallic systems that go beyond the limitations of conventional single-metal electrocatalysts.

2. Catalysis/ML

   Practical Deep-Learning Representation for Fast Heterogeneous Catalyst Screening

   Geun Ho Gu, Juhwan Noh, Sungwon Kim , and 3 more authors

   The Journal of Physical Chemistry Letters, Mar 2020
3. Catalysis/ML

   Enabling robust offline active learning for machine learning potentials using simple physics-based priors

   Muhammed Shuaibi, Saurabh Sivakumar, Rui Qi Chen , and 1 more author

   Machine Learning: Science and Technology, Dec 2020

   Abs

   Machine learning surrogate models for quantum mechanical simulations has enabled the field to efficiently and accurately study material and molecular systems. Developed models typically rely on a substantial amount of data to make reliable predictions of the potential energy landscape or careful active learning and uncertainty estimates. When starting with small datasets, convergence of active learning approaches is a major outstanding challenge which limited most demonstrations to online active learning. In this work we demonstrate a Δ-machine learning approach that enables stable convergence in offline active learning strategies by avoiding unphysical configurations with initial datasets as little as a single data point. We demonstrate our framework’s capabilities on a structural relaxation, transition state calculation, and molecular dynamics simulation, with the number of first principle calculations being cut down anywhere from 70-90%. The approach is incorporated and developed alongside AMP\textittorch, an open-source machine learning potential package, along with interactive Google Colab notebook examples.

4. Catalysis/ML

   Differentiable Optimization for the Prediction of Ground State Structures (DOGSS)

   Junwoong Yoon, and Zachary W Ulissi

   Physical Review Letters, Dec 2020
5. Catalysis/ML

   An Introduction to Electrocatalyst Design using Machine Learning for Renewable Energy Storage

   C Lawrence Zitnick, Lowik Chanussot, Abhishek Das , and 8 more authors

   arXiv preprint arXiv:2010.09435, Dec 2020

2019

1. Catalysis/ML

   Towards Predicting Intermetallics Surface Properties with High-Throughput DFT and Convolutional Neural Networks

   Aini Palizhati, Wen Zhong, Kevin Tran , and 2 more authors

   Journal of Chemical Information and Modeling, Dec 2019

2018

1. Catalysis/ML

   Active learning across intermetallics to guide discovery of electrocatalysts for CO2 reduction and H2 evolution

   Kevin Tran, and Zachary W. Ulissi

   Nature Catalysis, Sep 2018

2017

1. Catalysis/ML

   To address surface reaction network complexity using scaling relations machine learning and DFT calculations

   Zachary W. Ulissi, A. J. Medford, Thomas Bligaard , and 1 more author

   Nature Communications, Mar 2017
2. Catalysis/ML

   Machine-Learning Methods Enable Exhaustive Searches for Active Bimetallic Facets and Reveal Active Site Motifs for CO2 Reduction

   Zachary W. Ulissi, Michael T. Tang, Jianping Xiao , and 9 more authors

   ACS Catalysis, Oct 2017

2016

1. Catalysis/ML

   Automated Discovery and Construction of Surface Phase Diagrams using Machine Learning

   Zachary W Ulissi, Aayush R Singh, Charlie Tsai , and 1 more author

   The Journal of Physical Chemistry Letters, Oct 2016

   Abs

   Surface phase diagrams are necessary for understanding surface chemistry in electrochemical catalysis, where a range of adsorbates and coverages exist at varying applied potentials. These diagrams are typically constructed using intuition, which risks missing complex coverages and configurations at potentials of interest. More accurate cluster expansion methods are often difficult to implement quickly for new surfaces. We adopt a machine learning approach to rectify both issues. Using a Gaussian process regression model, the free energy of all possible adsorbate coverages for surfaces is predicted for a finite number of adsorption sites. Our result demonstrates a rational, simple, and systematic approach for generating accurate free-energy diagrams with reduced computational resources. The Pourbaix diagram for the IrO2(110) surface (with nine coverages from fully hydrogenated to fully oxygenated surfaces) is reconstructed using just 20 electronic structure relaxations, compared to approximately 90 using typical search methods. Similar efficiency is demonstrated for the MoS2 surface.