Lowering the Exponential Wall: Accelerating High-Entropy Alloy Catalysts Screening using Local Surface Energy Descriptors from Neural Network Potentials

• URL: http://arxiv.org/abs/2404.08413v1
• Date: Fri, 12 Apr 2024 11:54:06 GMT
• Title: Lowering the Exponential Wall: Accelerating High-Entropy Alloy Catalysts Screening using Local Surface Energy Descriptors from Neural Network Potentials
• Authors: Tomoya Shiota, Kenji Ishihara, Wataru Mizukami,
• Abstract summary: We propose a method to rapidly construct models that predict the properties of HEAs from data on monometallic systems. We make high-precision model development by employing both classical machine learning and quantum machine learning. Our approach allows accelerated exploration of the vast chemical space of HEAs facilitating the design of novel catalysts.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Computational screening is indispensable for the efficient design of high-entropy alloys (HEAs), which hold great potential for catalytic applications. However, the chemical space of HEAs is exponentially vast with respect to the number of constituent elements, and even screening calculations using machine learning potentials can be enormously time-consuming. To address this challenge, we propose a method to rapidly construct models that predict the properties of HEAs from data on monometallic systems (or few-component alloys). The core of our approach is a newly-introduced descriptor called local surface energy ($LSE$), which reflects the local reactivity of solid surfaces at atomic resolution. We successfully created a model using linear regression to screen the adsorption energies of molecules on HEAs based on LSEs from monometallic systems. Furthermore, we made high-precision model development by employing both classical machine learning and quantum machine learning. Using our method, we were able to complete the adsorption energy calculations of CO molecules on 1000 patterns of quinary nanoparticles consisting of 201 atoms within a few hours. These calculations would have taken hundreds of years and hundreds of days using density functional theory and a neural network potential, respectively. Our approach allows accelerated exploration of the vast chemical space of HEAs facilitating the design of novel catalysts.

Related papers

• Platinum-based Catalysts for Oxygen Reduction Reaction simulated with a Quantum Computer [2.8866494528688555]
  Hydrogen has emerged as a promising energy source, holding the key to achieve low-carbon and sustainable mobility. Here, we present a pioneering study that combines classical and quantum computational approaches to investigate the ORR on pure platinum and platinum/cobalt surfaces. Results highlight the involvement of strongly correlated species in the cobalt-containing catalyst, suggesting their potential as ideal candidates for showcasing quantum advantage in future applications.
  arXiv  Detail & Related papers  (2023-07-28T21:49:00Z)
• QH9: A Quantum Hamiltonian Prediction Benchmark for QM9 Molecules [69.25826391912368]
  We generate a new Quantum Hamiltonian dataset, named as QH9, to provide precise Hamiltonian matrices for 999 or 2998 molecular dynamics trajectories. We show that current machine learning models have the capacity to predict Hamiltonian matrices for arbitrary molecules.
  arXiv  Detail & Related papers  (2023-06-15T23:39:07Z)
• Molecular Geometry-aware Transformer for accurate 3D Atomic System modeling [51.83761266429285]
  We propose a novel Transformer architecture that takes nodes (atoms) and edges (bonds and nonbonding atom pairs) as inputs and models the interactions among them. Moleformer achieves state-of-the-art on the initial state to relaxed energy prediction of OC20 and is very competitive in QM9 on predicting quantum chemical properties.
  arXiv  Detail & Related papers  (2023-02-02T03:49:57Z)
• AdsorbML: A Leap in Efficiency for Adsorption Energy Calculations using Generalizable Machine Learning Potentials [8.636519538557001]
  We show machine learning potentials can be leveraged to identify low energy adsorbate-surface configurations more accurately and efficiently. Our algorithm provides a spectrum of trade-offs between accuracy and efficiency, with one balanced option finding the lowest energy configuration 87.36% of the time.
  arXiv  Detail & Related papers  (2022-11-29T18:54:55Z)
• PhAST: Physics-Aware, Scalable, and Task-specific GNNs for Accelerated Catalyst Design [102.9593507372373]
  Catalyst materials play a crucial role in the electrochemical reactions involved in industrial processes. Machine learning holds the potential to efficiently model materials properties from large amounts of data. We propose task-specific innovations applicable to most architectures, enhancing both computational efficiency and accuracy.
  arXiv  Detail & Related papers  (2022-11-22T05:24:30Z)
• Multi-Task Mixture Density Graph Neural Networks for Predicting Cu-based Single-Atom Alloy Catalysts for CO2 Reduction Reaction [61.9212585617803]
  Graph neural networks (GNNs) have drawn more and more attention from material scientists. We develop a multi-task (MT) architecture based on DimeNet++ and mixture density networks to improve the performance of such task.
  arXiv  Detail & Related papers  (2022-09-15T13:52:15Z)
• Deep learning techniques for energy clustering in the CMS ECAL [0.0]
  Reconstruction of electrons and photons in CMS depends on topological clustering of the energy deposited by an incident particle. New methods are being investigated that exploit state-of-the-art deep learning architectures like Graph Neural Networks (GNN) and self-attention algorithms.
  arXiv  Detail & Related papers  (2022-04-21T17:23:43Z)
• Nuclear two point correlation functions on a quantum-computer [105.89228861548395]
  We use current quantum hardware and error mitigation protocols to calculate response functions for a highly simplified nuclear model. In this work we use current quantum hardware and error mitigation protocols to calculate response functions for a modified Fermi-Hubbard model in two dimensions with three distinguishable nucleons on four lattice sites.
  arXiv  Detail & Related papers  (2021-11-04T16:25:33Z)
• Gaussian Moments as Physically Inspired Molecular Descriptors for Accurate and Scalable Machine Learning Potentials [0.0]
  We propose a machine learning method for constructing high-dimensional potential energy surfaces based on feed-forward neural networks. The accuracy of the developed approach in representing both chemical and configurational spaces is comparable to the one of several established machine learning models.
  arXiv  Detail & Related papers  (2021-09-15T16:46:46Z)
• BIGDML: Towards Exact Machine Learning Force Fields for Materials [55.944221055171276]
  Machine-learning force fields (MLFF) should be accurate, computationally and data efficient, and applicable to molecules, materials, and interfaces thereof. Here, we introduce the Bravais-Inspired Gradient-Domain Machine Learning approach and demonstrate its ability to construct reliable force fields using a training set with just 10-200 atoms.
  arXiv  Detail & Related papers  (2021-06-08T10:14:57Z)
• A Universal Framework for Featurization of Atomistic Systems [0.0]
  Reactive force fields based on physics or machine learning can be used to bridge the gap in time and length scales. We introduce the Gaussian multi-pole (GMP) featurization scheme that utilizes physically-relevant multi-pole expansions of the electron density around atoms. We demonstrate that GMP-based models can achieve chemical accuracy for the QM9 dataset, and their accuracy remains reasonable even when extrapolating to new elements.
  arXiv  Detail & Related papers  (2021-02-04T03:11:00Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.