利用机器学习和量子启发的相似性分析筛选高熵合金电催化剂

IF 17.3 1区 材料科学 Q1 MATERIALS SCIENCE, MULTIDISCIPLINARY Matter Pub Date : 2024-11-06 DOI:10.1016/j.matt.2024.10.001
Yuxin Chang , Ian Benlolo , Yang Bai , Christoff Reimer , Daojin Zhou , Hengrui Zhang , Hidetoshi Matsumura , Hitarth Choubisa , Xiao-Yan Li , Wei Chen , Pengfei Ou , Isaac Tamblyn , Edward H. Sargent
{"title":"利用机器学习和量子启发的相似性分析筛选高熵合金电催化剂","authors":"Yuxin Chang ,&nbsp;Ian Benlolo ,&nbsp;Yang Bai ,&nbsp;Christoff Reimer ,&nbsp;Daojin Zhou ,&nbsp;Hengrui Zhang ,&nbsp;Hidetoshi Matsumura ,&nbsp;Hitarth Choubisa ,&nbsp;Xiao-Yan Li ,&nbsp;Wei Chen ,&nbsp;Pengfei Ou ,&nbsp;Isaac Tamblyn ,&nbsp;Edward H. Sargent","doi":"10.1016/j.matt.2024.10.001","DOIUrl":null,"url":null,"abstract":"<div><div>The discovery of new electrocatalysts can be aided by density functional theory (DFT) computation of overpotentials based on the energies of chemical intermediates on prospective adsorption sites. We hypothesize that when training a machine learning model on DFT data, one could improve accuracy by introducing a quantitative measure of similarity among adsorption sites. When we augment a graph neural network-based machine learning workflow using similarity as an input feature, we find that the required training dataset size is decreased from 1,600 to 800, leading to a 2× acceleration: the number of DFT calculations required to train to a given level of accuracy is cut in half. This approach identifies Fe<sub>0.125</sub>Co<sub>0.125</sub>Ni<sub>0.229</sub>Ir<sub>0.229</sub>Ru<sub>0.292</sub> as a promising oxygen reduction reaction catalyst with an overpotential of 0.24 V, outperforming a Pt/C benchmark. We examine, by studying experimentally four additional HEAs, the predictive power of the computational approach.</div></div>","PeriodicalId":388,"journal":{"name":"Matter","volume":"7 11","pages":"Pages 4099-4113"},"PeriodicalIF":17.3000,"publicationDate":"2024-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"High-entropy alloy electrocatalysts screened using machine learning informed by quantum-inspired similarity analysis\",\"authors\":\"Yuxin Chang ,&nbsp;Ian Benlolo ,&nbsp;Yang Bai ,&nbsp;Christoff Reimer ,&nbsp;Daojin Zhou ,&nbsp;Hengrui Zhang ,&nbsp;Hidetoshi Matsumura ,&nbsp;Hitarth Choubisa ,&nbsp;Xiao-Yan Li ,&nbsp;Wei Chen ,&nbsp;Pengfei Ou ,&nbsp;Isaac Tamblyn ,&nbsp;Edward H. Sargent\",\"doi\":\"10.1016/j.matt.2024.10.001\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>The discovery of new electrocatalysts can be aided by density functional theory (DFT) computation of overpotentials based on the energies of chemical intermediates on prospective adsorption sites. We hypothesize that when training a machine learning model on DFT data, one could improve accuracy by introducing a quantitative measure of similarity among adsorption sites. When we augment a graph neural network-based machine learning workflow using similarity as an input feature, we find that the required training dataset size is decreased from 1,600 to 800, leading to a 2× acceleration: the number of DFT calculations required to train to a given level of accuracy is cut in half. This approach identifies Fe<sub>0.125</sub>Co<sub>0.125</sub>Ni<sub>0.229</sub>Ir<sub>0.229</sub>Ru<sub>0.292</sub> as a promising oxygen reduction reaction catalyst with an overpotential of 0.24 V, outperforming a Pt/C benchmark. We examine, by studying experimentally four additional HEAs, the predictive power of the computational approach.</div></div>\",\"PeriodicalId\":388,\"journal\":{\"name\":\"Matter\",\"volume\":\"7 11\",\"pages\":\"Pages 4099-4113\"},\"PeriodicalIF\":17.3000,\"publicationDate\":\"2024-11-06\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Matter\",\"FirstCategoryId\":\"88\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S2590238524005289\",\"RegionNum\":1,\"RegionCategory\":\"材料科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"MATERIALS SCIENCE, MULTIDISCIPLINARY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Matter","FirstCategoryId":"88","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2590238524005289","RegionNum":1,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"MATERIALS SCIENCE, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 0

摘要

密度泛函理论(DFT)根据化学中间体在预期吸附位点上的能量计算过电势,有助于发现新的电催化剂。我们假设,在根据 DFT 数据训练机器学习模型时,可以通过引入吸附位点之间相似性的定量指标来提高准确性。当我们使用相似性作为输入特征来增强基于图神经网络的机器学习工作流程时,我们发现所需的训练数据集从 1600 个减少到 800 个,从而提高了 2 倍的速度:训练到一定准确度所需的 DFT 计算量减少了一半。这种方法确定了 Fe0.125Co0.125Ni0.229Ir0.229Ru0.292 是一种很有前途的氧还原反应催化剂,其过电势为 0.24 V,优于 Pt/C 基准。我们通过实验研究了另外四种 HEA,检验了计算方法的预测能力。
本文章由计算机程序翻译,如有差异,请以英文原文为准。

摘要图片

摘要图片

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
High-entropy alloy electrocatalysts screened using machine learning informed by quantum-inspired similarity analysis
The discovery of new electrocatalysts can be aided by density functional theory (DFT) computation of overpotentials based on the energies of chemical intermediates on prospective adsorption sites. We hypothesize that when training a machine learning model on DFT data, one could improve accuracy by introducing a quantitative measure of similarity among adsorption sites. When we augment a graph neural network-based machine learning workflow using similarity as an input feature, we find that the required training dataset size is decreased from 1,600 to 800, leading to a 2× acceleration: the number of DFT calculations required to train to a given level of accuracy is cut in half. This approach identifies Fe0.125Co0.125Ni0.229Ir0.229Ru0.292 as a promising oxygen reduction reaction catalyst with an overpotential of 0.24 V, outperforming a Pt/C benchmark. We examine, by studying experimentally four additional HEAs, the predictive power of the computational approach.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
Matter
Matter MATERIALS SCIENCE, MULTIDISCIPLINARY-
CiteScore
26.30
自引率
2.60%
发文量
367
期刊介绍: Matter, a monthly journal affiliated with Cell, spans the broad field of materials science from nano to macro levels,covering fundamentals to applications. Embracing groundbreaking technologies,it includes full-length research articles,reviews, perspectives,previews, opinions, personnel stories, and general editorial content. Matter aims to be the primary resource for researchers in academia and industry, inspiring the next generation of materials scientists.
期刊最新文献
Biomimetic artificial neuromuscular fiber bundles with built-in adaptive feedback Massively multiplexed optical recording with polychromatic DNA frameworks Polyfunctional eutectogels with multiple hydrogen-bond-shielded amorphous networks for soft ionotronics Brilliant colorful daytime radiative cooling coating mimicking scarab beetle Soft bio-metamaterials with high acoustic transparency and gradient refractive index for tunable acoustic beamformer
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1