Beyond Scaling Relations for the Description of Catalytic Materials

Mie Andersen; Sergey V. Levchenko; Matthias Scheffler; Karsten Reuter

doi:10.1021/acscatal.8b04478

Beyond Scaling Relations for the Description of Catalytic Materials

Mie Andersen^*, Sergey V. Levchenko, Matthias Scheffler, Karsten Reuter

^*Corresponding author for this work

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaper › Journal article › Research › peer-review

196 Citations (Scopus)

Abstract

Computational screening for new and improved catalyst materials relies on accurate and low-cost predictions of key parameters such as adsorption energies. Here, we use recently developed compressed sensing methods to identify descriptors whose predictive power extends over a wide range of adsorbates, multimetallic transition metal surfaces, and facets. The descriptors are expressed as nonlinear functions of intrinsic properties of the clean catalyst surface, e.g. coordination numbers, d-band moments, and density of states at the Fermi level. From a single density functional theory calculation of these properties, we predict adsorption energies at all potential surface sites, and thereby also the most stable geometry. Compared to previous approaches such as scaling relations, we find our approach to be both more general and more accurate for the prediction of adsorption energies on alloys with mixed-metal surfaces, already when based on training data including only pure metals. This accuracy can be systematically improved by also adding alloy adsorption energies to the training data.

Original language	English
Journal	ACS Catalysis
Volume	9
Issue	4
Pages (from-to)	2752-2759
Number of pages	8
ISSN	2155-5435
DOIs	https://doi.org/10.1021/acscatal.8b04478
Publication status	Published - Apr 2019
Externally published	Yes

Keywords

adsorption energies
catalyst
computational screening
density functional theory
descriptor

Access to Document

10.1021/acscatal.8b04478

Cite this

@article{168c6e69f7e343ae8132b6a51746921a,

title = "Beyond Scaling Relations for the Description of Catalytic Materials",

abstract = "Computational screening for new and improved catalyst materials relies on accurate and low-cost predictions of key parameters such as adsorption energies. Here, we use recently developed compressed sensing methods to identify descriptors whose predictive power extends over a wide range of adsorbates, multimetallic transition metal surfaces, and facets. The descriptors are expressed as nonlinear functions of intrinsic properties of the clean catalyst surface, e.g. coordination numbers, d-band moments, and density of states at the Fermi level. From a single density functional theory calculation of these properties, we predict adsorption energies at all potential surface sites, and thereby also the most stable geometry. Compared to previous approaches such as scaling relations, we find our approach to be both more general and more accurate for the prediction of adsorption energies on alloys with mixed-metal surfaces, already when based on training data including only pure metals. This accuracy can be systematically improved by also adding alloy adsorption energies to the training data.",

keywords = "adsorption energies, catalyst, computational screening, density functional theory, descriptor",

author = "Mie Andersen and Levchenko, {Sergey V.} and Matthias Scheffler and Karsten Reuter",

note = "Funding Information: This project received funding from the European Unions Horizon 2020 research and innovation program under Grant 676580, The NOMAD Laboratory, a European Center of Excellence. The work was also supported by the Bavarian State Ministry of Science, Research, and Arts through the grant {\textquoteleft}Solar Technologies go Hybrid (SolTech){\textquoteright}. The authors gratefully acknowledge the Gauss Centre for Supercomputing e.V. (www.gauss-centre.eu) for funding this project by providing computing time on the GCS Supercomputer SuperMUC at Leibniz Supercomputing Centre (www.lrz.de) as well as computing time granted by the John von Neumann Institute for Computing (NIC) and provided on the supercomputer JURECA at J{\"u}lich Supercomputing Centre (JSC). We also thank Luca Ghiringhelli for helpful discussions and a careful proofreading of the manuscript. Runhai Ouyang wrote the multitask SISSO code18 used in the present work and provided technical assistance. Publisher Copyright: {\textcopyright} 2019 American Chemical Society. Copyright: Copyright 2019 Elsevier B.V., All rights reserved.",

year = "2019",

month = apr,

doi = "10.1021/acscatal.8b04478",

language = "English",

volume = "9",

pages = "2752--2759",

journal = "ACS Catalysis",

issn = "2155-5435",

publisher = "American Chemical Society",

number = "4",

}

TY - JOUR

T1 - Beyond Scaling Relations for the Description of Catalytic Materials

AU - Andersen, Mie

AU - Levchenko, Sergey V.

AU - Scheffler, Matthias

AU - Reuter, Karsten

N1 - Funding Information: This project received funding from the European Unions Horizon 2020 research and innovation program under Grant 676580, The NOMAD Laboratory, a European Center of Excellence. The work was also supported by the Bavarian State Ministry of Science, Research, and Arts through the grant ‘Solar Technologies go Hybrid (SolTech)’. The authors gratefully acknowledge the Gauss Centre for Supercomputing e.V. (www.gauss-centre.eu) for funding this project by providing computing time on the GCS Supercomputer SuperMUC at Leibniz Supercomputing Centre (www.lrz.de) as well as computing time granted by the John von Neumann Institute for Computing (NIC) and provided on the supercomputer JURECA at Jülich Supercomputing Centre (JSC). We also thank Luca Ghiringhelli for helpful discussions and a careful proofreading of the manuscript. Runhai Ouyang wrote the multitask SISSO code18 used in the present work and provided technical assistance. Publisher Copyright: © 2019 American Chemical Society. Copyright: Copyright 2019 Elsevier B.V., All rights reserved.

PY - 2019/4

Y1 - 2019/4

N2 - Computational screening for new and improved catalyst materials relies on accurate and low-cost predictions of key parameters such as adsorption energies. Here, we use recently developed compressed sensing methods to identify descriptors whose predictive power extends over a wide range of adsorbates, multimetallic transition metal surfaces, and facets. The descriptors are expressed as nonlinear functions of intrinsic properties of the clean catalyst surface, e.g. coordination numbers, d-band moments, and density of states at the Fermi level. From a single density functional theory calculation of these properties, we predict adsorption energies at all potential surface sites, and thereby also the most stable geometry. Compared to previous approaches such as scaling relations, we find our approach to be both more general and more accurate for the prediction of adsorption energies on alloys with mixed-metal surfaces, already when based on training data including only pure metals. This accuracy can be systematically improved by also adding alloy adsorption energies to the training data.

AB - Computational screening for new and improved catalyst materials relies on accurate and low-cost predictions of key parameters such as adsorption energies. Here, we use recently developed compressed sensing methods to identify descriptors whose predictive power extends over a wide range of adsorbates, multimetallic transition metal surfaces, and facets. The descriptors are expressed as nonlinear functions of intrinsic properties of the clean catalyst surface, e.g. coordination numbers, d-band moments, and density of states at the Fermi level. From a single density functional theory calculation of these properties, we predict adsorption energies at all potential surface sites, and thereby also the most stable geometry. Compared to previous approaches such as scaling relations, we find our approach to be both more general and more accurate for the prediction of adsorption energies on alloys with mixed-metal surfaces, already when based on training data including only pure metals. This accuracy can be systematically improved by also adding alloy adsorption energies to the training data.

KW - adsorption energies

KW - catalyst

KW - computational screening

KW - density functional theory

KW - descriptor

UR - http://www.scopus.com/inward/record.url?scp=85063987240&partnerID=8YFLogxK

U2 - 10.1021/acscatal.8b04478

DO - 10.1021/acscatal.8b04478

M3 - Journal article

AN - SCOPUS:85063987240

SN - 2155-5435

VL - 9

SP - 2752

EP - 2759

JO - ACS Catalysis

JF - ACS Catalysis

IS - 4

ER -

Beyond Scaling Relations for the Description of Catalytic Materials

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this