Extracting Thin Film Structures of Energy Materials Using Transformers

Chen Zhang; Valerie A. Niemann; Peter Benedek; Thomas F. Jaramillo; Mathieu Doucet

doi:10.1021/acsphyschemau.4c00054

Extracting Thin Film Structures of Energy Materials Using Transformers

Chen Zhang, Valerie A. Niemann, Peter Benedek, Thomas F. Jaramillo, Mathieu Doucet

Research output: Contribution to journal › Article › peer-review

Abstract

Neutron-Transformer Reflectometry Advanced Computation Engine (N-TRACE), a neural network model using a transformer architecture, is introduced for neutron reflectometry data analysis. It offers fast, accurate initial parameter estimations and efficient refinements, improving efficiency and precision for real-time data analysis of lithium-mediated nitrogen reduction for electrochemical ammonia synthesis, with relevance to other chemical transformations and batteries. Despite limitations in generalizing across systems, it shows promises for the use of transformers as the basis for models that could accelerate traditional approaches to modeling reflectometry data.

Original language	English
Journal	ACS Physical Chemistry Au
DOIs	https://doi.org/10.1021/acsphyschemau.4c00054
State	Accepted/In press - 2024

Funding

A portion of this research used resources at the Spallation Neutron Source (SNS), a Department of Energy (DOE) Office of Science User Facility operated by Oak Ridge National Laboratory. Neutron reflectometry measurements were carried out on the Liquids Reflectometer at the SNS, which is sponsored by the Scientific User Facilities Division, Office of Basic Energy Sciences, DOE. This research also used birthright cloud resources of the Compute and Data Environment for Science (CADES) at the Oak Ridge National Laboratory, which is supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC05-00OR22725. V.A.N., P.B., and T.F.J. acknowledge funding from the Villum Fonden (V-SUSTAIN grant 9455). V.A.N. was supported under the National Science Foundation Graduate Research Fellowship Program under grant no. DGE-1656518 and the Camille and Henry Dreyfus Foundation. V.A.N., P.B., and T.F.J. were supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, Chemical Sciences, Geosciences, and Biosciences Division, Catalysis Science Program through the SUNCAT Center for Interface Science and Catalysis.

Funders	Funder number
Basic Energy Sciences
Data Environment for Science
Camille and Henry Dreyfus Foundation
CADES
Scientific User Facilities Division
Office of Science
U.S. Department of Energy	DE-AC05-00OR22725
Villum Fonden	9455
National Science Foundation	DGE-1656518

Keywords

energy materials
machine learning
reflectometry
thin films
transformer encoder

Access to Document

10.1021/acsphyschemau.4c00054

Cite this

@article{57535506af5d4638907697dde9c8aa4c,

title = "Extracting Thin Film Structures of Energy Materials Using Transformers",

abstract = "Neutron-Transformer Reflectometry Advanced Computation Engine (N-TRACE), a neural network model using a transformer architecture, is introduced for neutron reflectometry data analysis. It offers fast, accurate initial parameter estimations and efficient refinements, improving efficiency and precision for real-time data analysis of lithium-mediated nitrogen reduction for electrochemical ammonia synthesis, with relevance to other chemical transformations and batteries. Despite limitations in generalizing across systems, it shows promises for the use of transformers as the basis for models that could accelerate traditional approaches to modeling reflectometry data.",

keywords = "energy materials, machine learning, reflectometry, thin films, transformer encoder",

author = "Chen Zhang and Niemann, {Valerie A.} and Peter Benedek and Jaramillo, {Thomas F.} and Mathieu Doucet",

note = "Publisher Copyright: {\textcopyright} 2024 The Authors, Oak Ridge National Laboratory. Published by American Chemical Society.",

year = "2024",

doi = "10.1021/acsphyschemau.4c00054",

language = "English",

journal = "ACS Physical Chemistry Au",

issn = "2694-2445",

publisher = "American Chemical Society",

}

TY - JOUR

T1 - Extracting Thin Film Structures of Energy Materials Using Transformers

AU - Zhang, Chen

AU - Niemann, Valerie A.

AU - Benedek, Peter

AU - Jaramillo, Thomas F.

AU - Doucet, Mathieu

PY - 2024

Y1 - 2024

N2 - Neutron-Transformer Reflectometry Advanced Computation Engine (N-TRACE), a neural network model using a transformer architecture, is introduced for neutron reflectometry data analysis. It offers fast, accurate initial parameter estimations and efficient refinements, improving efficiency and precision for real-time data analysis of lithium-mediated nitrogen reduction for electrochemical ammonia synthesis, with relevance to other chemical transformations and batteries. Despite limitations in generalizing across systems, it shows promises for the use of transformers as the basis for models that could accelerate traditional approaches to modeling reflectometry data.

AB - Neutron-Transformer Reflectometry Advanced Computation Engine (N-TRACE), a neural network model using a transformer architecture, is introduced for neutron reflectometry data analysis. It offers fast, accurate initial parameter estimations and efficient refinements, improving efficiency and precision for real-time data analysis of lithium-mediated nitrogen reduction for electrochemical ammonia synthesis, with relevance to other chemical transformations and batteries. Despite limitations in generalizing across systems, it shows promises for the use of transformers as the basis for models that could accelerate traditional approaches to modeling reflectometry data.

KW - energy materials

KW - machine learning

KW - reflectometry

KW - thin films

KW - transformer encoder

UR - http://www.scopus.com/inward/record.url?scp=85209213388&partnerID=8YFLogxK

U2 - 10.1021/acsphyschemau.4c00054

DO - 10.1021/acsphyschemau.4c00054

M3 - Article

AN - SCOPUS:85209213388

SN - 2694-2445

JO - ACS Physical Chemistry Au

JF - ACS Physical Chemistry Au

ER -

Extracting Thin Film Structures of Energy Materials Using Transformers

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this