Machine learning and artificial neural network prediction of interfacial thermal resistance between graphene and hexagonal boron nitride

Hong Yang, Zhongtao Zhang, Jingchao Zhang, Xiao C Zeng

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

High-performance thermal interface materials (TIMs) have attracted persistent attention for the design and development of miniaturized nanoelectronic devices; however, a large number of potential new materials exist to form these heterostructures and the explorations of their thermal properties are time consuming and expensive. In this work, we train several supervised machine learning (ML) and artificial neural network (ANN) models to predict the interfacial thermal resistance (R) between graphene and hexagonal boron-nitride (hBN) with only the knowledge of the system temperature, coupling strength between two layers, and in-plane tensile strains. The training data were obtained by high-throughput computations (HTCs) of R using classical molecular dynamics (MD) simulations. Four different ML models, i.e., linear regression, polynomial regression, decision tree and random forest, are explored. A pair of one dense layer ANNs and another pair of two dense layer deep neural networks (DNNs) are also investigated. It is reported that the DNN models provide better R prediction results compared to the ML models. The thermal property predictions using HTC and ML/ANN models are applicable to a wide range of materials and open up new perspectives in the explorations of TIMs.

Original languageEnglish (US)
Pages (from-to)19092-19099
Number of pages8
JournalNanoscale
Volume10
Issue number40
DOIs
StatePublished - Oct 28 2018

Fingerprint

Graphite
Boron nitride
Heat resistance
Graphene
Learning systems
Neural networks
Thermodynamic properties
Throughput
Nanoelectronics
Tensile strain
Decision trees
Linear regression
Heterojunctions
Molecular dynamics
boron nitride
Polynomials
Computer simulation
Temperature

ASJC Scopus subject areas

  • Materials Science(all)

Cite this

Machine learning and artificial neural network prediction of interfacial thermal resistance between graphene and hexagonal boron nitride. / Yang, Hong; Zhang, Zhongtao; Zhang, Jingchao; Zeng, Xiao C.

In: Nanoscale, Vol. 10, No. 40, 28.10.2018, p. 19092-19099.

Research output: Contribution to journalArticle

@article{49e4298a35d347a49371ca4e5c62256d,
title = "Machine learning and artificial neural network prediction of interfacial thermal resistance between graphene and hexagonal boron nitride",
abstract = "High-performance thermal interface materials (TIMs) have attracted persistent attention for the design and development of miniaturized nanoelectronic devices; however, a large number of potential new materials exist to form these heterostructures and the explorations of their thermal properties are time consuming and expensive. In this work, we train several supervised machine learning (ML) and artificial neural network (ANN) models to predict the interfacial thermal resistance (R) between graphene and hexagonal boron-nitride (hBN) with only the knowledge of the system temperature, coupling strength between two layers, and in-plane tensile strains. The training data were obtained by high-throughput computations (HTCs) of R using classical molecular dynamics (MD) simulations. Four different ML models, i.e., linear regression, polynomial regression, decision tree and random forest, are explored. A pair of one dense layer ANNs and another pair of two dense layer deep neural networks (DNNs) are also investigated. It is reported that the DNN models provide better R prediction results compared to the ML models. The thermal property predictions using HTC and ML/ANN models are applicable to a wide range of materials and open up new perspectives in the explorations of TIMs.",
author = "Hong Yang and Zhongtao Zhang and Jingchao Zhang and Zeng, {Xiao C}",
year = "2018",
month = "10",
day = "28",
doi = "10.1039/c8nr05703f",
language = "English (US)",
volume = "10",
pages = "19092--19099",
journal = "Nanoscale",
issn = "2040-3364",
publisher = "Royal Society of Chemistry",
number = "40",

}

TY - JOUR

T1 - Machine learning and artificial neural network prediction of interfacial thermal resistance between graphene and hexagonal boron nitride

AU - Yang, Hong

AU - Zhang, Zhongtao

AU - Zhang, Jingchao

AU - Zeng, Xiao C

PY - 2018/10/28

Y1 - 2018/10/28

N2 - High-performance thermal interface materials (TIMs) have attracted persistent attention for the design and development of miniaturized nanoelectronic devices; however, a large number of potential new materials exist to form these heterostructures and the explorations of their thermal properties are time consuming and expensive. In this work, we train several supervised machine learning (ML) and artificial neural network (ANN) models to predict the interfacial thermal resistance (R) between graphene and hexagonal boron-nitride (hBN) with only the knowledge of the system temperature, coupling strength between two layers, and in-plane tensile strains. The training data were obtained by high-throughput computations (HTCs) of R using classical molecular dynamics (MD) simulations. Four different ML models, i.e., linear regression, polynomial regression, decision tree and random forest, are explored. A pair of one dense layer ANNs and another pair of two dense layer deep neural networks (DNNs) are also investigated. It is reported that the DNN models provide better R prediction results compared to the ML models. The thermal property predictions using HTC and ML/ANN models are applicable to a wide range of materials and open up new perspectives in the explorations of TIMs.

AB - High-performance thermal interface materials (TIMs) have attracted persistent attention for the design and development of miniaturized nanoelectronic devices; however, a large number of potential new materials exist to form these heterostructures and the explorations of their thermal properties are time consuming and expensive. In this work, we train several supervised machine learning (ML) and artificial neural network (ANN) models to predict the interfacial thermal resistance (R) between graphene and hexagonal boron-nitride (hBN) with only the knowledge of the system temperature, coupling strength between two layers, and in-plane tensile strains. The training data were obtained by high-throughput computations (HTCs) of R using classical molecular dynamics (MD) simulations. Four different ML models, i.e., linear regression, polynomial regression, decision tree and random forest, are explored. A pair of one dense layer ANNs and another pair of two dense layer deep neural networks (DNNs) are also investigated. It is reported that the DNN models provide better R prediction results compared to the ML models. The thermal property predictions using HTC and ML/ANN models are applicable to a wide range of materials and open up new perspectives in the explorations of TIMs.

UR - http://www.scopus.com/inward/record.url?scp=85055077132&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85055077132&partnerID=8YFLogxK

U2 - 10.1039/c8nr05703f

DO - 10.1039/c8nr05703f

M3 - Article

VL - 10

SP - 19092

EP - 19099

JO - Nanoscale

JF - Nanoscale

SN - 2040-3364

IS - 40

ER -