Atom-Based Machine Learning Model for Quantitative Property-Structure Relationship of Electronic Properties of Fusenes and Substituted Fusenes

Tuan H Nguyen; Khang M Le; Lam H Nguyen; Thanh N Truong

doi:10.1021/acsomega.3c05212

Atom-Based Machine Learning Model for Quantitative Property-Structure Relationship of Electronic Properties of Fusenes and Substituted Fusenes

ACS Omega. 2023 Oct 2;8(41):38441-38451. doi: 10.1021/acsomega.3c05212. eCollection 2023 Oct 17.

Authors

Tuan H Nguyen¹, Khang M Le², Lam H Nguyen^{2

3}, Thanh N Truong⁴

Affiliations

¹ Faculty of Chemical Engineering, Ho Chi Minh City University of Technology, 268 Ly Thuong Kiet Street, District 10, Ho Chi Minh City 7000000, Vietnam.
² Faculty of Chemistry, VNUHCM-University of Science, 227 Nguyen Van Cu Street, Ho Chi Minh City 700000, Vietnam.
³ Institute for Computational Science and Technology, Ho Chi Minh City 700000, Vietnam.
⁴ Department of Chemistry, University of Utah, Salt Lake City, Utah 84112, United States.

Abstract

This study presents the development of machine-learning-based quantitative structure-property relationship (QSPR) models for predicting electron affinity, ionization potential, and band gap of fusenes from different chemical classes. Three variants of the atom-based Weisfeiler-Lehman (WL) graph kernel method and the machine learning model Gaussian process regressor (GPR) were used. The data pool comprises polycyclic aromatic hydrocarbons (PAHs), thienoacenes, cyano-substituted PAHs, and nitro-substituted PAHs computed with density functional theory (DFT) at the B3LYP-D3/6-31+G(d) level of theory. The results demonstrate that the GPR/WL kernel methods can accurately predict the electronic properties of PAHs and their derivatives with root-mean-square deviations of 0.15 eV. Additionally, we also demonstrate the effectiveness of the active learning protocol for the GPR/WL kernel methods pipeline, particularly for data sets with greater diversity. The interpretation of the model for contributions of individual atoms to the predicted electronic properties provides reasons for the success of our previous degree of π-orbital overlap model.