Title: Message-passing neural networks for high-throughput polymer screening

Abstract

Machine learning methods have shown promise in predicting molecular properties, and given sufficient training data, machine learning approaches can enable rapid high-throughput virtual screening of large libraries of compounds. Graph-based neural network architectures have emerged in recent years as the most successful approach for predictions based on molecular structure and have consistently achieved the best performance on benchmark quantum chemical datasets. However, these models have typically required optimized 3D structural information for the molecule to achieve the highest accuracy. These 3D geometries are costly to compute for high levels of theory, limiting the applicability and practicality of machine learning methods in high-throughput screening applications. In this study, we present a new database of candidate molecules for organic photovoltaic applications, comprising approximately 91,000 unique chemical structures. Compared to existing datasets, this dataset contains substantially larger molecules (up to 200 atoms) as well as extrapolated properties for long polymer chains. We show that message-passing neural networks trained with and without 3D structural information for these molecules achieve similar accuracy, comparable to state-of-the-art methods on existing benchmark datasets. Furthermore, these results therefore emphasize that for larger molecules with practical applications, near-optimal prediction results can be obtained without using optimized 3D geometry as anmore » input. We further show that learned molecular representations can be leveraged to reduce the training data required to transfer predictions to a new density functional theory functional.« less

Authors:

St. John, Peter C.  ^[1];

• Search DOE PAGES for author "St. John, Peter C."
• Search DOE PAGES for ORCID "0000-0002-7928-3722"

Phillips, Caleb T.  ^[1];

• Search DOE PAGES for author "Phillips, Caleb T."
• Search DOE PAGES for ORCID "0000-0002-3665-4239"

Kemper, Travis W. ^[1]; Wilson, A. Nolan  ^[1];

• Search DOE PAGES for author "Wilson, A. Nolan"
• Search DOE PAGES for ORCID "0000-0002-9002-3585"

Guan, Yanfei ^[2]; Crowley, Michael F.  ^[1];

• Search DOE PAGES for author "Crowley, Michael F."
• Search DOE PAGES for ORCID "0000-0001-5163-9398"

Nimlos, Mark R.  ^[1];

• Search DOE PAGES for author "Nimlos, Mark R."
• Search DOE PAGES for ORCID "0000-0001-7117-775X"

Larsen, Ross E.  ^[1]

• Search DOE PAGES for author "Larsen, Ross E."
• Search DOE PAGES for ORCID "0000-0002-2928-9835"

---

+ Show Author Affiliations

1. National Renewable Energy Lab. (NREL), Golden, CO (United States)
2. Colorado State Univ., Fort Collins, CO (United States)

Publication Date:

Wed Jun 19 00:00:00 EDT 2019

Research Org.:

National Renewable Energy Laboratory (NREL), Golden, CO (United States)

Sponsoring Org.:

USDOE Office of Energy Efficiency and Renewable Energy (EERE), Sustainable Transportation Office. Bioenergy Technologies Office

OSTI Identifier:

1543249

Alternate Identifier(s):

OSTI ID: 1527066

Report Number(s):

NREL/JA-2700-74021
Journal ID: ISSN 0021-9606

Grant/Contract Number:  

AC36-08GO28308

Resource Type:

Accepted Manuscript

Journal Name:

Journal of Chemical Physics

Additional Journal Information:

Journal Volume: 150; Journal Issue: 23; Journal ID: ISSN 0021-9606

Publisher:

American Institute of Physics (AIP)

Country of Publication:

United States

Language:

English

Subject:

14 SOLAR ENERGY; 71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSICS; chemical compounds and components; isomerism; photovoltaics; regression analysis; machine learning; optoelectronic properties; artificial neural networks; molecular properties