Design Principles and Top Non-Fullerene Acceptor Candidates for Organic Photovoltaics

By Steven A. Lopez, Benjamín Sánchez-Lengeling, J. Soares, Alán Aspuru-Guzik. Joule, 2017

193 citations recorded by the source corpus.

Summary Non-fullerene acceptors in organic photovoltaics (OPVs) continue to improve upon the shortcomings of many fullerene-based solar cells. We explored the chemical space of over 51,000 non-fullerene acceptors featuring 106 common moieties from the organic electronics literature, including naphthalene diimides, benzathiadiazoles, and fused fluoroanthenediimides. We identify top candidates featuring optimal energy level offsets, based on a well-studied electron-donor: poly[N-9′-heptadecanyl-2,7-carbazole-alt-5,5-(4′,7′-di-2-thienyl-2′,1′,3′-benzothiadiazole)]. The Harvard Clean Energy Project infrastructure was employed through the IBM World Community Grid to carry out quantum mechanical calculations using density functional theory. Gaussian processes regression was utilized to correct the computed frontier molecular orbital energies. Additional time-dependent density functional theory calculations on a subset of the top candidates refined the narrow band-gap chromophores for OPVs. These results on electron-acceptor materials complement the electronic structure calculations of over one million electron-donor materials publicly available through our website.

Subfields: Electrical and Electronic Engineering, Organic Chemistry, Polymers and Plastics

View Paper: https://www.semanticscholar.org/paper/cb3934a9f03672f4f479bad6814d96d1b2bd0992

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Electrical and Electronic Engineering

In Organic Chemistry

Alán Aspuru-Guzik

In Polymers and Plastics

Citations in the corpus

Self-referencing embedded strings (SELFIES): A 100% robust molecular string representation (2019, methodology): “ molecules. One of the datasets, denoted as QM9, consists of 134.000 molecules with up to 9 heavy atoms (C, O, N and F) [36, 37], while the other contains 50.000 small molecule organic semiconductors [38]. We encode them in the four different representations – the sizes of their alphabets and encodi”
Accelerating Materials Development via Automation, Machine Learning, and High-Performance Computing (2018, background): “24 Such an approach has been realized to help discover novel active layers in organic solar cells 11 and light-emitting diodes 12 , and metal alloys 13,27 , among many others.”