Polymer-Unit Fingerprint (PUFp): An Accessible Expression of Polymer Organic Semiconductors for Machine Learning

Abstract

High-performance organic semiconductors (OSCs) can be designed based on the identification of functional units and their role in the material properties. Herein, we present a polymer-unit fingerprint (PUFp) generation framework, “Python-based polymer-unit-recognition script” (PURS), to identify the subunits “polymer unit” in the polymer and generate polymer-unit fingerprint (PUFp). Using 678 collected OSC data, machine learning (ML) models can be used to determine structure–mobility relationships by using PUFp as a structural input, and the classification accuracy reaches 85.2%. A polymer-unit library consisting of 445 units is constructed, and the key polymer units affecting the mobility of OSCs are identified. By investigating the combinations of polymer units with mobility performance, a scheme for designing OSCs by combining ML approaches and PUFp information is proposed. This scheme not only passively predicts OSC mobility but also actively provides structural guidance for high-mobility OSC material design. The proposed scheme demonstrates the ability to screen materials through pre-evaluation and classification ML steps and is an alternative methodology for applying ML in high-mobility OSC discovery.

Beyond molecular structure: critically assessing machine learning for designing organic photovoltaic materials and devices

“…Recently, the polymer-unit fingerprint (PUFp) has also been introduced as a way to provide a more explainable fingerprint-based representation for conjugated molecules. 44…”

“…Recently, the polymer-unit ngerprint (PUFp) has also been introduced as a way to provide a more explainable ngerprintbased representation for conjugated molecules. 44 Molecular structure can naturally be represented by a graph where atoms are the nodes and bonds are the edges. However, encoding the graph is also non-trivial.…”

Polymer-Unit Graph: Advancing Interpretability in Graph Neural Network Machine Learning for Organic Polymer Semiconductor Materials

“…However, by leveraging the polymer unit concept in machine learning, this limitation can be overcome. In a previous study, we utilized the arrangement of polymer-units as molecular fingerprints to train various classifiers, including random forests, support vector machines, and multilayer perceptron models. The classification accuracy achieved was over 80%.…”

“…For each data collected in the OSC database, the PURS program is used to identify the polymer-units contained within it and the connection relationship between polymer-units . The methodology of polymer-unit recognition, as outlined in “Polymer-Unit Fingerprint ( PUFp ): An Accessible Expression of Polymer OSCs for Machine Learning”, is comprehensively elucidated within this document.…”

“…The role of the polymer unit in the study of polymer is crucial, making it highly significant to incorporate the concept of “polymer-unit” into GNN research within the polymer field. Recently, we proposed a Python-based script named polymer-unit-recognition script ( PURS ), which is a program for recognizing the polymer-unit based on SMILES . Its purpose is to identify the polymer-unit composition of polymers and macromolecules from SMILES.…”