SizeResGraph-Guided GFN2-xTB Screening of Low-Gap Thiophene-Containing Polymer Repeat Units
- 1Guangdong Pharmaceutical University , , 280 Waihuan East Road, Higher Education Mega Center , ,
Abstract
Low-gap conjugated polymers are widely studied for optoelectronic applications, but their vast repeat-unit space makes exhaustive quantum-chemical screening impractical. We developed SizeResGraph, a directed-bond graph neural network with a bounded size-dependent readout, to rank thiophene-containing repeat units by their hydrogen-capped GFN2-xTB HOMO–LUMO gaps. The model learned from 9,905 labeled structures curated from 540,650 public records. Across 10 scaffold partitions and five training seeds, SizeResGraph matched Chemprop D-MPNN within a prespecified ±0.010 eV RMSE equivalence margin. An end-to-end run ranked 419,564 prospective structures in 500.9 s using one RTX 4090 and 16 CPU threads. Direct GFN2-xTB calculations confirmed strong enrichment of the low-gap tail, with Low 5% hit fractions of 88.6% for SizeResGraph and 81.8% for equal-budget Chemprop. In a prediction-independent systematic sample, SizeResGraph predictions showed strong rank agreement with B3LYP-D4/def2-SVP gaps (ρ = 0.836). SizeResGraph combined with targeted quantum-chemical verification therefore provides a scalable route from large repeat-unit libraries to tractable sets of low-gap candidates.
Related
- Increasing Adhesion via Strong Dipoles: Oxazolidinone-containing Polyolefins as Nonreactive Tie Layers in Polymer Laminates
- Optimization and Evaluation of Polymer Inclusion Membrane (PIM) Containing 10% Copoly Eugenol Divinyl Benzene as Carrier Compound for Malachite Green Transport
- Carbon nanotube-containing polybutadiene-based electrospun fibres ultraviolet-cross-linked with pentaerythritol triacrylate as interpenetrating polymer networks
- State-Constrained SHAP Distills Polymer Property Models into Transferable Structural Rules for Fixed-Interval Screening
- A Simple Method for Polymer Screening for Selection of Food Packaging Material
BibTeX
@article{shi2026sizeresgraphguided,
title = {{SizeResGraph-Guided GFN2-xTB Screening of Low-Gap Thiophene-Containing Polymer Repeat Units}},
author = {Bowen Shi and Shuyan Zhang and Jiongzhou Wu and Jinghao Fang and Jin Yang},
journal = {ACS Omega},
year = {2026},
month = oct,
doi = {10.1021/acsomega.6c08651},
url = {https://pubs.acs.org/acsodf/article/11/42/65604/5452212/SizeResGraph-Guided-GFN2-xTB-Screening-of-Low-Gap},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - JOUR
TI - SizeResGraph-Guided GFN2-xTB Screening of Low-Gap Thiophene-Containing Polymer Repeat Units
AU - Bowen Shi
AU - Shuyan Zhang
AU - Jiongzhou Wu
AU - Jinghao Fang
AU - Jin Yang
PY - 2026
DA - 2026/10/15
JO - ACS Omega
DO - 10.1021/acsomega.6c08651
UR - https://pubs.acs.org/acsodf/article/11/42/65604/5452212/SizeResGraph-Guided-GFN2-xTB-Screening-of-Low-Gap
AB - Low-gap conjugated polymers are widely studied for optoelectronic applications, but their vast repeat-unit space makes exhaustive quantum-chemical screening impractical. We developed SizeResGraph, a directed-bond graph neural network with a bounded size-dependent readout, to rank thiophene-containing repeat units by their hydrogen-capped GFN2-xTB HOMO–LUMO gaps. The model learned from 9,905 labeled structures curated from 540,650 public records. Across 10 scaffold partitions and five training seeds, SizeResGraph matched Chemprop D-MPNN within a prespecified ±0.010 eV RMSE equivalence margin. An end-to-end run ranked 419,564 prospective structures in 500.9 s using one RTX 4090 and 16 CPU threads. Direct GFN2-xTB calculations confirmed strong enrichment of the low-gap tail, with Low 5% hit fractions of 88.6% for SizeResGraph and 81.8% for equal-budget Chemprop. In a prediction-independent systematic sample, SizeResGraph predictions showed strong rank agreement with B3LYP-D4/def2-SVP gaps (ρ = 0.836). SizeResGraph combined with targeted quantum-chemical verification therefore provides a scalable route from large repeat-unit libraries to tractable sets of low-gap candidates.
SN - 2470-1343
ER -