Exploring transition metal complexes with large language models
8 affiliations
- 1Kenneth S. Pitzer Theory Center, University of California Berkeley CA 94720 USA [email protected].
- 2Department of Chemistry, University of California Berkeley CA 94720 USA.
- 3Department of Chemical Engineering, Massachusetts Institute of Technology Cambridge MA 02139 USA.
- 4Chemical Sciences Division, Lawrence Berkeley National Laboratory Berkeley CA 94720 USA.
- 5Department of Materials Science and Engineering, Massachusetts Institute of Technology Cambridge MA 02139 USA.
- 6Department of Chemistry, Massachusetts Institute of Technology Cambridge MA 02139 USA.
- 7Department of Bioengineering, University of California Berkeley CA 94720 USA.
- 8Department of Chemical and Biomolecular Engineering, University of California Berkeley CA 94720 USA.
Abstract
Here we use a large language model for the exploration of chemical space of transition metal complexes (TMCs), fine-tuning the open-source Llama-3.2-1B model with a variation of the SMILES string representation tailored for coordination chemistry. We identify several key molecular properties that are critical to produce valid TMCs which we use for prompt engineering, creating TMC-Llama models that generate unique, novel, and structurally validated TMCs at high success rates, and exceeding the performance of both standard benchmarks as well as state-of-the-art generative methods in the literature. We use the pydentate workflow, Density Functional Theory calculations, and other benchmark metrics to show that the TMC-Llama model generates new TMC chemistries that are physically realistic, synthetically plausible, and geometrically consistent with large datasets of known experimental TMCs. Our study lays the groundwork for future TMC-Llama generative model applications to discover new TMCs for functional materials for catalysis, photochemistry, medicine, and sustainability technologies.
Related
- synthmc: a precedent-based synthesizability score for transition metal complexes
- Building the Data Foundations for the Design and Discovery of Transition Metal Complexes
- Graph Neural Networks for Accelerated Prediction of Excited-State Properties in Transition-Metal Complexes
- PARAMAGNETIC SUPRAMOLECULAR POLYMERS OF METAL-METAL BONDED DIRUTHENIUM COMPLEXES
- Intramolecular protonolysis of platinum(II) methyl complexes: competing mechanisms for metal-ligand proton tautomerism
BibTeX
@article{liu2026exploring,
title = {{Exploring transition metal complexes with large language models}},
author = {Yunsheng Liu and Jacob W Toney and Joseph M Cavanagh and Kunyang Sun and Andrew L Smith and Chung-Yueh Yuan and Roland G St Michel and Paul A Graggs and Dean Toste and Heather J Kulik and Teresa Head-Gordon},
journal = {Chemical science},
year = {2026},
month = sep,
doi = {10.1039/d6sc00436a},
url = {https://europepmc.org/article/PMC/PMC13752763},
note = {CC BY-NC},
}
RIS
TY - JOUR
TI - Exploring transition metal complexes with large language models
AU - Yunsheng Liu
AU - Jacob W Toney
AU - Joseph M Cavanagh
AU - Kunyang Sun
AU - Andrew L Smith
AU - Chung-Yueh Yuan
AU - Roland G St Michel
AU - Paul A Graggs
AU - Dean Toste
AU - Heather J Kulik
AU - Teresa Head-Gordon
PY - 2026
DA - 2026/09/09
JO - Chemical science
DO - 10.1039/d6sc00436a
UR - https://europepmc.org/article/PMC/PMC13752763
AB - Here we use a large language model for the exploration of chemical space of transition metal complexes (TMCs), fine-tuning the open-source Llama-3.2-1B model with a variation of the SMILES string representation tailored for coordination chemistry. We identify several key molecular properties that are critical to produce valid TMCs which we use for prompt engineering, creating TMC-Llama models that generate unique, novel, and structurally validated TMCs at high success rates, and exceeding the performance of both standard benchmarks as well as state-of-the-art generative methods in the literature. We use the pydentate workflow, Density Functional Theory calculations, and other benchmark metrics to show that the TMC-Llama model generates new TMC chemistries that are physically realistic, synthetically plausible, and geometrically consistent with large datasets of known experimental TMCs. Our study lays the groundwork for future TMC-Llama generative model applications to discover new TMCs for functional materials for catalysis, photochemistry, medicine, and sustainability technologies.
SN - 2041-6520
SN - 2041-6539
ER -