Explainable Synthesizability Prediction of Inorganic Crystal Polymorphs Using Large Language Models

Kim, Seongmin; Schrier, Joshua; Jung, Yousung

doi:10.1002/ange.202423950

Explainable Synthesizability Prediction of Inorganic Crystal Polymorphs Using Large Language Models

Journal Article · Sat Mar 22 00:00:00 EDT 2025 · Angewandte Chemie

DOI:https://doi.org/10.1002/ange.202423950· OSTI ID:2536774

Kim, Seongmin ^[1]; Schrier, Joshua ^[2]; Jung, Yousung ^[3]

Department of Chemical and Biological Engineering (BK21 four) Seoul National University 1 Gwanak‐ro Gwanak‐gu Seoul 08826 South Korea, Department of Chemical and Biomolecular Engineering Korea Advanced Institute of Science and Technology (KAIST) 291, Daehak‐ro Yuseong‐gu Daejeon 34141 South Korea
Department of Chemistry and Biochemistry Fordham University 441 E. Fordham Road The Bronx New York 10458 USA
Department of Chemical and Biological Engineering (BK21 four) Seoul National University 1 Gwanak‐ro Gwanak‐gu Seoul 08826 South Korea, Institute of Chemical Processes Seoul National University 1 Gwanak‐ro Gwanak‐gu Seoul 08826 South Korea, Institute of Engineering Research Seoul National University 1 Gwanak‐ro Gwanak‐gu Seoul 08826 South Korea

Abstract

We evaluate the ability of machine learning to predict whether a hypothetical crystal structure can be synthesized and explain those predictions to scientists. Fine‐tuned large language models (LLMs) trained on a human‐readable text description of the target crystal structure perform comparably to previous bespoke convolutional graph neural network methods, but better prediction quality can be achieved by training a positive‐unlabeled learning model on a text‐embedding representation of the structure. An LLM‐based workflow can then be used to generate human‐readable explanations for the types of factors governing synthesizability, extract the underlying physical rules, and assess the veracity of those rules. These explanations can guide chemists in modifying or optimizing non‐synthesizable hypothetical structures to make them more feasible for materials design.

View Journal Article

Sponsoring Organization:: USDOE

OSTI ID:: 2536774

Journal Information:: Angewandte Chemie, Journal Name: Angewandte Chemie Journal Issue: 19 Vol. 137; ISSN 0044-8249

Publisher:: Wiley Blackwell (John Wiley & Sons)Copyright Statement

Country of Publication:: Germany

Language:: English

References (63)

Conceptual Inorganic Materials Discovery - A Road Map Jansen, Martin Advanced Materials, Vol. 27, Issue 21 https://doi.org/10.1002/adma.201500143	journal	April 2015
Encoding the atomic structure for machine learning in materials science Li, Shunning; Liu, Yuanji; Chen, Dong WIREs Computational Molecular Science, Vol. 12, Issue 1 https://doi.org/10.1002/wcms.1558	journal	June 2021
Learning from positive and unlabeled data: a survey Bekker, Jessa; Davis, Jesse Machine Learning, Vol. 109, Issue 4 https://doi.org/10.1007/s10994-020-05877-5	journal	April 2020
Large-language models: The game-changers for materials science research Yu, Songlin; Ran, Nian; Liu, Jianjun Artificial Intelligence Chemistry, Vol. 2, Issue 2 https://doi.org/10.1016/j.aichem.2024.100076	journal	December 2024
High-Throughput Screening of Solid-State Li-Ion Conductors Using Lattice-Dynamics Descriptors Muy, Sokseiha; Voss, Johannes; Schlem, Roman iScience, Vol. 16 https://doi.org/10.1016/j.isci.2019.05.036	journal	June 2019
Synthesizability of materials stoichiometry using semi-supervised learning Jang, Jidon; Noh, Juhwan; Zhou, Lan Matter, Vol. 7, Issue 6 https://doi.org/10.1016/j.matt.2024.05.002	journal	June 2024
Explaining nonlinear classification decisions with deep Taylor decomposition Montavon, Grégoire; Lapuschkin, Sebastian; Binder, Alexander Pattern Recognition, Vol. 65 https://doi.org/10.1016/j.patcog.2016.11.008	journal	May 2017
A bagging SVM to learn from positive and unlabeled examples Mordelet, F.; Vert, J. -P. Pattern Recognition Letters, Vol. 37 https://doi.org/10.1016/j.patrec.2013.06.010	journal	February 2014
Artificial Intelligence Driving Materials Discovery? Perspective on the Article: Scaling Deep Learning for Materials Discovery Cheetham, Anthony K.; Seshadri, Ram Chemistry of Materials, Vol. 36, Issue 8 https://doi.org/10.1021/acs.chemmater.4c00643	journal	April 2024
Discovery of Hidden Classes of Layered Electrides by Extensive High-Throughput Material Screening Zhou, Jun; Shen, Lei; Yang, Ming Chemistry of Materials, Vol. 31, Issue 6 https://doi.org/10.1021/acs.chemmater.8b03021	journal	February 2019
Developing Quantitative Structure–Activity Relationship (QSAR) Models for Water Contaminants’ Activities/Properties by Fine-Tuning GPT-3 Models Zhong, Shifa; Guan, Xiaohong Environmental Science & Technology Letters, Vol. 10, Issue 10 https://doi.org/10.1021/acs.estlett.3c00599	journal	September 2023
Comment on “Comparing the Performance of College Chemistry Students with ChatGPT for Calculations Involving Acids and Bases” Schrier, Joshua Journal of Chemical Education, Vol. 101, Issue 5 https://doi.org/10.1021/acs.jchemed.4c00058	journal	April 2024
Label-Free Data Mining of Scientific Literature by Unsupervised Syntactic Distance Analysis Zhang, Baicheng; Xiao, Hengyu; Ye, Guilin The Journal of Physical Chemistry Letters, Vol. 15, Issue 1 https://doi.org/10.1021/acs.jpclett.3c03345	journal	December 2023
Prediction of Synthesis of 2D Metal Carbides and Nitrides (MXenes) and Their Precursors with Positive and Unlabeled Machine Learning Frey, Nathan C.; Wang, Jin; Vega Bellido, Gabriel Iván ACS Nano, Vol. 13, Issue 3 https://doi.org/10.1021/acsnano.8b08014	journal	March 2019
Predicting Synthesizability using Machine Learning on Databases of Existing Inorganic Materials Zhu, Ruiming; Tian, Siyu Isaac Parker; Ren, Zekun ACS Omega, Vol. 8, Issue 9 https://doi.org/10.1021/acsomega.2c04856	journal	February 2023
Evaluation of Tavorite-Structured Cathode Materials for Lithium-Ion Batteries Using High-Throughput Computing Mueller, Tim; Hautier, Geoffroy; Jain, Anubhav Chemistry of Materials, Vol. 23, Issue 17 https://doi.org/10.1021/cm200753g	journal	September 2011
Materials Cartography: Representing and Mining Materials Space Using Structural and Electronic Fingerprints Isayev, Olexandr; Fourches, Denis; Muratov, Eugene N. Chemistry of Materials, Vol. 27, Issue 3 https://doi.org/10.1021/cm503507h	journal	January 2015
Structure-Based Synthesizability Prediction of Crystals Using Partially Supervised Learning Jang, Jidon; Gu, Geun Ho; Noh, Juhwan Journal of the American Chemical Society, Vol. 142, Issue 44 https://doi.org/10.1021/jacs.0c07384	journal	October 2020
Rational Solid-State Synthesis Routes for Inorganic Materials Aykol, Muratahan; Montoya, Joseph H.; Hummelshøj, Jens Journal of the American Chemical Society, Vol. 143, Issue 24 https://doi.org/10.1021/jacs.1c04888	journal	June 2021
In Pursuit of the Exceptional: Research Directions for Machine Learning in Chemical and Materials Science Schrier, Joshua; Norquist, Alexander J.; Buonassisi, Tonio Journal of the American Chemical Society, Vol. 145, Issue 40 https://doi.org/10.1021/jacs.3c04783	journal	September 2023
ChatGPT Chemistry Assistant for Text Mining and the Prediction of MOF Synthesis Zheng, Zhiling; Zhang, Oufan; Borgs, Christian Journal of the American Chemical Society, Vol. 145, Issue 32 https://doi.org/10.1021/jacs.3c05819	journal	August 2023
Large Language Models for Inorganic Synthesis Predictions Kim, Seongmin; Jung, Yousung; Schrier, Joshua Journal of the American Chemical Society, Vol. 146, Issue 29 https://doi.org/10.1021/jacs.4c05840	journal	July 2024
Generative Pretrained Transformer for Heterogeneous Catalysts Mok, Dong Hyeon; Back, Seoin Journal of the American Chemical Society, Vol. 146, Issue 49 https://doi.org/10.1021/jacs.4c11504	journal	November 2024
Functional materials discovery using energy–structure–function maps Pulido, Angeles; Chen, Linjiang; Kaczorowski, Tomasz Nature, Vol. 543, Issue 7647 https://doi.org/10.1038/nature21419	journal	March 2017
Towards the computational design of solid catalysts Nørskov, J.; Bligaard, T.; Rossmeisl, J. Nature Chemistry, Vol. 1, Issue 1, p. 37-46 https://doi.org/10.1038/nchem.121	journal	April 2009
The high-throughput highway to computational materials design Curtarolo, Stefano; Hart, Gus L. W.; Nardelli, Marco Buongiorno Nature Materials, Vol. 12, Issue 3 https://doi.org/10.1038/nmat3568	journal	February 2013
Deep neural networks for accurate predictions of crystal stability Ye, Weike; Chen, Chi; Wang, Zhenbin Nature Communications, Vol. 9, Issue 1 https://doi.org/10.1038/s41467-018-06322-x	journal	September 2018
Identifying an efficient, thermally robust inorganic phosphor host via machine learning Zhuo, Ya; Mansouri Tehrani, Aria; Oliynyk, Anton O. Nature Communications, Vol. 9, Issue 1 https://doi.org/10.1038/s41467-018-06625-z	journal	October 2018
Robust and synthesizable photocatalysts for CO2 reduction: a data-driven materials discovery Singh, Arunima K.; Montoya, Joseph H.; Gregoire, John M. Nature Communications, Vol. 10, Issue 1 https://doi.org/10.1038/s41467-019-08356-1	journal	January 2019
The role of decomposition reactions in assessing first-principles predictions of solid stability Bartel, Christopher J.; Weimer, Alan W.; Lany, Stephan npj Computational Materials, Vol. 5, Issue 1 https://doi.org/10.1038/s41524-018-0143-2	journal	January 2019
A critical examination of compound stability predictions from machine-learned formation energies Bartel, Christopher J.; Trewartha, Amalie; Wang, Qi npj Computational Materials, Vol. 6, Issue 1 https://doi.org/10.1038/s41524-020-00362-y	journal	July 2020
Perovskite synthesizability using graph neural networks Gu, Geun Ho; Jang, Jidon; Noh, Juhwan npj Computational Materials, Vol. 8, Issue 1 https://doi.org/10.1038/s41524-022-00757-z	journal	April 2022
Predicting the synthesizability of crystalline inorganic materials from the data of known material compositions Antoniuk, Evan R.; Cheon, Gowoon; Wang, George npj Computational Materials, Vol. 9, Issue 1 https://doi.org/10.1038/s41524-023-01114-4	journal	August 2023
Leveraging language representation for materials exploration and discovery Qu, Jiaxing; Xie, Yuxuan Richard; Ciesielski, Kamil M. npj Computational Materials, Vol. 10, Issue 1 https://doi.org/10.1038/s41524-024-01231-8	journal	March 2024
Inverse design in search of materials with target functionalities Zunger, Alex Nature Reviews Chemistry, Vol. 2, Issue 4 https://doi.org/10.1038/s41570-018-0121	journal	March 2018
The case for data science in experimental chemistry: examples and recommendations Yano, Junko; Gaffney, Kelly J.; Gregoire, John Nature Reviews Chemistry, Vol. 6, Issue 5 https://doi.org/10.1038/s41570-022-00382-w	journal	April 2022
Machine learning for molecular and materials science Butler, Keith T.; Davies, Daniel W.; Cartwright, Hugh Nature, Vol. 559, Issue 7715 https://doi.org/10.1038/s41586-018-0337-2	journal	July 2018
Unsupervised word embeddings capture latent knowledge from materials science literature Tshitoyan, Vahe; Dagdelen, John; Weston, Leigh Nature, Vol. 571, Issue 7763 https://doi.org/10.1038/s41586-019-1335-8	journal	July 2019
Anthropogenic biases in chemical reaction data hinder exploratory inorganic synthesis Jia, Xiwen; Lynch, Allyson; Huang, Yuheng Nature, Vol. 573, Issue 7773 https://doi.org/10.1038/s41586-019-1540-5	journal	September 2019
Autonomous chemical research with large language models Boiko, Daniil A.; MacKnight, Robert; Kline, Ben Nature, Vol. 624, Issue 7992 https://doi.org/10.1038/s41586-023-06792-0	journal	December 2023
Novel inorganic crystal structures predicted using autonomous simulation agents Ye, Weike; Lei, Xiangyun; Aykol, Muratahan Scientific Data, Vol. 9, Issue 1 https://doi.org/10.1038/s41597-022-01438-8	journal	June 2022
Leveraging large language models for predictive chemistry Jablonka, Kevin Maik; Schwaller, Philippe; Ortega-Guerrero, Andres Nature Machine Intelligence, Vol. 6, Issue 2 https://doi.org/10.1038/s42256-023-00788-1	journal	February 2024
Augmenting large language models with chemistry tools M. Bran, Andres; Cox, Sam; Schilter, Oliver Nature Machine Intelligence, Vol. 6, Issue 5 https://doi.org/10.1038/s42256-024-00832-8	journal	May 2024
Predicting synthesizability of crystalline materials via deep learning Davariashtiyani, Ali; Kadkhodaie, Zahra; Kadkhodaei, Sara Communications Materials, Vol. 2, Issue 1 https://doi.org/10.1038/s43246-021-00219-x	journal	November 2021
Accelerated chemical science with AI Back, Seoin; Aspuru-Guzik, Alán; Ceriotti, Michele Digital Discovery, Vol. 3, Issue 1 https://doi.org/10.1039/D3DD00213F	journal	January 2024
Fine-tuning GPT-3 for machine learning electronic and functional properties of organic molecules Xie, Zikai; Evangelopoulos, Xenophon; Omar, Ömer H. Chemical Science, Vol. 15, Issue 2 https://doi.org/10.1039/D3SC04610A	journal	January 2024
How the AI-assisted discovery and synthesis of a ternary oxide highlights capability gaps in materials science Montoya, Joseph H.; Grimley, Carolyn; Aykol, Muratahan Chemical Science, Vol. 15, Issue 15 https://doi.org/10.1039/D3SC04823C	journal	January 2024
Materials science in the era of large language models: a perspective Lei, Ge; Docherty, Ronan; Cooper, Samuel J. Digital Discovery, Vol. 3, Issue 7 https://doi.org/10.1039/D4DD00074A	journal	January 2024
Exploring the expertise of large language models in materials science and metallurgical engineering Bajan, Christophe; Lambard, Guillaume Digital Discovery, Vol. 4, Issue 2 https://doi.org/10.1039/D4DD00319E	journal	January 2025
Mapping inorganic crystal chemical space Park, Hyunsoo; Onwuli, Anthony; Butler, Keith T. Faraday Discussions, Vol. 256 https://doi.org/10.1039/D4FD00063C	journal	January 2025
A critical reflection on attempts to machine-learn materials synthesis insights from text-mined literature recipes Sun, Wenhao; David, Nicholas Faraday Discussions, Vol. 256 https://doi.org/10.1039/D4FD00112E	journal	January 2025
A review of large language models and autonomous agents in chemistry Ramos, Mayk Caldas; Collison, Christopher J.; White, Andrew D. Chemical Science, Vol. 16, Issue 6 https://doi.org/10.1039/D4SC03921A	journal	January 2025
Assessment of fine-tuned large language models for real-world chemistry and material science applications Van Herck, Joren; Gil, María Victoria; Jablonka, Kevin Maik Chemical Science, Vol. 16, Issue 2 https://doi.org/10.1039/D4SC04401K	journal	January 2025
Commentary: The Materials Project: A materials genome approach to accelerating materials innovation Jain, Anubhav; Ong, Shyue Ping; Hautier, Geoffroy APL Materials, Vol. 1, Issue 1 https://doi.org/10.1063/1.4812323	journal	July 2013
Atoms as words: A novel approach to deciphering material properties using NLP-inspired machine learning on crystallographic information files (CIFs) Yadav, Lalit AIP Advances, Vol. 14, Issue 4 https://doi.org/10.1063/5.0187741	journal	April 2024
The thermodynamic scale of inorganic crystalline metastability Sun, Wenhao; Dacek, Stephen T.; Ong, Shyue Ping Science Advances, Vol. 2, Issue 11 https://doi.org/10.1126/sciadv.1600225	journal	November 2016
Thermodynamic limit for synthesis of metastable inorganic materials Aykol, Muratahan; Dwaraknath, Shyam S.; Sun, Wenhao Science Advances, Vol. 4, Issue 4 https://doi.org/10.1126/sciadv.aaq0148	journal	April 2018
XAI—Explainable artificial intelligence Gunning, David; Stefik, Mark; Choi, Jaesik Science Robotics, Vol. 4, Issue 37 https://doi.org/10.1126/scirobotics.aay7120	journal	December 2019
Estimating the synthetic accessibility of molecules with building block and reaction-aware SAScore Chen, Shuan; Jung, Yousung Journal of Cheminformatics, Vol. 16, Issue 1 https://doi.org/10.1186/s13321-024-00879-0	journal	July 2024
Robocrystallographer: automated crystal structure text descriptions and analysis Ganose, Alex M.; Jain, Anubhav MRS Communications, Vol. 9, Issue 3 https://doi.org/10.1557/mrc.2019.94	journal	July 2019
Similarity of materials and data-quality assessment by fingerprinting Kuban, Martin; Gabaj, Šimon; Aggoune, Wahib MRS Bulletin, Vol. 47, Issue 10 https://doi.org/10.1557/s43577-022-00339-w	journal	September 2022
Recovering True Classifier Performance in Positive-Unlabeled Learning Jain, Shantanu; White, Martha; Radivojac, Predrag Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 31, Issue 1 https://doi.org/10.1609/aaai.v31i1.10937	journal	February 2017
Fast Nonparametric Estimation of Class Proportions in the Positive-Unlabeled Classification Setting Zeiberg, Daniel; Jain, Shantanu; Radivojac, Predrag Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 34, Issue 04 https://doi.org/10.1609/aaai.v34i04.6151	journal	April 2020

Similar Records

Explainable Synthesizability Prediction of Inorganic Crystal Polymorphs Using Large Language Models

Journal Article · Fri Mar 21 20:00:00 EDT 2025 · Angewandte Chemie (International Edition) · OSTI ID:2536759

Predicting the synthesizability of crystalline inorganic materials from the data of known material compositions

Journal Article · Thu Aug 24 20:00:00 EDT 2023 · npj Computational Materials · OSTI ID:1996869

Machine learned synthesizability predictions aided by density functional theory

Journal Article · Tue Oct 11 20:00:00 EDT 2022 · Communications Materials · OSTI ID:1891929

Explainable Synthesizability Prediction of Inorganic Crystal Polymorphs Using Large Language Models

Citation Formats

References (63)

Similar Records

Related Subjects