PDGPT: A large language model for acquiring phase diagram information in magnesium alloys

Abstract

Magnesium alloys, known for their lightweight advantages, are increasingly in demand across a range of applications, from aerospace to the automotive industry. With rising requirements for strength and corrosion resistance, the development of new magnesium alloy systems has become critical. Phase diagrams play a crucial role in guiding the magnesium alloy design by providing key insights into phase stability, composition, and temperature ranges, enabling the optimization of alloy properties and processing conditions. However, accessing and interpreting phase diagram data with thermodynamic calculation software can be complex and time-consuming, often requiring intricate calculations and iterative refinement based on thermodynamic models. To address this challenge, we introduce PDGPT, a ChatGPT-based large language model designed to streamline the acquisition of magnesium alloys Phase Diagram information with high efficiency and accuracy. Enhanced by prompt-engineering, supervised fine-tuning and retrieval-augmented generation, PDGPT leverages the predictive and reasoning capabilities of large language models along with computational phase diagram data. By combining large language models with traditional phase diagram research tools, PDGPT not only improves the accessibility of critical phase diagram information but also sets the stage for future advancements in applying large language models to materials science.

Keywords

large language model / phase diagram prediction / prompt-engineering / retrieval-augmented generation / supervised fine-tuning

References

1	Esmaily M, Svensson J, Fajardo S, et al. Fundamentals and advances in magnesium alloy corrosion. Prog Mater Sci. 2017;89:92-193.
2	Jayasathyakawin S, Ravichandran M, Baskar N, Anand Chairman C, Balasundaram R. Mechanical properties and applications of Magnesium alloy –review. Mater Today Proc. 2020;27:909-913.
3	Yang Y, Xiong X, Chen J, Peng X, Chen D, Pan F. Research advances in magnesium and magnesium alloys worldwide in 2020. J Magnesium Alloys. 2021;9(3):705-747.
4	Zhang L, Cao Z, Liu Y, Su G, Cheng L. Effect of Al content on the microstructures and mechanical properties of Mg–Al alloys. Mater Sci Eng. 2009;508(1-2):129-133.
5	Song J, She J, Chen D, Pan F. Latest research advances on magnesium and magnesium alloys worldwide. J Magnesium Alloys. 2020;8(1):1-41.
6	Mezbahul-Islam M, Mostafa AO, Medraj M. Essential magnesium alloys binary phase diagrams and their thermochemical data. J Mater. 2014;2014(1):704283-704333.
7	Saunders N, Miodownik AP. CALPHAD (Calculation of Phase Diagrams): A Comprehensive Guide. Elsevier;1998.
8	Szymanski NJ, Rendy B, Fei Y, et al. An autonomous laboratory for the accelerated synthesis of novel materials. Nature. 2023;624(7990):86-91.
9	Rao Z, Tung PY, Xie R, et al. Machine learning–enabled high-entropy alloy discovery. Science. 2022;378(6615):78-85.
10	Srinivasan S, Batra R, Luo D, et al. Machine learning the metastable phase diagram of covalently bonded carbon. Nat Commun. 2022;13(1):3251.
11	Terayama K, Han K, Katsube R, et al. Acceleration of phase diagram construction by machine learning incorporating Gibbs’phase rule. Scripta Mater. 2022;208:114335.
12	Zhao WX, Zhou K, Li J, et al. A Survey of Large Language Models. Published online November 24, 2023. Accessed August 28, 2024.
13	Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need.
14	Li B, Jiang G, Li N, Song C. Research on large-scale structured and unstructured data processing based on Large Language Model. In: Proceedings of the International Conference on Machine Learning, Pattern Recognition and Automation Engineering. ACM;2024:111-116.
15	Qian C, Tang H, Yang Z, Liang H, Liu Y. Can Large Language Models Empower Molecular Property Prediction? Published online July 14, 2023. Accessed November 19, 2024.
16	Zhang D, Liu W, Tan Q, et al. ChemLLM: A Chemical Large Language Model. Published online February 9, 2024. Accessed March 3, 2024.
17	Liu X, Guo Y, Li H, et al. DrugLLM: Open Large Language Model for Few-Shot Molecule Generation.
18	Boiko DA, MacKnight R, Kline B, Gomes G. Autonomous chemical research with large language models. Nature. 2023;624(7992):570-578.
19	Zheng Z, Zhang O, Borgs C, Chayes JT, Yaghi OM. ChatGPT chemistry assistant for text mining and the prediction of MOF synthesis. J Am Chem Soc. 2023;145(32):18048-18062.
20	Xie T, Wan Y, Huang W, et al. DARWIN Series: Domain Specific Large Language Models for Natural Science. Published online August 24, 2023. Accessed September 12, 2024.
21	Zhang H, Song Y, Hou Z, Miret S, Liu B. HoneyComb: A Flexible LLM-Based Agent System for Materials Science. Published online August 29, 2024. Accessed September 10, 2024.
22	Gupta T, Zaki M, Krishnan NMA, Mausam. Mausam. MatSciBERT: a materials domain language model for text mining and information extraction. npj Comput Mater. 2022;8(1):102.
23	Trewartha A, Walker N, Huo H, et al. Quantifying the advantage of domain-specific pre-training on named entity recognition tasks in materials science. Patterns. 2022;3(4):100488.
24	Buehler MJ. Generative retrieval-augmented ontologic graph and multiagent strategies for interpretive large language model-based materials design. ACS Eng Au. 2024;4(2):241-277.
25	Jablonka KM, Ai Q, Al-Feghali A, et al. 14 examples of how LLMs can transform materials science and chemistry: a reflection on a large language model hackathon. Digital Discovery. 2023;2(5):1233-1250.
26	Choudhary K. AtomGPT: atomistic generative pretrained transformer for forward and inverse materials design. J Phys Chem Lett. 2024;15 (27):6909-6917.
27	Liu Z, Su Y, Guo Y, et al. Post-pretraining Large Language Model Enabled Reverse Design of MOFs for Hydrogen Storage. Published online March 21, 2024.
28	Liang P, Tarfa T, Robinson JA, et al. Experimental investigation and thermodynamic calculation of the Al–Mg–Zn system. Thermochim Acta. 1998;314(1):87-110.
29	Soudani H, Kanoulas E, Hasibi F. Fine Tuning vs. Retrieval Augmented Generation for Less Popular Knowledge. Published online September 27, 2024. Accessed October 7, 2024.
30	Gunel B, Du J, Conneau A, Stoyanov V. Supervised Contrastive Learning for Pre-trained Language Model Fine-tuning. Published online April 2, 2021.
31	Peeperkorn M, Kouwenhoven T, Brown D, Jordanous A. Is Temperature the Creativity Parameter of Large Language Models? Published online May 1, 2024.
32	Gao Y, Xiong Y, Gao X, et al. Retrieval-Augmented Generation for Large Language Models: A Survey. Published online March 27, 2024.
33	Kojima T, Gu SS, Reid M, Matsuo Y, Iwasawa Y. Large Language Models Are Zero-Shot Reasoners. Published online January 29, 2023. Accessed September 12, 2024.
34	Nigh M. mattnigh/ChatGPT3-Free-Prompt-List. Published online September 24, 2024. Accessed September 25, 2024.