Discovery of Sustainable Refrigerants through Physics-Informed RL Fine-Tuning of Sequence Models

Adrien Goldszal, Diego Calanzone, Vincent Taboga, Pierre-Luc Bacon

Keywords: Refrigerant Molecule Discovery, Climate Change, Physics-Informed, Reinforcement Learning, Large-Language-Models

Abstract: Most refrigerants currently used in air-conditioning systems, such as hydrofluorocarbons, are potent greenhouse gases and are being phased down. Large-scale molecular screening has been applied to the search for alternatives, but in practice only about 300 refrigerants are known, and only a few additional candidates have been suggested without experimental validation. This scarcity of reliable data limits the effectiveness of purely data-driven methods. We present `Refgen`, a generative pipeline that integrates machine learning with physics-grounded inductive biases. Alongside fine-tuning for valid molecular generation, `Refgen` incorporates predictive models for critical properties, equations of state, thermochemical polynomials, and full vapor compression cycle simulations. These models enable reinforcement learning fine-tuning under thermodynamic constraints, enforcing consistency and guiding discovery toward molecules that balance efficiency, safety, and environmental impact. By embedding physics into the learning process, `Refgen` leverages scarce data effectively and enables de novo refrigerant discovery beyond the known set of compounds.

Release To Public: Yes, please release this paper to the public

Submission Number: 2