We report the findings of a hackathon focused on exploring the diverse applications of large language models in molecular and materials science.

Most practical applications of solids in industry involve porous materials and\ud adsorption processes. A correct assessment of the equilibrium and kinetics of adsorption\ud is extremely important for the design and operation of adsorption based processes. In\ud our previous studies we focused on the evaluation of the equilibrium of CO2/CH4\ud adsorption on cherry stone-based carbons. In the present paper the kinetics of adsorption\ud of CO2 on two cherry stone-based activated carbons (CS-H2O and CS-CO2), previously\ud prepared in our laboratory, has been evaluated by means of transient breakthrough\ud experiments at different CO2/CH4 feed concentrations, at atmospheric pressure and\ud 30 °C. A commercial activated carbon, Calgon BPL, has also been evaluated for\ud reference purposes. Three models have been applied to estimate the rate parameters\ud during the adsorption of CO2 on these carbons, pseudo-first, pseudo-second and\ud Avrami´s fractional order kinetic models. Avrami´s model accurately predicted the\ud dynamic CO2 adsorption performance of the carbons for the different feed gas\ud compositions. To further investigate the mechanism of CO2 adsorption on CS-H2O, CSCO2\ud and Calgon BPL, intra-particle diffusion and Boyd´s film-diffusion models were\ud also evaluated. It was established that mass transfer during the adsorption of CO2 from\ud CO2/CH4 is a diffusion-based process and that the main diffusion mechanisms involved\ud are intra-particle and film diffusion. At the initial stages of adsorption, film diffusion\ud resistance governed the adsorption rate, whereas intra-particle diffusion resistance was\ud the predominant factor in the following stages of adsorption.This work has received financial support from the Spanish MINECO (Project ENE2011-23467), co-financed by the European Regional Development Fund (ERDF), and from the Gobierno del Principado de Asturias (PCTI 2013-2017 GRUPIN14-079). N.A-G. also acknowledges a fellowship awarded by the Spanish MINECO (FPI program), and co-financed by the European Social Fund.Peer reviewe

Large language models (LLMs) allow for the extraction of structured data from unstructured sources, such as scientific papers, with unprecedented accuracy and performance.