Abstract Adsorption-induced deformation in porous materials refers to the structural change that occurs due to the adsorption of guest molecules. This effect can be neglected for many systems because the variation in pore volume is practically zero. This fact led to the hypothesis of a rigid pore structure, which in turn gave rise to numerous adsorption models. However, for many new porous materials, such as metal-organic frameworks (MOFs), the effects of these deformations are significant and specific for different components and temperatures, requiring detailed models. The hybrid Monte Carlo/molecular dynamics method is the most direct approach for capturing volumetric deformation during adsorption. However, this method involves multiple iterations of Grand Canonical Monte Carlo (GCMC) and molecular dynamics simulations in an NPT ensemble (NPT-MD), which requires computer optimization to reduce computational time. In this work, we develop an efficient iterative GCMC/NPT-MD method, leveraging the GPU parallelization of gRASPA and GROMACS software, to study the effects of flexibility in IRMOF-1. Using this framework, we analyze the deformation of IRMOF-1 during the adsorption of CO ₂ and Ar, systematically evaluating the influence of key simulation parameters, including the size of the simulation box and the non-bonded cutoff radius.
Santos et al. (2026) studied this question.