A process-level perspective of the impact of molecular force fields on the computational screening of MOFs for carbon capture


The question we pose in this study is to what extent the ranking of metal organic frameworks (MOFs) for adsorption-based carbon capture, and the selection of top performers identified in PSA process modelling, depends on the choice of the commonly available forcefields. To answer this question, we first generated distributions of CO2 and N2 adsorption isotherms via molecular simulation in 690 MOFs using six typical forcefields: the UFF or Dreiding sets of Lennard-Jones parameters, in combination with partial charges derived from ab initio calculations (DDEC scheme) or by charge equilibration (Qeq an EQeq schemes). We then conducted a systematic uncertainty quantification study using PSA process-level modelling. We observe that: (i) the ranking of MOFs significantly depends on the choice of forcefield; (ii) partial charge assignment is the prevailing source of uncertainty, and that charge equilibration schemes produce results which are inconsistent with ab initio-derived charges; iii) the choice of Lennard-Jones parameters is a considerable source of uncertainty. Our work highlights that is not really possible to obtain material rankings with high resolution using a single molecular modelling approach and that, as a minimum, some uncertainty should be estimated for the performance of MOFs shortlisted using high throughput computational screening workflows. Future prospects for computational screening studies are also discussed.


Supplementary material

Supplementary information
Forcefield parameters; mixture adsorption predictions and validation; additional PSA simulation details; process sensitivity analysis; machine learning surrogate model details; additional notes and figures to support the results and discussion; comparison of CALF-20 process-level performance with CRAFTED-u MOFs within the CL1 class.