(618b) MOFs and COFs for Alternative Operating Conditions for Hydrogen Storage Explored by Machine Learning

Hydrogen fuel is attractive to power vehicles without emitting carbon, but onboard storage of sufficiently densified hydrogen at moderate pressure remains a significant challenge. Adsorption-based storage in porous crystals such as metal–organic frameworks (MOFs) and covalent organic frameworks (COFs) is attractive to reduce the storage pressure. It is, however, unclear to what extent volumetric storage targets can be met under constraints of material design and choice of operating conditions. Although the number of potential MOFs and COFs is already extremely large, the fact that different operating conditions likely correspond to different optimal MOF/COF designs makes the multi-objective optimization of material plus swing conditions and overwhelming task even for molecular simulation.

To overcome this challenge and help elucidate the tandem of attainable values for volumetric hydrogen storage metrics and corresponding optimal MOF/COF design, we “computationally synthesized” a library of porous frameworks and performed 18 000+ grand canonical Monte Carlo simulations to calculate hydrogen loadings at several T, P conditions. Leveraging all of the generated data, then we trained, for the first time, a single artificial neural network capable of predicting hydrogen loadings at multiple T, P conditions. Using this neural network, we were able to test the studied frameworks for more than a hundred operating conditions. The studied frameworks are based on 17 pore topologies and feature alchemical catecholate sites: sites whose interaction with hydrogen was artificially and systematically modified within the range of density functional theory-calculated hydrogen–catecholate binding energies found in the literature.

The corresponding characteristic of the optimal MOF/COF design for the studied operating conditions were analyzed. Materials with the tetrahedrally connected dia and qtz topologies tended to outperform other types of crystals for each “level” of hydrogen–alchemical site interaction strength. Porous crystals simultaneously featuring void fractions and volumetric surface areas in the 0.7–0.9 and 1300–1800 m2/cm3 ranges, respectively, were more susceptible to improvements in deliverable capacity for the 100 bar/77 K ↔ 5 bar/160 K swing by tuning their interactions with hydrogen. The latter swing conditions produced the highest optimal deliverable capacity (62 g/L with a 10 kJ/mol heat of adsorption) among the tested swings, which was 138% higher than the optimal deliverable capacity for the 100 bar ↔ 5 bar swing at ambient temperature (26 g/L with a 17 kJ/mol heat of adsorption). However, the use of the trained neural network allowed us to estimate that, for the non-isothermal 77 K ↔ 160 K swing, reducing the storage pressure from 100 to 35 bar only reduces the attainable deliverable capacity to 59 g/L, which may be an acceptable trade-off due to safety and compression cost implications.

As our trained neural network only uses simple descriptors as input, modelers and experimentalists alike could potentially use it to rapidly pre-assess the hydrogen storage capabilities of newly proposed framework designs at various swing conditions.

REFERENCES

[1] G Anderson, B Schweitzer, R Anderson, DA Gomez-Gualdron, J. Phys. Chem. C, 2019, 123 (1), pp 120–130