Project
H2STAR: A Gate-Validated Materials-to-System Model for Hydrogen Adsorption Storage
A self-directed, validation-gated study and codebase connecting cryogenic adsorption thermodynamics on carbons to onboard tank performance.
Motivation
My experiments can measure what a hydrogenated carbon does and predict almost none of what it would do inside a working system. TPD shows hydrogen stored and released; it says nothing about how the same material behaves in a tank cycling through real temperatures and pressures. H2STAR is two things at once: a self-directed, graduate-level course in hydrogen adsorption thermodynamics, and a validated codebase that carries a sorbent’s measured isotherm all the way to system-level capacity numbers.
Methods
The model core is the modified Dubinin-Astakhov isotherm (m = 2, ε = α + βT), anchored to the canonical AX-21 activated-carbon hydrogen parameters from Richard, Bénard, and Chahine (2009): n_max = 71.6 mol/kg, α = 3080 J/mol, β = 18.9 J/(mol·K), P₀ = 1470 MPa. All gas densities route through CoolProp’s reference multiparameter equation of state, and the wrapper is validated to better than 0.1 percent against NIST tables, with the honest caveat, recorded in the project journal, that this gate tests my unit handling rather than the physics, since CoolProp and NIST implement the same reference equation. The excess-versus-absolute distinction, the field’s most common fatal error, is guarded structurally: a unit test fails if the 77 K excess isotherm lacks its physical interior maximum near 30 to 40 bar, and the tank model enforces a dual-bookkeeping invariant in which total hydrogen counted as absolute plus void gas must equal excess plus all-pore-and-void gas to one part in 10⁹. Isosteric heat is extracted by Clausius-Clapeyron finite differences with a built-in 4 to 7 kJ/mol sanity band for carbons at low coverage, cross-checked against the analytic Dubinin-Astakhov limit. A registry of named failure modes (excess/absolute conflation, ideal-gas density, unit errors, parahydrogen selection, and others) turns the field’s known mistakes into explicit tests.
Current status
The equation-of-state and isotherm layers are implemented and gate-validated. The tank and system layer, which counts adsorbed hydrogen plus compressed void gas and translates material wt% into system-level capacity against U.S. DOE targets, is in progress and sits on the critical path.
Future work
Completing the tank layer closes the loop from a published isotherm to an onboard capacity estimate with quantified uncertainty. The validated layers are also pinned dependencies for CHASM’s system-projection module, so H2STAR’s near-term milestones are scheduled deliverables rather than open-ended study.