Solid hydride storage screening
HydrideStorageScreening bridges material-level calculations and process gas
thermodynamics. An atomistic tool such as ASE and Quantum ESPRESSO supplies
composition, crystal volume and a reaction enthalpy and entropy. NeqSim
supplies the gas composition, temperature, pressure, and hydrogen fugacity.
The utility does not perform density functional theory or predict hydride
chemistry from a formula alone.
For a hydride with $n_H$ hydrogen atoms per formula unit, hydrogen-free host molar mass $M_{host}$, $Z$ formula units per cell, and unit-cell volume $V$:
\[C_{wt}=100\frac{n_H M_H}{M_{host}+n_H M_H}\] \[\rho_{H_2}=\frac{Z n_H M_H}{N_A V}\]The first result is wt% of a fully hydrogenated solid. The second is grams of hydrogen per litre of ideal crystal, with $V$ converted from ų to litres. These are theoretical upper limits. Neither accounts for bed porosity, containers, incomplete conversion, cycling losses or usable operating window.
For a specified desorption reaction with unit solid activities, and reaction enthalpy and entropy per mole of released H₂:
\[\Delta G_{des}=\Delta H_{des}-T\Delta S_{des}+RT\ln(f_{H_2}/1\,\mathrm{bara})\]At equilibrium, $\Delta G_{des}=0$. The gas fugacity is $f_{H_2}=\phi_{H_2}y_{H_2}P$, with absolute pressure $P$ in bara. Obtain $\phi_{H_2}$ and $y_{H_2}$ from a NeqSim gas-phase calculation; a pressure alone is an ideal-gas approximation. A positive $\Delta G_{des}$ suppresses desorption for the assumed products, while a negative value favours it.
import neqsim.thermo.util.hydrogen.HydrideStorageScreening;
double capacityWtPercent =
HydrideStorageScreening.gravimetricCapacityPercent(6.94 + 10.81, 3);
double crystalDensityGPerL =
HydrideStorageScreening.volumetricHydrogenDensityGPerL(3, 1, 29.45);
double entropyJPerMolH2K = 130.7;
double enthalpyJPerMolH2 = 457.7 * entropyJPerMolH2K;
double equilibriumTemperatureK = HydrideStorageScreening.equilibriumTemperatureK(
10.0, enthalpyJPerMolH2, entropyJPerMolH2K);
The numerical methods in this example are exercised in
HydrideStorageScreeningTest. The LiBH₃ composition and 29.45 ų cell volume
are from Geldasa and Dejene (2026), DOI:10.1007/s00894-026-06861-x.
The illustrative 457.7 K reference temperature is their LiCuH₃ Table 5
result. It sets a separate reaction-parameter example and must not be
assigned to LiBH₃. Their abstract instead reports 382.7 K for LiCuH₃;
the discrepancy needs resolution before quantitative material ranking.
Formation energy in eV/atom cannot be substituted for desorption enthalpy in
J/mol H₂. The identity and free energy of competing solid products must be
checked before treating any pressure curve as a realizable phase boundary.
Further limits: constant $\Delta H$/$\Delta S$, a 1 bara fugacity standard, no temperature-dependent heat capacities, solid nonideality, phase transitions, diffusion, kinetics, hysteresis or cycling. The result is a thermodynamic screening calculation, not a validated material or vessel design.