Answers are grounded in this paper and cite their sources.
Grand canonical Monte Carlo (GCMC) isotherms are condensed into a few reported quantities: a Henry coefficient, an isosteric heat, fitted isotherm parameters, predicted mixture selectivities and a statement that each run was equilibrated. The post-processing code behind them is rarely tested. AdsorpQC is an open-source Python package that reads RASPA3 output and performs these steps; we validated each against known answers, using methane and carbon dioxide, pure and mixed, in silicalite-1 at 300 K. Henry coefficients extrapolated from the isotherms agreed with Widom insertion within 0.2%. The previous release, which fitted a line through the origin, was 4.5% low for methane and 22% low for carbon dioxide. The ideal adsorbed solution theory (IAST) solver agreed with pyIAST to 3 × 10⁻¹² in relative terms, and it predicted the mixture GCMC loadings within 1.5% up to 100 kPa and within 4% at 1 MPa. The previous release reported a Henry-limit ratio under the name IAST: it missed the rise of the CO₂/CH₄ selectivity from 5.9 to 6.9 at 1 MPa, and it was up to 7.7 times too high for model isotherms with unequal capacities. A stationarity test that compares the drift with its autocorrelation-corrected noise raised false alarms for 0.5% of stationary low-loading series, against 79% for the previous test. A small-sample information criterion selected the true isotherm model more reliably than AIC for two-parameter models but needed about ten points for four-parameter ones.
No takes yet. Share an insight, caveat, or question.
Hernández et al. (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: