The multi‐region thermodynamic Langmuir‐area model: A predictive tool for non‐ideal binary adsorption

Abstract BACKGROUND This study introduces two novel predictive models for binary adsorption equilibrium, developed from the classical Langmuir framework: the multi‐region thermodynamic Langmuir (MRTL) model and the MRTL‐area model, which explicitly incorporate the adsorbed molar area. Both models utilize the adsorption non‐random two‐liquid (aNRTL) model to calculate activity coefficients, accounting for non‐idealities in the adsorbed phase. While the MRTL model accounts for activity coefficients and the maximum adsorption capacity, the MRTL‐area model further refines the physical representation by dividing the adsorbent surface into areas occupied by each molecule. The models were validated against 21 single‐component and 12 binary adsorption systems from the NIST/ARPA‐E database. RESULTS The MRTL‐area model demonstrated superior performance, achieving an average root mean square error of 0.0737 for binary equilibria, outperforming not only the proposed MRTL but also established models like the ideal adsorbed solution theory (IAST) and the vacancy solution model with Wilson activity coefficients (VSM‐Wilson). CONCLUSION The key advancement of the MRTL‐area model lies in its ability to more accurately capture competitive adsorption mechanisms by accounting for molecular size and shape, resulting in a more thermodynamically consistent and predictive tool for the design and optimization of adsorption‐based separation processes. © 2026 Society of Chemical Industry (SCI).