Deep Learning Combined with IAST to Screen Thermodynamically Feasible MOFs for Adsorption-Based Separation of Multiple Binary Mixtures

This study demonstrates the coupling of a multipurpose multilayer perceptron (MLP) model that predicts single-component adsorption for a various molecules with ideal adsorption solution theory (IAST). The resulting computational framework predicts MOF separations properties for various binary mixtures at various compositions and pressures. The accuracy of the MLP+IAST framework was sufficiently high so that, for a given separation, MOFs in the 90th percentile from MLP+IAST-based screening contain ~87% of MOFs in the 95th percentile one would obtain from molecular simulation-based screening. Clustering algorithms were shown effective to identify so-called "privileged" MOFs that were high-performing for multiple separations. Free energy calculations were performed to determine privileged MOFs that were likely to be accesses synthetically, at least from a thermodynamic perspective.

Deep Learning Combined with IAST to Screen Thermodynamically Feasible MOFs for Adsorption-Based Separation of Multiple Binary Mixtures

This study demonstrates the coupling of a multipurpose multilayer perceptron (MLP) model that predicts single-component adsorption for a various molecules with ideal adsorption solution theory (IAST). The resulting computational framework predicts MOF separations properties for various binary mixtures at various compositions and pressures. The accuracy of the MLP+IAST framework was sufficiently high so that, for a given separation, MOFs in the 90th percentile from MLP+IAST-based screening contain ~87% of MOFs in the 95th percentile one would obtain from molecular simulation-based screening. Clustering algorithms were shown effective to identify so-called "privileged" MOFs that were high-performing for multiple separations. Free energy calculations were performed to determine privileged MOFs that were likely to be accesses synthetically, at least from a thermodynamic perspective.

Pretreatment Effects on Activated Carbon Adsorption of Humic Substances: Distributed Fictive Component Analysis

Diluted solutions of a peat water and a biological wastewater effluent were subjected to coagulation, ozonation, and chlorination. The effects of these pretreatments on the removal of humic substances by activated carbon adsorption were tested. Batch adsorption isotherms were analyzed using a distributed fictive component method, which assumed a logarithmic normal distribution of Freundlich K and a non-adsorbable fraction in the Ideal Adsorption Solution Theory for multicomponent adsorption. Coagulation treatment not only increased adsorbabilities but also decreased its heterogeneity in terms of multicomponent adsorbates. Ozonation decreased adsorbabilities and increased heterogeneity in adsorbabilities by producing weakly adsorbing compounds. The average Freundlich K decreased, while the Freundlich exponent, 1/n, and non-adsorbable fraction was increased with ozone consumption. Chlorination showed the same effects as ozonation. However, after ozonation, equilibrium capacity at low activated carbon doses was increased because of the reduced dissolved organic carbon concentration by ozonation.