Combining Machine Learning and Molecular Simulations to Unlock Gas Separation Potentials of MOF Membranes and MOF/Polymer MMMs

Abstract

Due to the enormous increase in the number of metal-organic frameworks (MOFs), combining molecular simulations with machine learning (ML) would be a very useful approach for the accurate and rapid assessment of the separation performances of thousands of materials. In this work, we combined these two powerful approaches, molecular simulations and ML, to evaluate MOF membranes and MOF/polymer mixed matrix membranes (MMMs) for six different gas separations: He/H2, He/N2, He/CH4, H2/N2, H2/CH4, and N2/CH4. Single-component gas uptakes and diffusivities were computed by grand canonical Monte Carlo (GCMC) and molecular dynamics (MD) simulations, respectively, and these simulation results were used to assess gas permeabilities and selectivities of MOF membranes. Physical, chemical, and energetic features of MOFs were used as descriptors, and eight different ML models were developed to predict gas adsorption and diffusion properties of MOFs. Gas permeabilities and membrane selectivities of 5249 MOFs and 31,494 MOF/polymer MMMs were predicted using these ML models. To examine the transferability of the ML models, we also focused on computer-generated, hypothetical MOFs (hMOFs) and predicted the gas permeability and selectivity of 1000 hMOF/polymer MMMs. The ML models that we developed accurately predict the uptake and diffusion properties of He, H2, N2, and CH4 gases in MOFs and will significantly accelerate the assessment of separation performances of MOF membranes and MOF/polymer MMMs. These models will also be useful to direct the extensive experimental efforts and computationally demanding molecular simulations to the fabrication and analysis of membrane materials offering high performance for a target gas separation.

Integrating Molecular Simulations with Machine Learning Guides in the Design and Synthesis of [BMIM][BF4]/MOF Composites for CO2/N2 Separation

“…As a result, four different ML models were developed to predict CO 2 and N 2 uptakes of MOFs and IL/MOF composites at 1 bar and 298 K using the tree-based pipeline optimization tool (TPOT) . For the development of the ML models, the Random Forest, , the Extra Tree, the GradientBoosting, and the Extreme Gradient Boosting (XGB) regressor algorithms were used with their optimized hyperparameters, as listed in Table S3. ML-predicted gas adsorption data of MOFs and IL/MOF composites were compared with the GCMC simulation results using the coefficient of determination ( R 2 ), mean absolute error (MAE), root-mean-square error (RMSE), and the Spearman rank-order correlation coefficient (SRCC) to assess the accuracy of the models.…”

Theoretical Design of MOFs and PSA Process for Efficient Separation of CF4/NF3

“…Nowadays, there have been over 1 000 000 MOFs and MOF-like materials, and screening the most promising material for CF 4 /NF 3 separation by experiments is less practical. Thus, high-throughput screening technology utilizing computational chemistry has become the choice for exploring MOFs with excellent adsorption separation performance that meets industrial needs. − Through high-throughput GCMC calculations, MOFs with excellent adsorption selectivity can be found quickly from the vast MOF database and the properties of MOFs can be systematically analyzed so that the method has performed a promising prospect for the gas separation application such as the separation of benzene/vinyl acetate, C 2 H 2 /CO 2 , , C 2 H 6 /CH 4 , and electronic gas Xe/Kr . MOFs have exhibited excellent adsorption performance in the field of electronic specialty gas separation.…”

“…Among them, the control of adsorption capacity by adjusting the temperature is called the temperature swing adsorption (TSA) process, 47 while the pressure swing adsorption (PSA) process is a representative adsorptionbased separation method where the adsorption occurs at the high pressure and desorption occurs under the low pressure. 10 We chose PSA technology rather than the TSA process to simulate the CF 4 /NF 3 separation process in industry due to its strengths of a large range of pressure control, short time consumption, low energy demand, and low investment cost. Currently, Henderson et al 44 have designed a PSA process for the production of purified NF 3 product gas from the feed gas contaminated with CF 4 impurities, showing that PSA adsorption technology is suitable for the separation of CF 4 / NF 3 mixtures.…”

Multi-Scale Computer-Aided Design of Covalent Organic Frameworks for CO2 Capture in Wet Flue Gas

“…Machine learning is an increasingly popular tool in material science due to its powerful ability in analyzing and extracting important features from big data. − The machine-learning-obtained useful information can contribute to design materials in a more effective way. For example, Qiao and co-workers combined machine learning and molecular fingerprint to identify the excellent bits (aromatic rings, double bonds, transition metals, halogens, and oxygen heterocycles) that could promote the non-methane hydrocarbon capture performance of MOFs.…”