Analysis of Influencing Factors on the Gas Separation Performance of Carbon Molecular Sieve Membrane Using Machine Learning Technique

Abstract

Gas separation performance of the carbon molecular sieve (CMS) membrane is influenced by multiple factors including the microstructural characteristics of carbon and gas properties. In this work, the support vector regression (SVR) method as a machine learning technique was applied to the correlation between the gas separation performance, the multiple membrane structure, and gas characteristic factors of the self-manufactured CMS membrane. A simple quantitative index based on the Robeson’s upper bound line, which indicated the gas permeability and selectivity simultaneously, was proposed to measure the gas separation performance of CMS membrane. Based on the calculation results, the inferred key factors affecting the gas permeability of CMS membrane were the fractional free volume (FFV) of the precursor, the average interlayer spacing of graphite-like carbon sheet, and the final carbonization temperature. Moreover, the most influential factors for the gas separation performance were supposed to be the two structural factors of precursor influencing the porosity of CMS membrane, the carbon residue and the FFV, and the ratio of the gas kinetic diameters. The results would be helpful to the structural optimization and the separation performance improvement of CMS membrane.

Machine Learning-Assisted Exploration of a Two-Dimensional Nanoslit for Blast Furnace Gas Separation

“…Six ML algorithms were applied to predict the diffusion property of gas molecules, including kernel ridge regression (KRR), random forest regression (RFR), least absolute shrinkage and selection operator (LASSO), support vector regression (SVR), decision tree regression (DTR) and extreme gradient boosting regression (XGBR) . All training models were developed in Python 3.7 using the scikit-learn module .…”

“…Recently, machine learning technology has emerged as a powerful tool to effectively predict the diffusion phenomenon of gas molecules, , exhibiting an efficacy that would prove arduous to attain through typical methods. For instance, molecular dynamics data and experimental data were used to build a robust machine learning model for predicting properties such as density and diffusion of fuels .…”

Construction and analysis of a detailed new database for gas separation membranes

“…A database with 648 records (including 36 types of MOFs and 41 types of polymers) was built containing permeation information for both pure gases and gas mixtures. Pan et al (2022) investigated Support Vector Machine (SVM) and Multiple Linear Regression (MLR) models to analyze the multiple factors that can influence the performance of membranes composed of Carbon Molecular Sieves (CMS). A database with 399 records was constructed containing permeation data for 5 different gases (CO2, CH4, N2, O2, and H2).…”

Performance of amine‐functionalized MIL‐53 incorporated thin‐film nanocomposite Pebax membranes for CO2/CH4 mixed gas separation

“…That is, polymers with lower gas permeability have higher gas selectivity, conversely. 11,12 Metal-organic frameworks (MOFs), known as porous particles with the potential to surpass the upper bound of Robeson, have been broadly utilized in mixed matrix membranes (MMMs) for gas separation [13][14][15] by developing new organic compounds such as molecular sieve materials, 16,17 silica, 18,19 covalent organic frameworks, [20][21][22] zeolites, [23][24][25] and carbon nanotubes. 26,27 Pebax 2533 consists of polyethylene oxide (PEO) soft and polyamide (PA) hard segments.…”

“…-CH 3 thin-film nanocomposite over the PSF sublayer was investigated by Ranjbar et al For all investigated gases, the highest permeability and selectivity were found with a POP-CH 3 loading of 5 wt%. By increasing POP-CH 3 loading to 5%, the CO 2 /CH 4 and CO 2 /N 2 selectivity of TFN membranes increased from 10.55 and 69.02 to 14.04 and 94 16,. respectively, at an applied feed pressure of 10 bar.…”