Prediction by Convolutional Neural Networks of CO₂/N₂ Selectivity in Porous Carbons from N₂ Adsorption Isotherm at 77 K

Abstract

Porous carbons are an important class of porous materials with many applications, including gas separation. An N 2 adsorption isotherm at 77 K is the most widely used approach to characterize porosity. Conventionally, textual properties such as surface area and pore volumes are derived from the N 2 adsorption isotherm at 77 K by fitting it to adsorption theory and then correlating it to gas separation performance (uptake and selectivity). Here the N 2 isotherm at 77 K was used directly as input (representing feature descriptors for the porosity) to train convolutional neural networks to predict gas separation performance (using CO 2 /N 2 as a test case) for porous carbons. The porosity space for porous carbons was explored for higher CO 2 /N 2 selectivity. Porous carbons with a bimodal pore-size distribution of well-separated mesopores (3-7 nm) and micropores (< 2 nm) were found to be most promising. This work will be useful in guiding experimental research of porous carbons with the desired porosity for gas separation and other applications.

Facile Construction of Robust and Powerful Biochar-Decorated Porous Foams with Super-Wettability for Automatic and Continuous Oil Spill Remediation

“…Large specific surface area and pore size will contribute to more active sites and continuous molecule diffusion pathways. 38 Therefore, it is expected that the constructed secondary nanostructures will help improve the performance of BMCF in oil absorption. Owing to the highly porous structure, the obtained BMCF exhibits ultralight weight characteristics with very low densities (∼7.5 mg cm −3 ).…”

“…N 2 gas absorption−desorption experiments are performed to evaluate the specific surface area and the pore-size distribution of the obtained samples. As shown in 38 Starting from 874 m 2 g −1 for CF, the BET surface area increased to 994 m 2 g −1 for BMCF, meaning that the introduction of the biochar mesh on the skeleton forms a larger surface area. Additionally, the pore sizes of the two samples are mainly distributed in the range of 0.5−2.0 nm, and the average pore sizes increase from about 0.8 nm for CF to 1.2 nm for BMCF.…”

“…Figure2c, both CF and BMCF show a typical type IV isotherm with strong N 2 absorption at a low relative pressure, which indicates the presence of abundant micropores 38. Starting from 874 m 2 g −1 for CF, the BET surface area increased to 994 m 2 g −1 for BMCF, meaning that the introduction of the biochar mesh on the skeleton forms a larger surface area.…”

Quantitative Structure–Property Relationship Analysis for the Prediction of Propylene Adsorption Capacity in Pure Silicon Zeolites at Various Pressure Levels

“…For instance, Durá et al 21 innovatively made a reasonable fitting for CO 2 adsorption capacity in porous carbons through a simple regression approach derived from microporous and mesoporous volumes. Subsequently, artificial neural network (ANN) methods using microporous and mesoporous volumes and Brunauer–Emmett–Teller surface area were used to predict CO 2 uptake, 22 nitrogen (N 2 ) uptake under ambient conditions, and CO 2 /N 2 selectivity 23 , 24 of porous carbons. Fernandez et al 31 accurately predicted methane (CH 4 ) uptake of MOFs at 100 bar based only on the dominant pore diameter, the maximum pore diameter, the void fraction, the gravimetric surface area, the volumetric surface area, and the framework density.…”