Accelerated screening and assembly of promising MOFs with open Cu sites for isobutene/isobutane separation using a data-driven approach

Abstract

As the by-product of catalytic cracking or alkane dehydrogenation, isobutene (2-methyl-propylene) and isobutane (2-methyl-propane) are important chemical feedstocks, while the separation of their mixture is one challenging issue in petrochemical...

Computer-Aided Design of Covalent Organic Frameworks for SF6 Capture: The Combination of High-Throughput Screening and Machine Learning

“…To balance COFs with S ads,SFd 6 /Nd 2 and ΔW SFd 6 , we employed the APS to screen COFs. 18,19,51 The color code intuitively revealed that red signifies COFs with the best performance (400 < APS <760), yellow indicates COFs with better performance (200 < APS < 400), green denotes COFs with moderate performance (100 < APS < 200), and blue represents lower adsorption (APS < 100) (Figure 3a). The 20 COFs with best and better APS were selected in order of APS from high to low (Tables S5 and S6).…”

“…The selectivity and working capacity of these adsorbents surpassed those experimentally synthesized COFs under VSA conditions. 18 The examples adequately demonstrate the significance of highthroughput screening studies, as they can expedite progress in experimental research and provide guidance for experimental work.…”

Combination of High-Throughput Screening and Assembly to Discover Efficient Metal–Organic Frameworks on Kr/Xe Adsorption Separation

“…After the high-performance MOFs were screened out from the three databases, the XBGoost machine learning model should be built, and then the important metal nodes and organic linkers, i.e., key "genes", were extracted by feature engineering finally to crossly assemble novel MOFs based on the material genomics strategy. The XGBoost model, incorporating a regularization coefficient into the cost function, as well as a second-order Taylor expansion to overcome overfitting, 64 was found to be an effective model in predicting high-performance MOFs in CoRE MOF, G-MOF, and hMOF databases through high-throughput screening by our group, 32,65,66 surpassing those of ridge regression, LASSO, Elastic Net, SVM, Bayes regression, and ANN. Thus, we also selected the XGBoost ML model to predict high-performance MOFs in this work.…”

Copper(ii)-MOFs for bio-applications

“…26 Furthermore, this advantage is related to their high specific surface area, which makes them ideal for use in biocatalysis, gas separation and energy storage, but especially for bioapplications. 28 Each of the reported copper-based MOFs has many unique advantages that make them attractive for use in a wide variety of applications, from catalysis to medicine. [29][30][31] For example, HKUST-1 (also known as MOF-199, Cu-BTC and Basolite C300), is one of the copper-based MOFs developed for the detection of biological signals, being the most reported and used for the detection of glucose, proteins and other biological molecules, 32 but it has also proven useful for the detection of environmental contaminants, separation and purification of chemical compounds, as well as the detection and removal of harmful gases, as well as drug delivery.…”

“…26 Furthermore, this advantage is related to their high specific surface area, which makes them ideal for use in biocatalysis, gas separation and energy storage, but especially for bioapplications. 28…”