Prediction of Carbon Dioxide Adsorption via Deep Learning

Zihao Zhang; Jennifer A. Schott; Miaomiao Liu; Hao Chen; Xiuyang Lu; Bobby G. Sumpter; Jie Fu; Sheng Dai

doi:10.1002/anie.201812363

Prediction of Carbon Dioxide Adsorption via Deep Learning

Zihao Zhang, Jennifer A. Schott, Miaomiao Liu, Hao Chen, Xiuyang Lu, Bobby G. Sumpter, Jie Fu, Sheng Dai

Research output: Contribution to journal › Article › peer-review

97 Scopus citations

Abstract

Porous carbons with different textural properties exhibit great differences in CO₂ adsorption capacity. It is generally known that narrow micropores contribute to higher CO₂ adsorption capacity. However, it is still unclear what role each variable in the textural properties plays in CO₂ adsorption. Herein, a deep neural network is trained as a generative model to direct the relationship between CO₂ adsorption of porous carbons and corresponding textural properties. The trained neural network is further employed as an implicit model to estimate its ability to predict the CO₂ adsorption capacity of unknown porous carbons. Interestingly, the practical CO₂ adsorption amounts are in good agreement with predicted values using surface area, micropore and mesopore volumes as the input values simultaneously. This unprecedented deep learning neural network (DNN) approach, a type of machine learning algorithm, exhibits great potential to predict gas adsorption and guide the development of next-generation carbons.

Original language	English
Pages (from-to)	259-263
Number of pages	5
Journal	Angewandte Chemie - International Edition
Volume	58
Issue number	1
DOIs	https://doi.org/10.1002/anie.201812363
State	Published - Jan 2 2019

Funding

This work was supported by the Division of Chemical Sciences, Geosciences, and Biosciences, Office of Basic Energy Sciences, US Department of Energy. J.F. was supported by the National Natural Science Foundation of China (No. 21436007, 21676243, 21706228), the Zhejiang Provincial Natural Science Foundation of China (No. LR17B060002). Z.Z. thanks the China Scholarship Council for financial support as a joint PhD student.

Keywords

CO adsorption
machine learning
porous carbon
textural properties

Access to Document

10.1002/anie.201812363

Cite this

@article{8e1e716aab204826bcb2c22f4e066134,

title = "Prediction of Carbon Dioxide Adsorption via Deep Learning",

abstract = "Porous carbons with different textural properties exhibit great differences in CO2 adsorption capacity. It is generally known that narrow micropores contribute to higher CO2 adsorption capacity. However, it is still unclear what role each variable in the textural properties plays in CO2 adsorption. Herein, a deep neural network is trained as a generative model to direct the relationship between CO2 adsorption of porous carbons and corresponding textural properties. The trained neural network is further employed as an implicit model to estimate its ability to predict the CO2 adsorption capacity of unknown porous carbons. Interestingly, the practical CO2 adsorption amounts are in good agreement with predicted values using surface area, micropore and mesopore volumes as the input values simultaneously. This unprecedented deep learning neural network (DNN) approach, a type of machine learning algorithm, exhibits great potential to predict gas adsorption and guide the development of next-generation carbons.",

keywords = "CO adsorption, machine learning, porous carbon, textural properties",

author = "Zihao Zhang and Schott, {Jennifer A.} and Miaomiao Liu and Hao Chen and Xiuyang Lu and Sumpter, {Bobby G.} and Jie Fu and Sheng Dai",

note = "Publisher Copyright: {\textcopyright} 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim",

year = "2019",

month = jan,

day = "2",

doi = "10.1002/anie.201812363",

language = "English",

volume = "58",

pages = "259--263",

journal = "Angewandte Chemie - International Edition",

issn = "1433-7851",

publisher = "wiley",

number = "1",

}

TY - JOUR

T1 - Prediction of Carbon Dioxide Adsorption via Deep Learning

AU - Zhang, Zihao

AU - Schott, Jennifer A.

AU - Liu, Miaomiao

AU - Chen, Hao

AU - Lu, Xiuyang

AU - Sumpter, Bobby G.

AU - Fu, Jie

AU - Dai, Sheng

PY - 2019/1/2

Y1 - 2019/1/2

N2 - Porous carbons with different textural properties exhibit great differences in CO2 adsorption capacity. It is generally known that narrow micropores contribute to higher CO2 adsorption capacity. However, it is still unclear what role each variable in the textural properties plays in CO2 adsorption. Herein, a deep neural network is trained as a generative model to direct the relationship between CO2 adsorption of porous carbons and corresponding textural properties. The trained neural network is further employed as an implicit model to estimate its ability to predict the CO2 adsorption capacity of unknown porous carbons. Interestingly, the practical CO2 adsorption amounts are in good agreement with predicted values using surface area, micropore and mesopore volumes as the input values simultaneously. This unprecedented deep learning neural network (DNN) approach, a type of machine learning algorithm, exhibits great potential to predict gas adsorption and guide the development of next-generation carbons.

AB - Porous carbons with different textural properties exhibit great differences in CO2 adsorption capacity. It is generally known that narrow micropores contribute to higher CO2 adsorption capacity. However, it is still unclear what role each variable in the textural properties plays in CO2 adsorption. Herein, a deep neural network is trained as a generative model to direct the relationship between CO2 adsorption of porous carbons and corresponding textural properties. The trained neural network is further employed as an implicit model to estimate its ability to predict the CO2 adsorption capacity of unknown porous carbons. Interestingly, the practical CO2 adsorption amounts are in good agreement with predicted values using surface area, micropore and mesopore volumes as the input values simultaneously. This unprecedented deep learning neural network (DNN) approach, a type of machine learning algorithm, exhibits great potential to predict gas adsorption and guide the development of next-generation carbons.

KW - CO adsorption

KW - machine learning

KW - porous carbon

KW - textural properties

UR - http://www.scopus.com/inward/record.url?scp=85058024245&partnerID=8YFLogxK

U2 - 10.1002/anie.201812363

DO - 10.1002/anie.201812363

M3 - Article

C2 - 30511416

AN - SCOPUS:85058024245

SN - 1433-7851

VL - 58

SP - 259

EP - 263

JO - Angewandte Chemie - International Edition

JF - Angewandte Chemie - International Edition

IS - 1

ER -

Prediction of Carbon Dioxide Adsorption via Deep Learning

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this