skip to main content

DOE PAGESDOE PAGES

This content will become publicly available on December 4, 2019

Title: Prediction of Carbon Dioxide Adsorption via Deep Learning

Porous carbons with different textural properties exhibit great differences in CO 2 adsorption capacity. It is generally known that narrow micropores contribute to higher CO 2 adsorption capacity. However, it is still unclear what role each variable in the textural properties plays in CO 2 adsorption. Herein, a deep neural network is trained as a generative model to direct the relationship between CO 2 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 CO 2 adsorption capacity of unknown porous carbons. Interestingly, the practical CO 2 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.
Authors:
 [1] ;  [2] ;  [3] ;  [4] ;  [4] ;  [5] ;  [4] ; ORCiD logo [2]
  1. Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States); Univ. of Tennessee, Knoxville, TN (United States); Zhejiang Univ., Hangzhou (China)
  2. Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States); Univ. of Tennessee, Knoxville, TN (United States)
  3. Univ. of Tennessee, Knoxville, TN (United States)
  4. Zhejiang Univ., Hangzhou (China)
  5. Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Publication Date:
Grant/Contract Number:
AC05-00OR22725
Type:
Accepted Manuscript
Journal Name:
Angewandte Chemie (International Edition)
Additional Journal Information:
Journal Name: Angewandte Chemie (International Edition); Journal Volume: 130; Journal Issue: n/a; Journal ID: ISSN 1433-7851
Publisher:
Wiley
Research Org:
Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Sponsoring Org:
USDOE Office of Science (SC), Basic Energy Sciences (BES) (SC-22)
Country of Publication:
United States
Language:
English
Subject:
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CHEMISTRY; CO2 adsorption; machine learning; porous carbon; textural properties
OSTI Identifier:
1486930
Alternate Identifier(s):
OSTI ID: 1489089

Zhang, Zihao, Schott, Jennifer A., Liu, Miaomiao, Chen, Hao, Lu, Xiuyang, Sumpter, Bobby G., Fu, Jie, and Dai, Sheng. Prediction of Carbon Dioxide Adsorption via Deep Learning. United States: N. p., Web. doi:10.1002/anie.201812363.
Zhang, Zihao, Schott, Jennifer A., Liu, Miaomiao, Chen, Hao, Lu, Xiuyang, Sumpter, Bobby G., Fu, Jie, & Dai, Sheng. Prediction of Carbon Dioxide Adsorption via Deep Learning. United States. doi:10.1002/anie.201812363.
Zhang, Zihao, Schott, Jennifer A., Liu, Miaomiao, Chen, Hao, Lu, Xiuyang, Sumpter, Bobby G., Fu, Jie, and Dai, Sheng. 2018. "Prediction of Carbon Dioxide Adsorption via Deep Learning". United States. doi:10.1002/anie.201812363.
@article{osti_1486930,
title = {Prediction of Carbon Dioxide Adsorption via Deep Learning},
author = {Zhang, Zihao and Schott, Jennifer A. and Liu, Miaomiao and Chen, Hao and Lu, Xiuyang and Sumpter, Bobby G. and Fu, Jie and Dai, Sheng},
abstractNote = {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.},
doi = {10.1002/anie.201812363},
journal = {Angewandte Chemie (International Edition)},
number = n/a,
volume = 130,
place = {United States},
year = {2018},
month = {12}
}

Works referenced in this record:

Synthesis of Mesoporous Carbon Materials via Enhanced Hydrogen-Bonding Interaction
journal, April 2006
  • Liang, Chengdu; Dai, Sheng
  • Journal of the American Chemical Society, Vol. 128, Issue 16, p. 5316-5317
  • DOI: 10.1021/ja060242k

Sustainable carbon materials
journal, January 2015
  • Titirici, Maria-Magdalena; White, Robin J.; Brun, Nicolas
  • Chemical Society Reviews, Vol. 44, Issue 1, p. 250-290
  • DOI: 10.1039/C4CS00232F

High-Throughput Synthesis of Zeolitic Imidazolate Frameworks and Application to CO2 Capture
journal, February 2008

Carbon Dioxide Capture: Prospects for New Materials
journal, July 2010
  • D'Alessandro, Deanna?M.; Smit, Berend; Long, Jeffrey?R.
  • Angewandte Chemie International Edition, Vol. 49, Issue 35, p. 6058-6082
  • DOI: 10.1002/anie.201000431

CO2 capture by adsorption with nitrogen enriched carbons
journal, September 2007

Recent Progress in the Synthesis of Porous Carbon Materials
journal, August 2006