Machine Learning in the Development of Adsorbents for Clean Energy Application and Greenhouse Gas Capture

Mai, Haoxin; Le, Tu C.; Chen, Dehong; Winkler, David A.; Caruso, Rachel A.

doi:10.1002/advs.202203899

Cited by 9 publications

(2 citation statements)

References 239 publications

(356 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Looking ahead, molecular simulations will continue to be an excellent complementary tool to experiments, but other computational techniques can be incorporated into the systematic search for adsorbents for CO 2 capture, including guided machine learning 50 to reliably cover the huge amount of materials and their available information. At the atomic and molecular levels, other techniques such as the implementation of reactive force fields 51 can be used for the detailed analysis of the chemical reactions of CO 2 with the adsorbents and other species.…”

Section: Discussionmentioning

confidence: 99%

Importance of Bridging Molecular and Process Modeling to Design Optimal Adsorbents for Large-Scale CO₂ Capture

Vega,

Bahamon

2023

Acc. Chem. Res.

View full text Add to dashboard Cite

Metrics & MoreArticle Recommendations CONSPECTUS: Carbon capture, utilization, and storage have been identified as key technologies to decarbonize the energy and industrial sectors. Although postcombustion CO 2 capture by absorption in aqueous amines is a mature technology, the required high regeneration energy, losses due to degradation and evaporation, and corrosion carry a high economic cost, precluding this technology to be used today at the scale required to mitigate climate change.Solid adsorbent-based systems with high CO 2 capacities, high selectivity, and lower regeneration energy are becoming an attractive alternative for this purpose. Conscious of this opportunity, the search for optimal adsorbents for the capture of CO 2 has become an urgent task. To accurately assess the performance of CO 2 separation by adsorption at the needed scale, adsorbents should be synthesized and fully characterized under the required operating conditions, and the proper design and simulation of the process should be implemented along with techno-economic and environmental assessments. Several works have examined pure CO 2 single-component adsorption or binary mixtures of CO 2 with nitrogen for different families of adsorbents, primarily addressing their CO 2 adsorption capacity and selectivity; however, very limited data is available under other conditions and/or with impurities, mainly due to the intensive experimental (modeling) efforts required for the large number of adsorbents to be studied, posing a challenge for their assessment under the needed conditions. In this regard, molecular simulations can be employed in synergy with experiments, reliably generating missing adsorption properties of mixtures while providing understanding at the molecular level of the mechanism of the adsorption process.This Account provides an outlook on strategies used for the rational design of materials for CO 2 capture from different sources from the understanding of the adsorption mechanism at the molecular level. We illustrate with practical examples from our work and others' work how molecular simulations can be reliably used to link the molecular knowledge of novel adsorbents for which limited data exist for CO 2 capture adsorption processes. Molecular simulation results of different adsorbents, including MOFs, zeolites, and carbon-based and silica-based materials, are discussed, focusing on understanding the role of physical and chemical adsorption obtained from simulations and quantifying the impact of impurities in the performance of the materials. Furthermore, simulation results can be used for screening adsorbents from basic key performance indicators, such as cycling the working capacity, selectivity, and energy requirement, or for feeding detailed dynamic models to assess their performance in swing adsorption processes on the industrial scale, additionally including monetized performance indicators such as operating expenses, equipment sizes, and compression cost. Moreover, we highlight the role of molecular simulations in guid...

show abstract

Section: Discussionmentioning

confidence: 99%

Importance of Bridging Molecular and Process Modeling to Design Optimal Adsorbents for Large-Scale CO₂ Capture

Vega,

Bahamon

2023

Acc. Chem. Res.

View full text Add to dashboard Cite

show abstract

“…[314] In brief, ML is a data-driven method that can build up relationships between variables using training data and appropriate algorithms to predict the structure or property of a material. [315,316] In recent years, algorithmic models based on ML such as artificial neural network (ANN) and random forest (RF), have been established to predict properties and performances of the BCMs according to the feedstock characteristics and processing conditions. ANN, one of the so called "black box" models, is able to reveal complicated relationships between multiple inputs and outputs without knowing the mathematical description of the phenomena.…”

Section: Applications Of Machine Learning In the Precise Control Of B...mentioning

confidence: 99%

Functional Carbon from Nature: Biomass‐Derived Carbon Materials and the Recent Progress of Their Applications

Zhang

et al. 2023

Advanced Science

View full text Add to dashboard Cite

Biomass is considered as a promising source to fabricate functional carbon materials for its sustainability, low cost, and high carbon content. Biomass-derived-carbon materials (BCMs) have been a thriving research field. Novel structures, diverse synthesis methods, and versatile applications of BCMs have been reported. However, there has been no recent review of the numerous studies of different aspects of BCMs-related research. Therefore, this paper presents a comprehensive review that summarizes the progress of BCMs related research. Herein, typical types of biomass used to prepare BCMs are introduced. Variable structures of BCMs are summarized as the performance and properties of BCMs are closely related to their structures. Representative synthesis strategies, including both their merits and drawbacks are reviewed comprehensively. Moreover, the influence of synthetic conditions on the structure of as-prepared carbon products is discussed, providing important information for the rational design of the fabrication process of BCMs. Recent progress in versatile applications of BCMs based on their morphologies and physicochemical properties is reported. Finally, the remaining challenges of BCMs, are highlighted. Overall, this review provides a valuable overview of current knowledge and recent progress of BCMs, and it outlines directions for future research development of BCMs.

show abstract

Machine Learning in the Development of Adsorbents for Clean Energy Application and Greenhouse Gas Capture

Mai

Chen

et al. 2022

Advanced Science

Self Cite

View full text Add to dashboard Cite

Addressing climate change challenges by reducing greenhouse gas levels requires innovative adsorbent materials for clean energy applications. Recent progress in machine learning has stimulated technological breakthroughs in the discovery, design, and deployment of materials with potential for high-performance and low-cost clean energy applications. This review summarizes basic machine learning methods-data collection, featurization, model generation, and model evaluation-and reviews their use in the development of robust adsorbent materials. Key case studies are provided where these methods are used to accelerate adsorbent materials design and discovery, optimize synthesis conditions, and understand complex feature-property relationships. The review provides a concise resource for researchers wishing to use machine learning methods to rapidly develop effective adsorbent materials with a positive impact on the environment.

show abstract

Machine Learning in the Development of Adsorbents for Clean Energy Application and Greenhouse Gas Capture

Cited by 9 publications

References 239 publications

Importance of Bridging Molecular and Process Modeling to Design Optimal Adsorbents for Large-Scale CO₂ Capture

Importance of Bridging Molecular and Process Modeling to Design Optimal Adsorbents for Large-Scale CO₂ Capture

Functional Carbon from Nature: Biomass‐Derived Carbon Materials and the Recent Progress of Their Applications

Machine Learning in the Development of Adsorbents for Clean Energy Application and Greenhouse Gas Capture

Contact Info

Product

Resources

About

Machine Learning in the Development of Adsorbents for Clean Energy Application and Greenhouse Gas Capture

Cited by 9 publications

References 239 publications

Importance of Bridging Molecular and Process Modeling to Design Optimal Adsorbents for Large-Scale CO2 Capture

Importance of Bridging Molecular and Process Modeling to Design Optimal Adsorbents for Large-Scale CO2 Capture

Functional Carbon from Nature: Biomass‐Derived Carbon Materials and the Recent Progress of Their Applications

Machine Learning in the Development of Adsorbents for Clean Energy Application and Greenhouse Gas Capture

Contact Info

Product

Resources

About

Importance of Bridging Molecular and Process Modeling to Design Optimal Adsorbents for Large-Scale CO₂ Capture

Importance of Bridging Molecular and Process Modeling to Design Optimal Adsorbents for Large-Scale CO₂ Capture