Active Learning-Guided Optimization of Moisture Swing Adsorption for Direct Air Capture.
- 1Department of Civil and Environmental Engineering, Case Western Reserve University, Cleveland, Ohio44106, United States.
- 2Department of Earth and Environmental Engineering, Columbia University, New York, New York10027, United States.
Abstract
Moisture swing adsorption (MSA) offers an energy-efficient route for direct air capture of CO2 by exploiting humidity-driven sorbent regeneration, yet its performance remains difficult to predict due to the coupled effects of sorbent chemistry, counterion properties, and operating conditions. Here, we present a data-efficient experimental and modeling strategy that systematically maps these interactions across a high-dimensional design space. Using active learning to target regions of maximal uncertainty, we expand experimental coverage from 19 to 77 conditions and, together with curated literature data, assemble a unified dataset of 290 data points (201 distinct experimental conditions) spanning 20 commercial ion-exchange resins. Interpretable machine-learning models trained on this dataset predict CO2 capture capacity (R2 = 0.702, RMSE = 0.164) and identify functional groups, anion basicity, humidity, and CO2 concentration as dominant performance determinants. Guided by these insights, we experimentally validate model-selected resin-anion combinations for two practical scenarios─maximum CO2 capture capacity and moisture-swing efficiency. This work converts fragmented experimental observations into a predictive structure-property framework and demonstrates a generalizable approach for understanding and optimizing complex adsorption systems under data and experimental constraints.
Related
- High-affinity phosphate capture on MgO/Mg(OH)2-decorated wattle bark biochar: adsorption mechanism and regeneration performance
- Natural CaCO₃ improves the moisture resistance and mechanical performance of hemp-starch biocomposites: A multi-objective optimization study
- Modeling and optimization of adsorption parameters for hexavalent chromium removal from wastewater using Box-Behnken design and response surface methodology
- Optimization of Turbidity and Organic Matter Removal from Palm Oil Mill Effluent by Batch Adsorption Using Gasification-Derived Pinus patula Biochar
- The role of soot oxidation on the optimization of chitosan/soot aerogel beads for heavy metals and dyes adsorption
BibTeX
@article{yu2026active,
title = {{Active Learning-Guided Optimization of Moisture Swing Adsorption for Direct Air Capture.}},
author = {Peiling Yu and Kai Zhang and Xiaoyang Shi and Zihan Zheng and Huichun Zhang},
journal = {Environmental science \& technology},
year = {2026},
month = oct,
doi = {10.1021/acs.est.6c07244},
url = {https://europepmc.org/article/MED/42915970},
note = {CC BY},
}
RIS
TY - JOUR
TI - Active Learning-Guided Optimization of Moisture Swing Adsorption for Direct Air Capture.
AU - Peiling Yu
AU - Kai Zhang
AU - Xiaoyang Shi
AU - Zihan Zheng
AU - Huichun Zhang
PY - 2026
DA - 2026/10/01
JO - Environmental science & technology
DO - 10.1021/acs.est.6c07244
UR - https://europepmc.org/article/MED/42915970
AB - Moisture swing adsorption (MSA) offers an energy-efficient route for direct air capture of CO2 by exploiting humidity-driven sorbent regeneration, yet its performance remains difficult to predict due to the coupled effects of sorbent chemistry, counterion properties, and operating conditions. Here, we present a data-efficient experimental and modeling strategy that systematically maps these interactions across a high-dimensional design space. Using active learning to target regions of maximal uncertainty, we expand experimental coverage from 19 to 77 conditions and, together with curated literature data, assemble a unified dataset of 290 data points (201 distinct experimental conditions) spanning 20 commercial ion-exchange resins. Interpretable machine-learning models trained on this dataset predict CO2 capture capacity (R2 = 0.702, RMSE = 0.164) and identify functional groups, anion basicity, humidity, and CO2 concentration as dominant performance determinants. Guided by these insights, we experimentally validate model-selected resin-anion combinations for two practical scenarios─maximum CO2 capture capacity and moisture-swing efficiency. This work converts fragmented experimental observations into a predictive structure-property framework and demonstrates a generalizable approach for understanding and optimizing complex adsorption systems under data and experimental constraints.
SN - 0013-936X
SN - 1520-5851
ER -