Predicting the interfacial tension of CO2 and NaCl aqueous solution with machine learning

Abstract Achieving carbon neutrality requires effective strategies to reduce CO2 emissions, and geological sequestration of CO2 is considered among the most promising and economically viable options. The interfacial tension (IFT) between the CO2 and the surrounding liquid (underground salt water or...

Full description

Saved in:
Bibliographic Details
Main Authors: Kashif Liaqat, Daniel J. Preston, Laura Schaefer
Format: Article
Language:English
Published: Nature Portfolio 2025-07-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-10274-w
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849763692978110464
author Kashif Liaqat
Daniel J. Preston
Laura Schaefer
author_facet Kashif Liaqat
Daniel J. Preston
Laura Schaefer
author_sort Kashif Liaqat
collection DOAJ
description Abstract Achieving carbon neutrality requires effective strategies to reduce CO2 emissions, and geological sequestration of CO2 is considered among the most promising and economically viable options. The interfacial tension (IFT) between the CO2 and the surrounding liquid (underground salt water or brine, NaCl) is a key parameter that affects the storage capacity of CO2 in saline aquifers; however, the experimental measurement of IFT is often time-consuming, labor-intensive, and reliant on expensive equipment, and empirical correlations demonstrate a low level of accuracy. Machine learning (ML) techniques have been suggested as an alternative approach, and the current literature related to interfacial phenomena utilizes a wide array of basic and advanced ML models for predicting IFT, though often without a comparative analysis, raising the question of which model is most appropriate for this specific application. In this work, multiple machine learning models, including linear regression (LR), support vector machine (SVM), decision tree regressor (DTR), random forest regressor (RFR), and multilayer perceptron (MLP), are used to predict the IFT of the CO2 and aqueous solution of NaCl. Models are trained using an experimental dataset that covers a wide range of temperature, pressure, and salinity (NaCl) conditions for CO2-brine IFT. Hyperparameter tuning algorithms are utilized to optimize each model, and the performance is evaluated using metrics such as mean absolute error (MAE) and mean absolute percentage error (MAPE). The best-performing algorithms are found to be SVM and MLP, with a MAPE of 0.97% and 0.99% and a MAE of 0.39 mN/m and 0.40 mN/m, respectively. The linear regression model demonstrated the worst performance with a MAPE of 4.25% and an MAE of 1.7 mN/m. The feature importance analysis reveals that pressure is the main parameter affecting the IFT. Our findings indicate a notable enhancement in prediction accuracy over previous ML studies in this area. Moreover, the results from this study suggest that even the basic ML models that were investigated, when properly tuned and optimized, are sufficient for accurate IFT predictions. This demonstrates that ML models offer a cost-effective and efficient alternative to experimental methods, potentially optimizing designs for CO2 sequestration.
format Article
id doaj-art-acc2bfb033e54c44a0b98db37615b003
institution DOAJ
issn 2045-2322
language English
publishDate 2025-07-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-acc2bfb033e54c44a0b98db37615b0032025-08-20T03:05:21ZengNature PortfolioScientific Reports2045-23222025-07-0115111610.1038/s41598-025-10274-wPredicting the interfacial tension of CO2 and NaCl aqueous solution with machine learningKashif Liaqat0Daniel J. Preston1Laura Schaefer2Department of Mechanical Engineering, Rice UniversityDepartment of Mechanical Engineering, Rice UniversityDepartment of Mechanical Engineering, Rice UniversityAbstract Achieving carbon neutrality requires effective strategies to reduce CO2 emissions, and geological sequestration of CO2 is considered among the most promising and economically viable options. The interfacial tension (IFT) between the CO2 and the surrounding liquid (underground salt water or brine, NaCl) is a key parameter that affects the storage capacity of CO2 in saline aquifers; however, the experimental measurement of IFT is often time-consuming, labor-intensive, and reliant on expensive equipment, and empirical correlations demonstrate a low level of accuracy. Machine learning (ML) techniques have been suggested as an alternative approach, and the current literature related to interfacial phenomena utilizes a wide array of basic and advanced ML models for predicting IFT, though often without a comparative analysis, raising the question of which model is most appropriate for this specific application. In this work, multiple machine learning models, including linear regression (LR), support vector machine (SVM), decision tree regressor (DTR), random forest regressor (RFR), and multilayer perceptron (MLP), are used to predict the IFT of the CO2 and aqueous solution of NaCl. Models are trained using an experimental dataset that covers a wide range of temperature, pressure, and salinity (NaCl) conditions for CO2-brine IFT. Hyperparameter tuning algorithms are utilized to optimize each model, and the performance is evaluated using metrics such as mean absolute error (MAE) and mean absolute percentage error (MAPE). The best-performing algorithms are found to be SVM and MLP, with a MAPE of 0.97% and 0.99% and a MAE of 0.39 mN/m and 0.40 mN/m, respectively. The linear regression model demonstrated the worst performance with a MAPE of 4.25% and an MAE of 1.7 mN/m. The feature importance analysis reveals that pressure is the main parameter affecting the IFT. Our findings indicate a notable enhancement in prediction accuracy over previous ML studies in this area. Moreover, the results from this study suggest that even the basic ML models that were investigated, when properly tuned and optimized, are sufficient for accurate IFT predictions. This demonstrates that ML models offer a cost-effective and efficient alternative to experimental methods, potentially optimizing designs for CO2 sequestration.https://doi.org/10.1038/s41598-025-10274-wGeological sequestrationArtificial neural networksSupport vector machinesCO2-brineCarbon capture and storage
spellingShingle Kashif Liaqat
Daniel J. Preston
Laura Schaefer
Predicting the interfacial tension of CO2 and NaCl aqueous solution with machine learning
Scientific Reports
Geological sequestration
Artificial neural networks
Support vector machines
CO2-brine
Carbon capture and storage
title Predicting the interfacial tension of CO2 and NaCl aqueous solution with machine learning
title_full Predicting the interfacial tension of CO2 and NaCl aqueous solution with machine learning
title_fullStr Predicting the interfacial tension of CO2 and NaCl aqueous solution with machine learning
title_full_unstemmed Predicting the interfacial tension of CO2 and NaCl aqueous solution with machine learning
title_short Predicting the interfacial tension of CO2 and NaCl aqueous solution with machine learning
title_sort predicting the interfacial tension of co2 and nacl aqueous solution with machine learning
topic Geological sequestration
Artificial neural networks
Support vector machines
CO2-brine
Carbon capture and storage
url https://doi.org/10.1038/s41598-025-10274-w
work_keys_str_mv AT kashifliaqat predictingtheinterfacialtensionofco2andnaclaqueoussolutionwithmachinelearning
AT danieljpreston predictingtheinterfacialtensionofco2andnaclaqueoussolutionwithmachinelearning
AT lauraschaefer predictingtheinterfacialtensionofco2andnaclaqueoussolutionwithmachinelearning