Overall
- Language
- english
- Conflict of Interest
- In relation to this article, we declare that there is no conflict of interest.
- Publication history
-
Received March 1, 2026
Revised April 6, 2026
Accepted April 10, 2026
Available online April 17, 2026
-
This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/bync/3.0) which permits
unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Copyright © KIChE. All rights reserved.
Most Cited
Machine Learning for Deep Eutectic Solvent Density: Impact of Feature Representations and Dataset Complexity on Predictive Reliability
Hongik University
Korean Chemical Engineering Research, May 2026, 64(2), 105163
https://doi.org/10.9713/kcer.2026.64.2.105163
https://doi.org/10.9713/kcer.2026.64.2.105163
Download PDF
Abstract
Accurately predicting the density of deep eutectic solvents (DESs) is crucial for optimizing green separation processes. This study investigated the impact of feature representations (ChemBERTa, hybrid, and critical property models) and data partitioning (binary, ternary, and comprehensive datasets) on machine learning predictions. Evaluating RF, XGBoost, CatBoost, and ANN models revealed that tree-based ensembles were highly robust, consistently achieving R > 0.93 on limited datasets. Conversely, ANNs required explicit physical descriptors or massive datasets (>12,000 points) to prevent overfitting. Cross-domain validations demonstrated that extrapolating from simple to complex systems failed due to restricted thermodynamic diversity, whereas specializing from a comprehensive dataset ensured excellent transferability. These findings established that combining large, diverse datasets with ensemble algorithms or physics-informed features were essential for the reliable computational design of multicomponent DES properties.
References
1. Costa, S. P. F., Azevedo, A. M. O., Pinto, P. C. A. G. and Sara- iva, M. L. M. F. S., “Environmental Impact of Ionic Liquids: Recent Advances in (eco)toxicology and (bio)degra-dability,” ChemSus- Chem, 10, 2321-2347(2017).
2. Płotka-Wasylka, J., Guardia, M., Andruch, V. and Vilková, M., “Deep Eutectic Solvents vs. Ionic Liquids: Similarities and Dif- ferences,” MicroChem. J., 159, 105539(2020).
3. Jung, M. and Park, Y., “Thermophysical Properties and Liquid- liquid Equilibria of Pseudoternary Systems {toluene + n-heptane + deep Eutectic Solvents Based on Levulinic Acid,” J. Chem. Eng. Data, 67, 416-427(2022).
4. Shang, M., Wang, P., Cheng, Y., Zhou, Y. and Lei, Z., “Separation of Azeotropes in the Dimethoxymethane Production Using Deep Eutectic Solvents: Liquid-liquid Extraction Experiments and Molec- ular Insights,” Fuel, 406, 137184(2026).
5. Wang, D., Wang, Y., Ma, R., Ma, J., Xu, M., Ai, L., Leng, C., Ma, Q., Jia, D., Wang, L. and Guo, N., “Efficient and Green Separation of Phenolics by Halogen Free Deep Eutectic Solvents,” J. Mol. Liquids, 417, 126602(2025).
6. Alnajjar, A. and Onaizi, S., “CO2 Capture and Conversion Using Deep Eutectics Solvents: Recent Progress and Outlooks,” J. Mol. Liquids, 421, 126832(2025).
7. Abranches, D. O., Silva, L. P., Martins, M. A. R., Pinho, S. P. and Coutinho, J. A. P., “Understanding the Formation of Deep Eutectic Solvents: Betaine as a Universal Hydrogen Bond Acceptor”, ChemSusChem, 13, 4916-4921(2020).
8. Mulyono, S., Hizaddin, H. F., Alnashef, I. M., Hashim, M. A., Fakeeha, A. F. and Hadj-Kali, M. K., “Separation of BTEX Aromat- ics from n-octane Using a (tetrabutylammonia bromide + sulfolane) Deep Eutectic Solvent – experiments and COSCO-RS Predic- tion,” RSC Adv., 4, 17597(2014).
9. Ji, Q., Yu, X., Yagoub, A. Chen, L. and Zhou, C., “Efficient Removal of Lignin from Vegetable Waste by Ultrasonic and Microwaves- assisted Treatment with Ternary Deep Eutectic Solvent,” Industrial Crops and Products, 149, 112357(2020).
10. Park, Y., The Solubility of Deep Eutectic Solvents Derived from Allyltriphenylphosphonium Bromide in Supercritical Carbon Diox- ide in the Presence of Ethanol as a Cosolvent, Korean J. Chem. Eng., 41, 2091-2097(2024).
11. Boublia, A., Lemaoui, T., Almustafa, G., Darwish, A. S., Bengu- erba, Y., Banat, F. and AlNashef, I. M., “Critical Properties of Ter- nary Deep Eutectic Solvents Using Group Contribution with Extended Lee–Kesler Mixing Rules,” ACS Omega, 8, 13177(2023).
12. Almustafa, G., Darwish, A. S., Lemaoui, T., O’Conner, M. J., Amin, S., Arafat, H. A. and AlNashef, I., Liquification of 2,2,4-trimethyl- 1,3-pentanediol Into Hydrophobic Eutectic Mixtures: A Multi- criteria Design for Eco-efficient Boron Recovery,” Chem. Eng. J., 426, 131342(2021).
13. Park, Y., Liquid-liquid Equilibria of Cylcohexene-cyclohexane with Betaine-glyceol DES: Experiments and Correlation,” Korean Chem. Eng. Res., 63, 334-340(2025).
14. Di Pietro, T., Cesari, L. and Mutelet, F., Group Contribution Models for Densities and Heat Capacities of Deep Eutectic Solvents,” Fluid Phase Eq., 572, 113854(2023).
15. Hu, J., Peng, D., Huang, X., Wang, N., Liu, B., Di, D., Liu, J., Qu, Q., Pei, D., “COSMO-SAC and QSPR Combined Models: A Flexible and Reliable Strategy for Screening the Extraction Efficiency of Deep Eutectic Solvents,” Sep. Purif. Technol., 315, 123699(2023).
16. Were, K., Bui, D. T., Dick, O. B. and Singh, B. R., “A Compar- ative Assessment of Support Vector Regression, Artificial Neu- ral Networks, and Random Forests for Predicting and Mapping Soil Organic Carbon Stocks Across an Afromontrane Landscape,” Ecol. Ind., 52, 394(2015).
17. Wang, Y. X., Hou, X. J., Zeng, J., Wu, K. J. and He, Y. C., Ran-dom Forest Models to Predict the Densities and Surface Ten- sions of Deep Eutectic Solvents,” AIChE J., 69, e18095(2023).
18. Chen, T. and Guestrin, C., “XGBoost: A Scalable Tree Boosting System, In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD’16), San Francisco, CA, USA, August 13–17, 2016; pp. 785–794.
19. Prokhorenkova, L., Gusev, G., Vorobev, A., Dorogush, A. V. and Gulin A., “CatBoost: Unbiased Boosting with Categorical Fea- tures,” In Proceedings of the 32nd International conference on neu- ral information processing systems, pp. 6639-6649.
20. Chithraranda, S., Grand, G. and Ramsundar, B., “ChemBERTa: Large-scale Self-supervised Pretraining for Molecular Property Prediction,” arXiv:2010.09885(2020).
21. Wu, T., Song, J., Hu, Y., Qiu, Q., Tang, J., Peng, W., Fu, X., Shi, C. and Lin, X., “Methodological Roadmap for Machine Learning in Deep Eutectic Solvent Research: A Framework-driven Review and Perspective,” Ind. Eng. Chem. Res., 64, 16443-16465(2025).
22. Omar, K. A. and Sadeghi, R., “Database of Deep Eutectic Sol- vents and Their Physical Properties: A Review,” J. Mol. Liquids, 384, 121899(2023).
23. Gajardo-Parra, N. F., Controneo-Figueroa, V. P., Aravena, P., Vesovic, V. and Canales, R. I., “Viscosity of Choline Chloride- based Deep Eutectic Solvents: Experiments and Modeling,” J. Chem. Eng. Data, 65, 5581-5591(2020).
24. Yadav, A., Kar, J. R., Verma, M., Naqvi, S. and Pandey, S., “Densities of Aqueous Mixtures of (choline chloride+ethylene glycol) and (choline chloride+malonic acid) Deep Eutectic Solvents in Tem- perature Range 283.15-363.15K,” Thermo. Acta, 600, 95-101 (2015).
25. Wu, T., Zhan, P., Chen, W., Lin, M., Qiu, Q., Hu, Y., Song, J. and Lin, X., “ChemBERTa Embeddings and Ensemble Learning for Prediction of Density and Melting Point of Deep Eutectic Solvents with Hybrid Features,” Comput. Chem. Eng., 196, 109065 (2025).
26. Tropsha, A., “Best Practices for QSAR Model Development, Validation, and Assessment,” Mol. Inf., 29, 476-488(2010).
27. Hawkins, D. M., “The Problem of Overfitting,” J. Chem. Inf. Com- put. Sci., 44, 1-12(2004).
28. Mjalli, F. S., “Mass Connectivity Index-based Density Prediction of Deep Eutectic Solvents,” Fluid Phase Equilibria, 409, 312- 317(2016).
29. Wu, Z., Ramsundar, B., Feinberg, E. N., Gomes, J., Geniesse, C., Pappu, A. S., Leswing, K. and Pande, V., “MoleculeNet: a Benchmark for Molecular Machine Learning,” Chem. Sci., 9, 513-530 (2018).
30. Goh, G. B., Hodas, N. O. and Vishnu, A., “Deep Learning for Computational Chemistry,” J. Comput. Chem., 38, 1291-1307(2017).
31. Liu, Y., Hong, W. and Cao, B., “Machine Learning for Predicting Thermodynamic Properties of Pure Fluids and Mixtures,” Energy, 188, 116091(2019).
32. Yang, K., Swanson, K., Jin, W., Coley, C., Eiden, P., Gao, H., Guzman-Perez, A., Hopper, T., Kelly, B., Mathea, M., Palmer, A., Settels, V., Jaakkola, T., Jensen, K. and Barzilay, R., Analyzing Learned Molecular Representations for Property Prediction,” J. Chem. Inf. Model., 59, 3370-3388(2019).
2. Płotka-Wasylka, J., Guardia, M., Andruch, V. and Vilková, M., “Deep Eutectic Solvents vs. Ionic Liquids: Similarities and Dif- ferences,” MicroChem. J., 159, 105539(2020).
3. Jung, M. and Park, Y., “Thermophysical Properties and Liquid- liquid Equilibria of Pseudoternary Systems {toluene + n-heptane + deep Eutectic Solvents Based on Levulinic Acid,” J. Chem. Eng. Data, 67, 416-427(2022).
4. Shang, M., Wang, P., Cheng, Y., Zhou, Y. and Lei, Z., “Separation of Azeotropes in the Dimethoxymethane Production Using Deep Eutectic Solvents: Liquid-liquid Extraction Experiments and Molec- ular Insights,” Fuel, 406, 137184(2026).
5. Wang, D., Wang, Y., Ma, R., Ma, J., Xu, M., Ai, L., Leng, C., Ma, Q., Jia, D., Wang, L. and Guo, N., “Efficient and Green Separation of Phenolics by Halogen Free Deep Eutectic Solvents,” J. Mol. Liquids, 417, 126602(2025).
6. Alnajjar, A. and Onaizi, S., “CO2 Capture and Conversion Using Deep Eutectics Solvents: Recent Progress and Outlooks,” J. Mol. Liquids, 421, 126832(2025).
7. Abranches, D. O., Silva, L. P., Martins, M. A. R., Pinho, S. P. and Coutinho, J. A. P., “Understanding the Formation of Deep Eutectic Solvents: Betaine as a Universal Hydrogen Bond Acceptor”, ChemSusChem, 13, 4916-4921(2020).
8. Mulyono, S., Hizaddin, H. F., Alnashef, I. M., Hashim, M. A., Fakeeha, A. F. and Hadj-Kali, M. K., “Separation of BTEX Aromat- ics from n-octane Using a (tetrabutylammonia bromide + sulfolane) Deep Eutectic Solvent – experiments and COSCO-RS Predic- tion,” RSC Adv., 4, 17597(2014).
9. Ji, Q., Yu, X., Yagoub, A. Chen, L. and Zhou, C., “Efficient Removal of Lignin from Vegetable Waste by Ultrasonic and Microwaves- assisted Treatment with Ternary Deep Eutectic Solvent,” Industrial Crops and Products, 149, 112357(2020).
10. Park, Y., The Solubility of Deep Eutectic Solvents Derived from Allyltriphenylphosphonium Bromide in Supercritical Carbon Diox- ide in the Presence of Ethanol as a Cosolvent, Korean J. Chem. Eng., 41, 2091-2097(2024).
11. Boublia, A., Lemaoui, T., Almustafa, G., Darwish, A. S., Bengu- erba, Y., Banat, F. and AlNashef, I. M., “Critical Properties of Ter- nary Deep Eutectic Solvents Using Group Contribution with Extended Lee–Kesler Mixing Rules,” ACS Omega, 8, 13177(2023).
12. Almustafa, G., Darwish, A. S., Lemaoui, T., O’Conner, M. J., Amin, S., Arafat, H. A. and AlNashef, I., Liquification of 2,2,4-trimethyl- 1,3-pentanediol Into Hydrophobic Eutectic Mixtures: A Multi- criteria Design for Eco-efficient Boron Recovery,” Chem. Eng. J., 426, 131342(2021).
13. Park, Y., Liquid-liquid Equilibria of Cylcohexene-cyclohexane with Betaine-glyceol DES: Experiments and Correlation,” Korean Chem. Eng. Res., 63, 334-340(2025).
14. Di Pietro, T., Cesari, L. and Mutelet, F., Group Contribution Models for Densities and Heat Capacities of Deep Eutectic Solvents,” Fluid Phase Eq., 572, 113854(2023).
15. Hu, J., Peng, D., Huang, X., Wang, N., Liu, B., Di, D., Liu, J., Qu, Q., Pei, D., “COSMO-SAC and QSPR Combined Models: A Flexible and Reliable Strategy for Screening the Extraction Efficiency of Deep Eutectic Solvents,” Sep. Purif. Technol., 315, 123699(2023).
16. Were, K., Bui, D. T., Dick, O. B. and Singh, B. R., “A Compar- ative Assessment of Support Vector Regression, Artificial Neu- ral Networks, and Random Forests for Predicting and Mapping Soil Organic Carbon Stocks Across an Afromontrane Landscape,” Ecol. Ind., 52, 394(2015).
17. Wang, Y. X., Hou, X. J., Zeng, J., Wu, K. J. and He, Y. C., Ran-dom Forest Models to Predict the Densities and Surface Ten- sions of Deep Eutectic Solvents,” AIChE J., 69, e18095(2023).
18. Chen, T. and Guestrin, C., “XGBoost: A Scalable Tree Boosting System, In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD’16), San Francisco, CA, USA, August 13–17, 2016; pp. 785–794.
19. Prokhorenkova, L., Gusev, G., Vorobev, A., Dorogush, A. V. and Gulin A., “CatBoost: Unbiased Boosting with Categorical Fea- tures,” In Proceedings of the 32nd International conference on neu- ral information processing systems, pp. 6639-6649.
20. Chithraranda, S., Grand, G. and Ramsundar, B., “ChemBERTa: Large-scale Self-supervised Pretraining for Molecular Property Prediction,” arXiv:2010.09885(2020).
21. Wu, T., Song, J., Hu, Y., Qiu, Q., Tang, J., Peng, W., Fu, X., Shi, C. and Lin, X., “Methodological Roadmap for Machine Learning in Deep Eutectic Solvent Research: A Framework-driven Review and Perspective,” Ind. Eng. Chem. Res., 64, 16443-16465(2025).
22. Omar, K. A. and Sadeghi, R., “Database of Deep Eutectic Sol- vents and Their Physical Properties: A Review,” J. Mol. Liquids, 384, 121899(2023).
23. Gajardo-Parra, N. F., Controneo-Figueroa, V. P., Aravena, P., Vesovic, V. and Canales, R. I., “Viscosity of Choline Chloride- based Deep Eutectic Solvents: Experiments and Modeling,” J. Chem. Eng. Data, 65, 5581-5591(2020).
24. Yadav, A., Kar, J. R., Verma, M., Naqvi, S. and Pandey, S., “Densities of Aqueous Mixtures of (choline chloride+ethylene glycol) and (choline chloride+malonic acid) Deep Eutectic Solvents in Tem- perature Range 283.15-363.15K,” Thermo. Acta, 600, 95-101 (2015).
25. Wu, T., Zhan, P., Chen, W., Lin, M., Qiu, Q., Hu, Y., Song, J. and Lin, X., “ChemBERTa Embeddings and Ensemble Learning for Prediction of Density and Melting Point of Deep Eutectic Solvents with Hybrid Features,” Comput. Chem. Eng., 196, 109065 (2025).
26. Tropsha, A., “Best Practices for QSAR Model Development, Validation, and Assessment,” Mol. Inf., 29, 476-488(2010).
27. Hawkins, D. M., “The Problem of Overfitting,” J. Chem. Inf. Com- put. Sci., 44, 1-12(2004).
28. Mjalli, F. S., “Mass Connectivity Index-based Density Prediction of Deep Eutectic Solvents,” Fluid Phase Equilibria, 409, 312- 317(2016).
29. Wu, Z., Ramsundar, B., Feinberg, E. N., Gomes, J., Geniesse, C., Pappu, A. S., Leswing, K. and Pande, V., “MoleculeNet: a Benchmark for Molecular Machine Learning,” Chem. Sci., 9, 513-530 (2018).
30. Goh, G. B., Hodas, N. O. and Vishnu, A., “Deep Learning for Computational Chemistry,” J. Comput. Chem., 38, 1291-1307(2017).
31. Liu, Y., Hong, W. and Cao, B., “Machine Learning for Predicting Thermodynamic Properties of Pure Fluids and Mixtures,” Energy, 188, 116091(2019).
32. Yang, K., Swanson, K., Jin, W., Coley, C., Eiden, P., Gao, H., Guzman-Perez, A., Hopper, T., Kelly, B., Mathea, M., Palmer, A., Settels, V., Jaakkola, T., Jensen, K. and Barzilay, R., Analyzing Learned Molecular Representations for Property Prediction,” J. Chem. Inf. Model., 59, 3370-3388(2019).
