Publicação
Machine learning-driven prediction of deep eutectic solvents’ heat capacity for sustainable process design
| Resumo: | Heat capacity, a crucial physical property for chemical processes, is often understudied in Deep Eutectic Solvents (DESs), which in turn are promising green alternatives to environmentally hazardous conventional solvents. This work addresses this gap by developing a machine learning model to predict DES heat capacity and identify key structural features influencing it. We employed a dataset of 530 DESs with corresponding experimental heat capacity values. Quantum-chemical COSMO-RS-based descriptors, capturing detailed information about DES structures, were calculated for each data point. Various machine learning algorithms, namely k-Nearest Neighbours (kNN), Random Forests (RF), Neural Network Multilayer Perceptron (MLP), and Support Vector Machines (SVM) were explored alongside a linear model (Multiple Linear Regression, MLR). Hyperparameter optimisation ensured all models were fine-tuned for optimal performance. The most successful model, based on the MLP technique, achieved remarkably low Average Absolute Relative Deviation (AARD) values of 0.500 % and 3.999 % for the training and test sets, respectively. This signifies a significant improvement in prediction accuracy compared to traditional methods. Furthermore, by applying a SHapley Additive exPlanations (SHAP) analysis, we identified the most crucial structural factors within DES components that govern their heat capacity. This comprehensive investigation offers valuable insights that can pave the way for an efficient design of novel DESs in the future. |
|---|---|
| Autores principais: | Halder, Amit Kumar |
| Outros Autores: | Haghbakhsh, Reza; Ferreira, Elisabete S. C.; Duarte, Ana Rita C.; Cordeiro, M. Natália D. S. |
| Assunto: | COSMO-RS Deep eutectic solvents Heat capacity prediction Machine learning Thermodynamic modelling Electronic, Optical and Magnetic Materials Atomic and Molecular Physics, and Optics Condensed Matter Physics Spectroscopy Physical and Theoretical Chemistry Materials Chemistry |
| Ano: | 2025 |
| País: | Portugal |
| Tipo de documento: | artigo |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade Nova de Lisboa |
| Idioma: | inglês |
| Origem: | Repositório Institucional da UNL |
| _version_ | 1879446788970119168 |
|---|---|
| author | Halder, Amit Kumar |
| author2 | Haghbakhsh, Reza Ferreira, Elisabete S. C. Duarte, Ana Rita C. Cordeiro, M. Natália D. S. |
| author2_role | author author author author |
| author_facet | Halder, Amit Kumar Haghbakhsh, Reza Ferreira, Elisabete S. C. Duarte, Ana Rita C. Cordeiro, M. Natália D. S. |
| author_role | author |
| contributor_name_str_mv | LAQV@REQUIMTE DQ - Departamento de Química Elsevier RUN |
| country_str | PT |
| creators_json_txt | [{\"Person.name\":\"Halder, Amit Kumar\"},{\"Person.name\":\"Haghbakhsh, Reza\"},{\"Person.name\":\"Ferreira, Elisabete S. C.\"},{\"Person.name\":\"Duarte, Ana Rita C.\"},{\"Person.name\":\"Cordeiro, M. Natália D. S.\"}] |
| datacite.contributors.contributor.contributorName.fl_str_mv | LAQV@REQUIMTE DQ - Departamento de Química Elsevier RUN |
| datacite.creators.creator.creatorName.fl_str_mv | Halder, Amit Kumar Haghbakhsh, Reza Ferreira, Elisabete S. C. Duarte, Ana Rita C. Cordeiro, M. Natália D. S. |
| datacite.date.Accepted.fl_str_mv | 2025-01-15T00:00:00Z |
| datacite.date.available.fl_str_mv | 2025-07-09T21:16:34Z |
| datacite.date.embargoed.fl_str_mv | 2025-07-09T21:16:34Z |
| datacite.rights.fl_str_mv | http://purl.org/coar/access_right/c_abf2 |
| datacite.subjects.subject.fl_str_mv | COSMO-RS Deep eutectic solvents Heat capacity prediction Machine learning Thermodynamic modelling Electronic, Optical and Magnetic Materials Atomic and Molecular Physics, and Optics Condensed Matter Physics Spectroscopy Physical and Theoretical Chemistry Materials Chemistry |
| datacite.titles.title.fl_str_mv | Machine learning-driven prediction of deep eutectic solvents’ heat capacity for sustainable process design |
| dc.contributor.none.fl_str_mv | LAQV@REQUIMTE DQ - Departamento de Química Elsevier RUN |
| dc.creator.none.fl_str_mv | Halder, Amit Kumar Haghbakhsh, Reza Ferreira, Elisabete S. C. Duarte, Ana Rita C. Cordeiro, M. Natália D. S. |
| dc.date.Accepted.fl_str_mv | 2025-01-15T00:00:00Z |
| dc.date.available.fl_str_mv | 2025-07-09T21:16:34Z |
| dc.date.embargoed.fl_str_mv | 2025-07-09T21:16:34Z |
| dc.format.none.fl_str_mv | application/pdf |
| dc.identifier.none.fl_str_mv | http://hdl.handle.net/10362/185008 |
| dc.language.none.fl_str_mv | eng |
| dc.rights.none.fl_str_mv | http://purl.org/coar/access_right/c_abf2 |
| dc.subject.none.fl_str_mv | COSMO-RS Deep eutectic solvents Heat capacity prediction Machine learning Thermodynamic modelling Electronic, Optical and Magnetic Materials Atomic and Molecular Physics, and Optics Condensed Matter Physics Spectroscopy Physical and Theoretical Chemistry Materials Chemistry |
| dc.title.fl_str_mv | Machine learning-driven prediction of deep eutectic solvents’ heat capacity for sustainable process design |
| dc.type.none.fl_str_mv | http://purl.org/coar/resource_type/c_6501 |
| description | Heat capacity, a crucial physical property for chemical processes, is often understudied in Deep Eutectic Solvents (DESs), which in turn are promising green alternatives to environmentally hazardous conventional solvents. This work addresses this gap by developing a machine learning model to predict DES heat capacity and identify key structural features influencing it. We employed a dataset of 530 DESs with corresponding experimental heat capacity values. Quantum-chemical COSMO-RS-based descriptors, capturing detailed information about DES structures, were calculated for each data point. Various machine learning algorithms, namely k-Nearest Neighbours (kNN), Random Forests (RF), Neural Network Multilayer Perceptron (MLP), and Support Vector Machines (SVM) were explored alongside a linear model (Multiple Linear Regression, MLR). Hyperparameter optimisation ensured all models were fine-tuned for optimal performance. The most successful model, based on the MLP technique, achieved remarkably low Average Absolute Relative Deviation (AARD) values of 0.500 % and 3.999 % for the training and test sets, respectively. This signifies a significant improvement in prediction accuracy compared to traditional methods. Furthermore, by applying a SHapley Additive exPlanations (SHAP) analysis, we identified the most crucial structural factors within DES components that govern their heat capacity. This comprehensive investigation offers valuable insights that can pave the way for an efficient design of novel DESs in the future. |
| dirty | 0 |
| eu_rights_str_mv | openAccess |
| format | article |
| fulltext.url.fl_str_mv | https://run.unl.pt/bitstreams/04eaf875-7e6b-4da3-bb0b-471326e4f29c/download |
| funder_facet_str_mv | FCT{{{_:::_}}}Fundação para a Ciência e a Tecnologia FCT{{{_:::_}}}Fundação para a Ciência e a Tecnologia |
| funding.funder.alternateName_str_mv | FCT FCT |
| funding.funder.identifier_str_mv | http://doi.org/10.13039/501100001871 http://doi.org/10.13039/501100001871 |
| funding.funder.name_str_mv | Fundação para a Ciência e a Tecnologia Fundação para a Ciência e a Tecnologia |
| funding.identifier_str_mv | 2020.01423.CEECIND/CP1596/CT0003 2022.05803.CEECIND/CP1725/CT0003 |
| funding.name_str_mv | CEEC IND 3ed CEEC IND5ed |
| funding_str_mv | 2020.01423.CEECIND/CP1596/CT0003 info:eu-repo/grantAgreement/FCT/CEEC IND 3ed/2020.01423.CEECIND%2FCP1596%2FCT0003/PT 2022.05803.CEECIND/CP1725/CT0003 info:eu-repo/grantAgreement/FCT/CEEC IND5ed/2022.05803.CEECIND%2FCP1725%2FCT0003/PT |
| id | run_53d048a79dc18818481cf6850732bb48 |
| identifier.url.fl_str_mv | http://hdl.handle.net/10362/185008 |
| inst_facet_str | urn:organizationAcronym:unl{{{_:::_}}}Universidade Nova de Lisboa |
| instacron_str | unl |
| institution | Universidade Nova de Lisboa |
| instname_str | Universidade Nova de Lisboa |
| language | eng |
| network_acronym_str | run |
| network_name_str | Repositório Institucional da UNL |
| oai_identifier_str | oai:run.unl.pt:10362/185008 |
| organization_str_mv | urn:organizationAcronym:unl |
| person_str_mv | Halder, Amit Kumar Haghbakhsh, Reza Ferreira, Elisabete S. C. Duarte, Ana Rita C. Cordeiro, M. Natália D. S. |
| publishDate | 2025 |
| repo_facet_str | urn:repositoryAcronym:run{{{_:::_}}}Repositório Institucional da UNL |
| reponame_str | Repositório Institucional da UNL |
| repository_id_str | urn:repositoryAcronym:run |
| service_str_mv | urn:repositoryAcronym:run |
| spelling | engenHeat capacity, a crucial physical property for chemical processes, is often understudied in Deep Eutectic Solvents (DESs), which in turn are promising green alternatives to environmentally hazardous conventional solvents. This work addresses this gap by developing a machine learning model to predict DES heat capacity and identify key structural features influencing it. We employed a dataset of 530 DESs with corresponding experimental heat capacity values. Quantum-chemical COSMO-RS-based descriptors, capturing detailed information about DES structures, were calculated for each data point. Various machine learning algorithms, namely k-Nearest Neighbours (kNN), Random Forests (RF), Neural Network Multilayer Perceptron (MLP), and Support Vector Machines (SVM) were explored alongside a linear model (Multiple Linear Regression, MLR). Hyperparameter optimisation ensured all models were fine-tuned for optimal performance. The most successful model, based on the MLP technique, achieved remarkably low Average Absolute Relative Deviation (AARD) values of 0.500 % and 3.999 % for the training and test sets, respectively. This signifies a significant improvement in prediction accuracy compared to traditional methods. Furthermore, by applying a SHapley Additive exPlanations (SHAP) analysis, we identified the most crucial structural factors within DES components that govern their heat capacity. This comprehensive investigation offers valuable insights that can pave the way for an efficient design of novel DESs in the future.application/pdfenMachine learning-driven prediction of deep eutectic solvents’ heat capacity for sustainable process designHalder, Amit KumarHaghbakhsh, RezaFerreira, Elisabete S. C.Duarte, Ana Rita C.Cordeiro, M. Natália D. S.LAQV@REQUIMTEDQ - Departamento de QuímicaElsevierHostingInstitutionOrganizationalRUNe-mailmailto:run@unl.ptrun@unl.ptISSNIsPartOf0167-7322URNIsPartOfPURE: 121436495URNIsPartOfPURE UUID: e247afb7-9f86-48f9-9b12-623bdbe868feURNIsPartOfScopus: 85211970998URNIsPartOfORCID: /0000-0003-0800-0112/work/187642775DOIIsPartOf10.1016/j.molliq.2024.1267072025-07-09T21:16:34Z2025-01-152025-01-15T00:00:00ZHandlehttp://hdl.handle.net/10362/185008http://purl.org/coar/access_right/c_abf2open accessCOSMO-RSDeep eutectic solventsHeat capacity predictionMachine learningThermodynamic modellingElectronic, Optical and Magnetic MaterialsAtomic and Molecular Physics, and OpticsCondensed Matter PhysicsSpectroscopyPhysical and Theoretical ChemistryMaterials Chemistry2259940 bytesFundação para a Ciência e a TecnologiaDevelopment of Biocatalysts for Industrial Applications: From plastic degradation to the pharmaceutical chemistryinfo:eu-repo/grantAgreement/FCT/CEEC IND 3ed/2020.01423.CEECIND%2FCP1596%2FCT0003/PT2020.01423.CEECIND/CP1596/CT0003CEEC IND 3edCrossref Funder IDhttp://doi.org/10.13039/501100001871Fundação para a Ciência e a TecnologiaDeveloping extensive thermodynamic and machine learning models to predict various physical properties and phase equilibria of deep eutectic solvents and incorporation into a comprehensive commercial softwareinfo:eu-repo/grantAgreement/FCT/CEEC IND5ed/2022.05803.CEECIND%2FCP1725%2FCT0003/PT2022.05803.CEECIND/CP1725/CT0003CEEC IND5edCrossref Funder IDhttp://doi.org/10.13039/501100001871literaturehttp://purl.org/coar/resource_type/c_6501journal articlehttp://purl.org/coar/access_right/c_abf2application/pdffulltexthttps://run.unl.pt/bitstreams/04eaf875-7e6b-4da3-bb0b-471326e4f29c/download |
| spellingShingle | Machine learning-driven prediction of deep eutectic solvents’ heat capacity for sustainable process design Halder, Amit Kumar COSMO-RS Deep eutectic solvents Heat capacity prediction Machine learning Thermodynamic modelling Electronic, Optical and Magnetic Materials Atomic and Molecular Physics, and Optics Condensed Matter Physics Spectroscopy Physical and Theoretical Chemistry Materials Chemistry |
| status | SINGLETON |
| subject.fl_str_mv | COSMO-RS Deep eutectic solvents Heat capacity prediction Machine learning Thermodynamic modelling Electronic, Optical and Magnetic Materials Atomic and Molecular Physics, and Optics Condensed Matter Physics Spectroscopy Physical and Theoretical Chemistry Materials Chemistry |
| title | Machine learning-driven prediction of deep eutectic solvents’ heat capacity for sustainable process design |
| title_full | Machine learning-driven prediction of deep eutectic solvents’ heat capacity for sustainable process design |
| title_fullStr | Machine learning-driven prediction of deep eutectic solvents’ heat capacity for sustainable process design |
| title_full_unstemmed | Machine learning-driven prediction of deep eutectic solvents’ heat capacity for sustainable process design |
| title_short | Machine learning-driven prediction of deep eutectic solvents’ heat capacity for sustainable process design |
| title_sort | Machine learning-driven prediction of deep eutectic solvents’ heat capacity for sustainable process design |
| topic | COSMO-RS Deep eutectic solvents Heat capacity prediction Machine learning Thermodynamic modelling Electronic, Optical and Magnetic Materials Atomic and Molecular Physics, and Optics Condensed Matter Physics Spectroscopy Physical and Theoretical Chemistry Materials Chemistry |
| topic_facet | COSMO-RS Deep eutectic solvents Heat capacity prediction Machine learning Thermodynamic modelling Electronic, Optical and Magnetic Materials Atomic and Molecular Physics, and Optics Condensed Matter Physics Spectroscopy Physical and Theoretical Chemistry Materials Chemistry |
| url | http://hdl.handle.net/10362/185008 |
| visible | 1 |