Hybrid Approach for Predicting Melting Points in Nonionic Eutectic Solvents Using Thermodynamics and Machine Learning
In this work, a hybrid approach combining solution thermodynamics and machine learning (ML) methods is presented as a means of estimating solid–liquid equilibria (SLE) in nonionic eutectic solvents. The models were developed based on a data set comprising 141 binary mixtures and 1668 experimental me...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | , |
| Published: |
2025
|
| Subjects: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1852018047942918144 |
|---|---|
| author | Dmitriy M. Makarov (1600432) |
| author2 | Vasiliy Golubev (21008304) Arkadiy M. Kolker (1600426) |
| author2_role | author author |
| author_facet | Dmitriy M. Makarov (1600432) Vasiliy Golubev (21008304) Arkadiy M. Kolker (1600426) |
| author_role | author |
| dc.creator.none.fl_str_mv | Dmitriy M. Makarov (1600432) Vasiliy Golubev (21008304) Arkadiy M. Kolker (1600426) |
| dc.date.none.fl_str_mv | 2025-07-31T14:13:34Z |
| dc.identifier.none.fl_str_mv | 10.1021/acs.iecr.5c01348.s001 |
| dc.relation.none.fl_str_mv | https://figshare.com/articles/dataset/Hybrid_Approach_for_Predicting_Melting_Points_in_Nonionic_Eutectic_Solvents_Using_Thermodynamics_and_Machine_Learning/29716544 |
| dc.rights.none.fl_str_mv | CC BY-NC 4.0 info:eu-repo/semantics/openAccess |
| dc.subject.none.fl_str_mv | Biophysics Biochemistry Medicine Microbiology Ecology Biological Sciences not elsewhere classified Mathematical Sciences not elsewhere classified Chemical Sciences not elsewhere classified Physical Sciences not elsewhere classified semiempirical associated solution predicting melting points nonionic eutectic solvents calculate melting points ml first predicts direct ml predictions one fitting parameter asl model ’ two fitting parameters work compares models fitting parameters parameter asl ω ′,< ω )), two versions machine learning k </ interchange energy hybrid approach heteroassociation constant developed based 7 k |
| dc.title.none.fl_str_mv | Hybrid Approach for Predicting Melting Points in Nonionic Eutectic Solvents Using Thermodynamics and Machine Learning |
| dc.type.none.fl_str_mv | Dataset info:eu-repo/semantics/publishedVersion dataset |
| description | In this work, a hybrid approach combining solution thermodynamics and machine learning (ML) methods is presented as a means of estimating solid–liquid equilibria (SLE) in nonionic eutectic solvents. The models were developed based on a data set comprising 141 binary mixtures and 1668 experimental melting points. The semiempirical Associated Solution and Lattice (ASL) method was employed to characterize the SLE in two versions: with one fitting parameter, representing the interchange energy (ASL(ω)), and with two fitting parameters, representing the interchange energy and the heteroassociation constant (ASL(ω′,<i>K</i>)). This work compares models for predicting mixture melting points using direct ML and a hybrid approach. In the hybrid method, ML first predicts the ASL model’s fitting parameters, which are then used to calculate melting points. The single-parameter ASL approach showed better predictive performance than both the two-parameter ASL and direct ML predictions, achieving the lowest average absolute deviation of 8.7 K. |
| eu_rights_str_mv | openAccess |
| id | Manara_ca1a06c1905ecf4fbc2387f52f6388fd |
| identifier_str_mv | 10.1021/acs.iecr.5c01348.s001 |
| network_acronym_str | Manara |
| network_name_str | ManaraRepo |
| oai_identifier_str | oai:figshare.com:article/29716544 |
| publishDate | 2025 |
| repository.mail.fl_str_mv | |
| repository.name.fl_str_mv | |
| repository_id_str | |
| rights_invalid_str_mv | CC BY-NC 4.0 |
| spelling | Hybrid Approach for Predicting Melting Points in Nonionic Eutectic Solvents Using Thermodynamics and Machine LearningDmitriy M. Makarov (1600432)Vasiliy Golubev (21008304)Arkadiy M. Kolker (1600426)BiophysicsBiochemistryMedicineMicrobiologyEcologyBiological Sciences not elsewhere classifiedMathematical Sciences not elsewhere classifiedChemical Sciences not elsewhere classifiedPhysical Sciences not elsewhere classifiedsemiempirical associated solutionpredicting melting pointsnonionic eutectic solventscalculate melting pointsml first predictsdirect ml predictionsone fitting parameterasl model ’two fitting parameterswork compares modelsfitting parametersparameter aslω ′,<ω )),two versionsmachine learningk </interchange energyhybrid approachheteroassociation constantdeveloped based7 kIn this work, a hybrid approach combining solution thermodynamics and machine learning (ML) methods is presented as a means of estimating solid–liquid equilibria (SLE) in nonionic eutectic solvents. The models were developed based on a data set comprising 141 binary mixtures and 1668 experimental melting points. The semiempirical Associated Solution and Lattice (ASL) method was employed to characterize the SLE in two versions: with one fitting parameter, representing the interchange energy (ASL(ω)), and with two fitting parameters, representing the interchange energy and the heteroassociation constant (ASL(ω′,<i>K</i>)). This work compares models for predicting mixture melting points using direct ML and a hybrid approach. In the hybrid method, ML first predicts the ASL model’s fitting parameters, which are then used to calculate melting points. The single-parameter ASL approach showed better predictive performance than both the two-parameter ASL and direct ML predictions, achieving the lowest average absolute deviation of 8.7 K.2025-07-31T14:13:34ZDatasetinfo:eu-repo/semantics/publishedVersiondataset10.1021/acs.iecr.5c01348.s001https://figshare.com/articles/dataset/Hybrid_Approach_for_Predicting_Melting_Points_in_Nonionic_Eutectic_Solvents_Using_Thermodynamics_and_Machine_Learning/29716544CC BY-NC 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/297165442025-07-31T14:13:34Z |
| spellingShingle | Hybrid Approach for Predicting Melting Points in Nonionic Eutectic Solvents Using Thermodynamics and Machine Learning Dmitriy M. Makarov (1600432) Biophysics Biochemistry Medicine Microbiology Ecology Biological Sciences not elsewhere classified Mathematical Sciences not elsewhere classified Chemical Sciences not elsewhere classified Physical Sciences not elsewhere classified semiempirical associated solution predicting melting points nonionic eutectic solvents calculate melting points ml first predicts direct ml predictions one fitting parameter asl model ’ two fitting parameters work compares models fitting parameters parameter asl ω ′,< ω )), two versions machine learning k </ interchange energy hybrid approach heteroassociation constant developed based 7 k |
| status_str | publishedVersion |
| title | Hybrid Approach for Predicting Melting Points in Nonionic Eutectic Solvents Using Thermodynamics and Machine Learning |
| title_full | Hybrid Approach for Predicting Melting Points in Nonionic Eutectic Solvents Using Thermodynamics and Machine Learning |
| title_fullStr | Hybrid Approach for Predicting Melting Points in Nonionic Eutectic Solvents Using Thermodynamics and Machine Learning |
| title_full_unstemmed | Hybrid Approach for Predicting Melting Points in Nonionic Eutectic Solvents Using Thermodynamics and Machine Learning |
| title_short | Hybrid Approach for Predicting Melting Points in Nonionic Eutectic Solvents Using Thermodynamics and Machine Learning |
| title_sort | Hybrid Approach for Predicting Melting Points in Nonionic Eutectic Solvents Using Thermodynamics and Machine Learning |
| topic | Biophysics Biochemistry Medicine Microbiology Ecology Biological Sciences not elsewhere classified Mathematical Sciences not elsewhere classified Chemical Sciences not elsewhere classified Physical Sciences not elsewhere classified semiempirical associated solution predicting melting points nonionic eutectic solvents calculate melting points ml first predicts direct ml predictions one fitting parameter asl model ’ two fitting parameters work compares models fitting parameters parameter asl ω ′,< ω )), two versions machine learning k </ interchange energy hybrid approach heteroassociation constant developed based 7 k |