Pub Date : 2025-12-03DOI: 10.1016/j.fluid.2025.114651
Yi Zhang , Xiaowei Cheng , Shiyu Sun , Baoshu Liu , Hua Sun
The solubility of active pharmaceutical ingredients is vital throughout the drug design, development processes and manufacture. However, solubility prediction remains a challenging task in the pharmaceutical field. Therefore, BCS class II drugs solubility prediction model was developed on the basis of the machine learning algorithms and molecular descriptors through Bayesian Optimization, cosine similarity and sparse principal component analyses, revealing XGBoost model exhibited the better accuracy and suitability. Besides, the generalization of XGBoost model was confirmed by the solubility data prediction in the uncommon solvents and unseen solutes. Influences of molecular descriptors on the predicted solubility data were evaluated through Shapley Additive Explanations analysis, exposing the temperature exhibited a positive effect on the predicted solubility and the double bonds number of the solvent molecule presented a negative effect on the predicted solubility data. The various molecular descriptor contributions to the solubility prediction of XGBoost model were analyzed through feature importance, exposing the molecular descriptor contributions followed the order: Chi0 > SMR_VSA1 > MolMR > ExactMolWt > T > NumValenceElectrons > fr_C_O. In addition, it revealed the studied molecular descriptors must synergistically contribute to the solubility data prediction of XGBoost model according to prediction results comparison of simple and original XGBoost models.
{"title":"Solubility prediction of BCS class II drugs through combining machine learning and molecular descriptor","authors":"Yi Zhang , Xiaowei Cheng , Shiyu Sun , Baoshu Liu , Hua Sun","doi":"10.1016/j.fluid.2025.114651","DOIUrl":"10.1016/j.fluid.2025.114651","url":null,"abstract":"<div><div>The solubility of active pharmaceutical ingredients is vital throughout the drug design, development processes and manufacture. However, solubility prediction remains a challenging task in the pharmaceutical field. Therefore, BCS class II drugs solubility prediction model was developed on the basis of the machine learning algorithms and molecular descriptors through Bayesian Optimization, cosine similarity and sparse principal component analyses, revealing XGBoost model exhibited the better accuracy and suitability. Besides, the generalization of XGBoost model was confirmed by the solubility data prediction in the uncommon solvents and unseen solutes. Influences of molecular descriptors on the predicted solubility data were evaluated through Shapley Additive Explanations analysis, exposing the temperature exhibited a positive effect on the predicted solubility and the double bonds number of the solvent molecule presented a negative effect on the predicted solubility data. The various molecular descriptor contributions to the solubility prediction of XGBoost model were analyzed through feature importance, exposing the molecular descriptor contributions followed the order: <em>Chi0</em> > <em>SMR_VSA1</em> > <em>MolMR</em> > <em>ExactMolWt</em> > <em>T</em> > <em>NumValenceElectrons</em> > <em>fr_C_O</em>. In addition, it revealed the studied molecular descriptors must synergistically contribute to the solubility data prediction of XGBoost model according to prediction results comparison of simple and original XGBoost models.</div></div>","PeriodicalId":12170,"journal":{"name":"Fluid Phase Equilibria","volume":"603 ","pages":"Article 114651"},"PeriodicalIF":2.7,"publicationDate":"2025-12-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145682347","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-02DOI: 10.1016/j.fluid.2025.114650
Adina Werner, Jongmin Kim, Fabian Mauss
Excess volumes can be calculated generally via equations of state. In this work, the excess volumes are obtained using the UNIQUAC model with two approaches of a temperature- and pressure-dependent binary interaction parameter. The pressure dependency is required as the excess volume is derived from the pressure dependency of the excess free enthalpy. Both UNIQUAC approaches are successfully able to predict the vapor-liquid equilibrium as well as the excess volume of methanol-water mixtures over a temperature range between 288.15–473 K and a pressure range between 0.1519–134 bar using a single optimized parameter set.
{"title":"Excess volumes calculated from UNIQUAC model using the example of methanol - water mixtures","authors":"Adina Werner, Jongmin Kim, Fabian Mauss","doi":"10.1016/j.fluid.2025.114650","DOIUrl":"10.1016/j.fluid.2025.114650","url":null,"abstract":"<div><div>Excess volumes can be calculated generally via equations of state. In this work, the excess volumes are obtained using the UNIQUAC model with two approaches of a temperature- and pressure-dependent binary interaction parameter. The pressure dependency is required as the excess volume is derived from the pressure dependency of the excess free enthalpy. Both UNIQUAC approaches are successfully able to predict the vapor-liquid equilibrium as well as the excess volume of methanol-water mixtures over a temperature range between 288.15–473 K and a pressure range between 0.1519–134 bar using a single optimized parameter set.</div></div>","PeriodicalId":12170,"journal":{"name":"Fluid Phase Equilibria","volume":"603 ","pages":"Article 114650"},"PeriodicalIF":2.7,"publicationDate":"2025-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145682426","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Two binary systems containing CO2 and a branched alkane were newly studied in this work: CO2 + 2,3,4-trimethylpentane and CO2 + 2,2,4,6,6-pentamethylheptane. Their fluid phase behavior was investigated using a high-pressure variable-volume cell and a synthetic method. Saturation pressures were visually measured from (293.15 et 363.15) K and for 10 different CO2 mole fractions for both mixtures.
Whatever the considered system, no liquid-liquid immiscibility was observed in this temperature range, suggesting a continuous vapor-liquid critical curve between the pure components critical points. A total of 160 points were acquired: 149 bubble points and 11 dew points.
These new experimental data were satisfactory modeled by using the Peng-Robinson equation of state with a temperature-dependent interaction parameter, kij(T), fitted against the data.
{"title":"The CO2 + 2,3,4-trimethylpentane and CO2 + 2,2,4,6,6-pentamethylheptane binary systems: high–pressure phase equilibria measurements","authors":"Stéphane Vitu , Vincent Caqueret , Jean-Luc Daridon , Jean-Patrick Bazile","doi":"10.1016/j.fluid.2025.114648","DOIUrl":"10.1016/j.fluid.2025.114648","url":null,"abstract":"<div><div>Two binary systems containing CO<sub>2</sub> and a branched alkane were newly studied in this work: CO<sub>2</sub> + 2,3,4-trimethylpentane and CO<sub>2</sub> + 2,2,4,6,6-pentamethylheptane. Their fluid phase behavior was investigated using a high-pressure variable-volume cell and a synthetic method. Saturation pressures were visually measured from (293.15 et 363.15) K and for 10 different CO<sub>2</sub> mole fractions for both mixtures.</div><div>Whatever the considered system, no liquid-liquid immiscibility was observed in this temperature range, suggesting a continuous vapor-liquid critical curve between the pure components critical points. A total of 160 points were acquired: 149 bubble points and 11 dew points.</div><div>These new experimental data were satisfactory modeled by using the Peng-Robinson equation of state with a temperature-dependent interaction parameter, <em>k<sub>ij</sub></em>(<em>T</em>), fitted against the data.</div></div>","PeriodicalId":12170,"journal":{"name":"Fluid Phase Equilibria","volume":"603 ","pages":"Article 114648"},"PeriodicalIF":2.7,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145682419","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-01DOI: 10.1016/j.fluid.2025.114649
Haziq Ridwan Bin Asmuni , Romain Privat , Saifuddin Ahmed , Marc Bonnissel , Jean-Noël Jaubert
An extended version of the I-PC-SAFT equation of state (EoS) which incorporates the association term is presented, and a parameterisation protocol is proposed for water and three well-known homologous series of self-associating compounds: linear alcohols (from C1 to C18, except C14), linear carboxylic acids (from C2 to C17) and linear amines (from C1 to C12). The protocol follows the same constraints used in the original non-associative I-PC-SAFT (NonA-I-PC-SAFT), i.e., an exact reproduction of the critical temperature and pressure, acentric factor as well as the saturated liquid density at a reduced temperature of 0.8.
Seven (1A, 2A, 2B, 3A, 3B, 4A and 4B) and eight classical association schemes (1A, 2A, 2B, 3A, 3B, 4A, 4B and 4C) were examined for the three homologous series and water respectively, in addition to the reference variant without the association term (NonA-I-PC-SAFT). For each component, the optimal reduced association parameters were determined by minimising an objective function that prioritises the vapour pressure, followed by liquid density and finally thermal properties.
Compared to the NonA-I-PC-SAFT EoS, the association term substantially improved predictions for strongly associating fluids. For alcohols, the mean absolute percentage error (MAPE) in vapour pressures decreases from ∼15 % to ∼2 %, heat capacity and heat of vaporisation errors decrease from ∼16 % and ∼10 % to <5 %. Similar improvements are observed for carboxylic acids: from ∼8 % to less than ∼1 % for the vapour pressures, from ∼14 % to 9 % for the heat of vaporisation, while the errors of the heat capacities were barely reduced from ∼8 % to ∼7 %. For water, the association term reduces MAPE for vapour pressures from ∼5 % to ∼1 %, for heat of vaporisation from ∼4 % to ∼2 % and for heat capacity from ∼24 % to ∼4 %. In contrast, minimal improvements are observed for the amines that are weakly associating. The liquid densities errors remained close to ∼5 % for all systems.
In the case of alcohols, amines and water, all association schemes gave comparable results. The optimal association scheme was found to be 2B for both amines and alcohols, 1A for the acids and 4B for water. These results demonstrate that the association term is essential for strongly associating alcohols, carboxylic acids and water, whilst the NonA-I-PC-SAFT EoS offers a simpler alternative for amines without significantly compromising the accuracy of the predictions.
{"title":"How to parameterise the association term in SAFT models? Insights from the I-PC-SAFT Equation of State","authors":"Haziq Ridwan Bin Asmuni , Romain Privat , Saifuddin Ahmed , Marc Bonnissel , Jean-Noël Jaubert","doi":"10.1016/j.fluid.2025.114649","DOIUrl":"10.1016/j.fluid.2025.114649","url":null,"abstract":"<div><div>An extended version of the <em>I</em>-PC-SAFT equation of state (EoS) which incorporates the association term is presented, and a parameterisation protocol is proposed for water and three well-known homologous series of self-associating compounds: linear alcohols (from C<sub>1</sub> to C<sub>18</sub>, except C<sub>14</sub>), linear carboxylic acids (from C<sub>2</sub> to C<sub>17</sub>) and linear amines (from C<sub>1</sub> to C<sub>12</sub>). The protocol follows the same constraints used in the original non-associative <em>I</em>-PC-SAFT (NonA-<em>I</em>-PC-SAFT), i.e., an exact reproduction of the critical temperature and pressure, acentric factor as well as the saturated liquid density at a reduced temperature of 0.8.</div><div>Seven (1A, 2A, 2B, 3A, 3B, 4A and 4B) and eight classical association schemes (1A, 2A, 2B, 3A, 3B, 4A, 4B and 4C) were examined for the three homologous series and water respectively, in addition to the reference variant without the association term (NonA-<em>I</em>-PC-SAFT). For each component, the optimal reduced association parameters <span><math><mrow><mo>(</mo><mrow><msup><mrow><mi>κ</mi></mrow><mrow><mi>A</mi><mi>B</mi></mrow></msup><mo>,</mo><mrow><msup><mrow><mrow><mi>ε</mi></mrow></mrow><mrow><mi>A</mi><mi>B</mi></mrow></msup><mo>/</mo><mrow><mi>ε</mi></mrow></mrow></mrow><mo>)</mo></mrow></math></span> were determined by minimising an objective function that prioritises the vapour pressure, followed by liquid density and finally thermal properties.</div><div>Compared to the NonA-<em>I</em>-PC-SAFT EoS, the association term substantially improved predictions for strongly associating fluids. For alcohols, the mean absolute percentage error (MAPE) in vapour pressures decreases from ∼15 % to ∼2 %, heat capacity and heat of vaporisation errors decrease from ∼16 % and ∼10 % to <5 %. Similar improvements are observed for carboxylic acids: from ∼8 % to less than ∼1 % for the vapour pressures, from ∼14 % to 9 % for the heat of vaporisation, while the errors of the heat capacities were barely reduced from ∼8 % to ∼7 %. For water, the association term reduces MAPE for vapour pressures from ∼5 % to ∼1 %, for heat of vaporisation from ∼4 % to ∼2 % and for heat capacity from ∼24 % to ∼4 %. In contrast, minimal improvements are observed for the amines that are weakly associating. The liquid densities errors remained close to ∼5 % for all systems.</div><div>In the case of alcohols, amines and water, all association schemes gave comparable results. The optimal association scheme was found to be 2B for both amines and alcohols, 1A for the acids and 4B for water. These results demonstrate that the association term is essential for strongly associating alcohols, carboxylic acids and water, whilst the NonA-<em>I</em>-PC-SAFT EoS offers a simpler alternative for amines without significantly compromising the accuracy of the predictions.</div></div>","PeriodicalId":12170,"journal":{"name":"Fluid Phase Equilibria","volume":"604 ","pages":"Article 114649"},"PeriodicalIF":2.7,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145838509","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-30DOI: 10.1016/j.fluid.2025.114647
Yizhao Wang , Li Zhu , Hongwei Yang
This study proposes an optimization strategy for the modified embedded atom method (MEAM) potentials of Pb-Au and Pb-Sn alloys and develops an approach for calculating binary alloy vapor-liquid equilibrium (VLE). A sensitivity of the potential parameters to the properties of the metal reveals that the potential parameters A and β(0) of the unary MEAM have a significant impact on the melting point (Tm). The binary MEAM potential parameters ρA:ρB, Cmin(A-B-B) and Cmin(A-B-A) are dominant of mixing enthalpy (ΔHm) for binary alloy. Based on this, the MEAM potentials of the Pb-Au and Pb-Sn systems are developed. These potentials accurately predict the Tm of pure Pb, Au, and Sn, though slight deviations are observed in the predictions of isobaric heat capacities (Cp). Regarding the Pb-Au and Pb-Sn binary systems, the calculated mixing enthalpies for both systems exhibit the mean absolute percentage errors (MAPE) within 15 %, indicating acceptable agreement with experimental data. To bridge atomic-scale simulation and macroscopic phase equilibrium, this study combines MEAM with the Wilson equation. The component activities in both the Pb-Au and Pb-Sn systems and the VLE under vacuum conditions are predicted based on this method. The VLE diagram reveals that reduced system pressure significantly lowers the temperature required in vacuum distillation. Furthermore, the method quantifies the optimal conditions for achieving high-purity metal extraction by vacuum distillation.
{"title":"MEAM potential optimization and vapor-liquid equilibrium modeling for Pb-Au and Pb-Sn alloys","authors":"Yizhao Wang , Li Zhu , Hongwei Yang","doi":"10.1016/j.fluid.2025.114647","DOIUrl":"10.1016/j.fluid.2025.114647","url":null,"abstract":"<div><div>This study proposes an optimization strategy for the modified embedded atom method (MEAM) potentials of Pb-Au and Pb-Sn alloys and develops an approach for calculating binary alloy vapor-liquid equilibrium (VLE). A sensitivity of the potential parameters to the properties of the metal reveals that the potential parameters <em>A</em> and <em>β</em><sup>(0)</sup> of the unary MEAM have a significant impact on the melting point (<em>T</em><sub>m</sub>). The binary MEAM potential parameters <em>ρ</em><sup>A</sup>:<em>ρ</em><sup>B</sup>, <em>C<sub>min</sub></em>(A-B-B) and <em>C<sub>min</sub></em>(A-B-A) are dominant of mixing enthalpy (Δ<em>H</em><sub>m</sub>) for binary alloy. Based on this, the MEAM potentials of the Pb-Au and Pb-Sn systems are developed. These potentials accurately predict the <em>T</em><sub>m</sub> of pure Pb, Au, and Sn, though slight deviations are observed in the predictions of isobaric heat capacities (<em>C<sub>p</sub></em>). Regarding the Pb-Au and Pb-Sn binary systems, the calculated mixing enthalpies for both systems exhibit the mean absolute percentage errors (MAPE) within 15 %, indicating acceptable agreement with experimental data. To bridge atomic-scale simulation and macroscopic phase equilibrium, this study combines MEAM with the Wilson equation. The component activities in both the Pb-Au and Pb-Sn systems and the VLE under vacuum conditions are predicted based on this method. The VLE diagram reveals that reduced system pressure significantly lowers the temperature required in vacuum distillation. Furthermore, the method quantifies the optimal conditions for achieving high-purity metal extraction by vacuum distillation.</div></div>","PeriodicalId":12170,"journal":{"name":"Fluid Phase Equilibria","volume":"603 ","pages":"Article 114647"},"PeriodicalIF":2.7,"publicationDate":"2025-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145682422","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-28DOI: 10.1016/j.fluid.2025.114643
Yanyan Hou, Zhentao Zhang, Ziyue Zhang, Yi Yu, Li Xu
The solubility of glutaric anhydride (GA) in ten single organic solvents containing methyl acetate, ethyl acetate, n-propyl acetate, butyl acetate, acetone, acetonitrile, dichloromethane, 1,2-dichloroethane, DMF, DMAC, was determined by static method. In the experimental temperature range (278.15 ∼ 313.15 K), the solubility of the measured GA in each solvent increased with the increase of temperature. Among the solvents selected, GA had the highest solubility in DMAC and the lowest solubility in butyl acetate. The solubility of GA in DMAC was about 2 ∼ 3 times higher than in butyl acetate. Solvent effects in different solvents were explained by the Hansen solubility parameter (HSP) and the physicochemical properties of the solvents (polarity, hydrogen bond donor-acceptor tendency, and cohesion energy density). The analysis of solvation effect showed that the changes of GA solubility were the results of the combined effect of multiple factors, among which the effect of solvent polarity was more significant. Based on density functional theory (DFT), the electrostatic potential energy surface and solute-solvent interaction of GA were analyzed at the molecular level. In addition, the solubility data were fitted using the van't Hoff, modified Apelblat, Yaws, λh and Wilson models. The fitting results showed that the Yaws model had the best correlation with the smallest ARD (3.74 %) and RMSD (0.82 %). The analysis results of the apparent thermodynamic properties indicated that GA dissolution was a spontaneous, endothermic, entropy-driven process.
{"title":"Solubility, solvent effects, correlation, and thermodynamic properties of glutaric anhydride in ten individual organic solvents from 278.15 to 313.15 K","authors":"Yanyan Hou, Zhentao Zhang, Ziyue Zhang, Yi Yu, Li Xu","doi":"10.1016/j.fluid.2025.114643","DOIUrl":"10.1016/j.fluid.2025.114643","url":null,"abstract":"<div><div>The solubility of glutaric anhydride (GA) in ten single organic solvents containing methyl acetate, ethyl acetate, <em>n</em>-propyl acetate, butyl acetate, acetone, acetonitrile, dichloromethane, 1,2-dichloroethane, DMF, DMAC, was determined by static method. In the experimental temperature range (278.15 ∼ 313.15 K), the solubility of the measured GA in each solvent increased with the increase of temperature. Among the solvents selected, GA had the highest solubility in DMAC and the lowest solubility in butyl acetate. The solubility of GA in DMAC was about 2 ∼ 3 times higher than in butyl acetate. Solvent effects in different solvents were explained by the Hansen solubility parameter (HSP) and the physicochemical properties of the solvents (polarity, hydrogen bond donor-acceptor tendency, and cohesion energy density). The analysis of solvation effect showed that the changes of GA solubility were the results of the combined effect of multiple factors, among which the effect of solvent polarity was more significant. Based on density functional theory (DFT), the electrostatic potential energy surface and solute-solvent interaction of GA were analyzed at the molecular level. In addition, the solubility data were fitted using the van't Hoff, modified Apelblat, Yaws, <em>λh</em> and Wilson models. The fitting results showed that the Yaws model had the best correlation with the smallest <em>ARD</em> (3.74 %) and <em>RMSD</em> (0.82 %). The analysis results of the apparent thermodynamic properties indicated that GA dissolution was a spontaneous, endothermic, entropy-driven process.</div></div>","PeriodicalId":12170,"journal":{"name":"Fluid Phase Equilibria","volume":"603 ","pages":"Article 114643"},"PeriodicalIF":2.7,"publicationDate":"2025-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145682420","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-28DOI: 10.1016/j.fluid.2025.114644
Neha Sawant , Noor Alomari , Kendall Nore , Niah Thurman , James Springstead , Santiago Aparicio , Mert Atilhan
This study investigates ten hydrophobic deep eutectic solvent (HDES) systems for removing methyl paraben (MP) and bisphenol A (Bis-A) from contaminated water. The HDES were synthesized using various combinations of terpenes, fatty acids, and quaternary ammonium salts, including Cineole-Oleic Acid, Cineole-Decanoic Acid, Carvone-Menthol, Cineole-Linoleic Acid, Geraniol-Linoleic Acid, Geraniol-Oleic Acid, Trihexyl (tetradecyl)phosphonium Chloride-Linoleic Acid, Geraniol-Decanoic Acid, Tetra-n-octyl ammonium Bromide-Linoleic Acid, and Tetrabutylammonium Chloride-Linoleic Acid. Extraction efficiency was determined by analyzing the change in contaminant concentration after treatment with HDES, measured using UV-Vis spectroscopy. Classical molecular dynamics (MD) simulations were also conducted to elucidate the molecular interactions between the HDES and contaminants. Radial and spatial distribution functions, and hydrogen bonding distributions were analyzed to understand the extraction mechanisms.
Experimental and computational studies demonstrated the potential of HDES as an efficient for personal care product treatment in water. P66614Cl-LnA and N8888Br-LnA were the most effective, removing up to 95.44% of Bis-A and 99.30% of MP. Cin-LnA, Cin-OleA, and Ger-DeA also showed high removal efficiencies, exceeding 90% for Bis-A and 92% for MP in most cases. In contrast, CAR-MEN and N4444Cl-LnA showed lower efficiencies. These findings highlight the potential of HDES, particularly P66614Cl-LnA and N8888Br-LnA, for water purification applications.
{"title":"Efficient removal of bisphenol A and methyl paraben from water using hydrophobic deep eutectic solvents: Experimental and molecular dynamics insights","authors":"Neha Sawant , Noor Alomari , Kendall Nore , Niah Thurman , James Springstead , Santiago Aparicio , Mert Atilhan","doi":"10.1016/j.fluid.2025.114644","DOIUrl":"10.1016/j.fluid.2025.114644","url":null,"abstract":"<div><div>This study investigates ten hydrophobic deep eutectic solvent (HDES) systems for removing methyl paraben (MP) and bisphenol A (Bis-A) from contaminated water. The HDES were synthesized using various combinations of terpenes, fatty acids, and quaternary ammonium salts, including Cineole-Oleic Acid, Cineole-Decanoic Acid, Carvone-Menthol, Cineole-Linoleic Acid, Geraniol-Linoleic Acid, Geraniol-Oleic Acid, Trihexyl (tetradecyl)phosphonium Chloride-Linoleic Acid, Geraniol-Decanoic Acid, Tetra-n-octyl ammonium Bromide-Linoleic Acid, and Tetrabutylammonium Chloride-Linoleic Acid. Extraction efficiency was determined by analyzing the change in contaminant concentration after treatment with HDES, measured using UV-Vis spectroscopy. Classical molecular dynamics (MD) simulations were also conducted to elucidate the molecular interactions between the HDES and contaminants. Radial and spatial distribution functions, and hydrogen bonding distributions were analyzed to understand the extraction mechanisms.</div><div>Experimental and computational studies demonstrated the potential of HDES as an efficient for personal care product treatment in water. P66614Cl-LnA and N8888Br-LnA were the most effective, removing up to 95.44% of Bis-A and 99.30% of MP. Cin-LnA, Cin-OleA, and Ger-DeA also showed high removal efficiencies, exceeding 90% for Bis-A and 92% for MP in most cases. In contrast, CAR-MEN and N4444Cl-LnA showed lower efficiencies. These findings highlight the potential of HDES, particularly P66614Cl-LnA and N8888Br-LnA, for water purification applications.</div></div>","PeriodicalId":12170,"journal":{"name":"Fluid Phase Equilibria","volume":"603 ","pages":"Article 114644"},"PeriodicalIF":2.7,"publicationDate":"2025-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145682423","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The phase behavior and properties of confined fluids play a critical role in subsurface energy and environmental operations. Predicting these behaviors in porous media typically relies on Molecular Dynamics (MD) simulations, which, while accurate, are prohibitively expensive for large-scale applications. Deep learning (DL) has recently emerged as a promising alternative for developing surrogate models of such processes. However, conventional DL architectures require large volumes of training data—an impractical requirement given the high cost of generating MD datasets. To address this challenge, transfer learning can be employed: models are first trained on related, lower-cost tasks and subsequently adapted to the target task with limited data. This strategy has been highly effective in domains such as natural language processing and computer vision, but its application to confined fluid modeling remains underexplored.
In this work, we present NanoSG, a science-guided deep learning framework for emulating MD simulations of fluid mixtures in confinement. NanoSG integrates domain knowledge with pre-trained representations to enhance learning efficiency and physical consistency. Through extensive experimentation, we show that NanoSG achieves robust generalization, with a minimum performance improvement of 16.26% over baseline models, while maintaining consistency with established scientific principles despite being trained on limited MD data. Our results highlight the potential of science-guided transfer learning to accelerate predictive modeling of confined fluids under data-scarce conditions, opening new avenues for scalable simulation in energy and subsurface applications.
{"title":"Science-guided transfer learning for molecular dynamics of confined fluids in shale nanopores","authors":"Nikhil Muralidhar , Mohamed Mehana , Naren Ramakrishnan , Anuj Karpatne , Nicholas Lubbers","doi":"10.1016/j.fluid.2025.114646","DOIUrl":"10.1016/j.fluid.2025.114646","url":null,"abstract":"<div><div>The phase behavior and properties of confined fluids play a critical role in subsurface energy and environmental operations. Predicting these behaviors in porous media typically relies on Molecular Dynamics (MD) simulations, which, while accurate, are prohibitively expensive for large-scale applications. Deep learning (DL) has recently emerged as a promising alternative for developing surrogate models of such processes. However, conventional DL architectures require large volumes of training data—an impractical requirement given the high cost of generating MD datasets. To address this challenge, transfer learning can be employed: models are first trained on related, lower-cost tasks and subsequently adapted to the target task with limited data. This strategy has been highly effective in domains such as natural language processing and computer vision, but its application to confined fluid modeling remains underexplored.</div><div>In this work, we present NanoSG, a science-guided deep learning framework for emulating MD simulations of fluid mixtures in confinement. NanoSG integrates domain knowledge with pre-trained representations to enhance learning efficiency and physical consistency. Through extensive experimentation, we show that NanoSG achieves robust generalization, with a minimum performance improvement of 16.26% over baseline models, while maintaining consistency with established scientific principles despite being trained on limited MD data. Our results highlight the potential of science-guided transfer learning to accelerate predictive modeling of confined fluids under data-scarce conditions, opening new avenues for scalable simulation in energy and subsurface applications.</div></div>","PeriodicalId":12170,"journal":{"name":"Fluid Phase Equilibria","volume":"603 ","pages":"Article 114646"},"PeriodicalIF":2.7,"publicationDate":"2025-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145682425","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-27DOI: 10.1016/j.fluid.2025.114645
Yohan Lee , Sivakumar Subramanian , Douglas Estanga , Amadeu K. Sum
Gas hydrates are an ever-present concern in oil and gas production, as their undesirable formation results in significant operational loss in addition to creating potentially hazardous conditions. Although predictions of gas hydrate phase equilibria to prevent flowline blockage are mostly reliable, the data for hydrate phase equilibria in the literature are mostly limited to mild production conditions in flowlines. However, uncertainties in the predictions are significant when considering more severe (high pressures) and unusual systems (high salinity) with less common salts, such as Zinc salts. This study assesses the reliability and accuracy of hydrate dissociation conditions for brine blends to be used in completion fluids containing ZnBr2, with measurements of hydrate dissociation and comparison among common prediction tools. Hydrate phase equilibria for four brine blends containing NaBr, CaBr2, and ZnBr2 with a synthetic gas mixture were determined at pressures up to 2000 bar and compared with different prediction tools. The predictions with commercial software showed unreliable results for the brine blends, whereas the HLS correlation provided reliable and accurate predictions of the hydrate phase equilibrium boundary for the ZnBr2 containing brine blends with consideration of water activity in the brine blends. These results establish a more robust basis for predicting hydrate risks in completion fluids under extreme field conditions.
{"title":"Gas hydrate phase equilibria for aqueous solutions of NaBr, CaBr2, and ZnBr2 at high salinity and pressures up to 2000 bar","authors":"Yohan Lee , Sivakumar Subramanian , Douglas Estanga , Amadeu K. Sum","doi":"10.1016/j.fluid.2025.114645","DOIUrl":"10.1016/j.fluid.2025.114645","url":null,"abstract":"<div><div>Gas hydrates are an ever-present concern in oil and gas production, as their undesirable formation results in significant operational loss in addition to creating potentially hazardous conditions. Although predictions of gas hydrate phase equilibria to prevent flowline blockage are mostly reliable, the data for hydrate phase equilibria in the literature are mostly limited to mild production conditions in flowlines. However, uncertainties in the predictions are significant when considering more severe (high pressures) and unusual systems (high salinity) with less common salts, such as Zinc salts. This study assesses the reliability and accuracy of hydrate dissociation conditions for brine blends to be used in completion fluids containing ZnBr<sub>2,</sub> with measurements of hydrate dissociation and comparison among common prediction tools. Hydrate phase equilibria for four brine blends containing NaBr, CaBr<sub>2</sub>, and ZnBr<sub>2</sub> with a synthetic gas mixture were determined at pressures up to 2000 bar and compared with different prediction tools. The predictions with commercial software showed unreliable results for the brine blends, whereas the HLS correlation provided reliable and accurate predictions of the hydrate phase equilibrium boundary for the ZnBr<sub>2</sub> containing brine blends with consideration of water activity in the brine blends. These results establish a more robust basis for predicting hydrate risks in completion fluids under extreme field conditions.</div></div>","PeriodicalId":12170,"journal":{"name":"Fluid Phase Equilibria","volume":"603 ","pages":"Article 114645"},"PeriodicalIF":2.7,"publicationDate":"2025-11-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145682424","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-25DOI: 10.1016/j.fluid.2025.114642
Sugata P. Tan, Jeffrey S. Kargel
This is an erratum to the paper by Tan & Kargel published in Fluid Phase Equilibria 458 (2018) 153–169. Two typos originating from misprints in the manuscript that went undetected through proofreading were found in equations that appeared in Appendix A of the paper.
{"title":"Erratum to “Multiphase-equilibria analysis: Application in modeling the atmospheric and lacustrine chemical systems of Saturn's moon Titan” [Fluid Phase Equil. 458 (2018) 153-169]","authors":"Sugata P. Tan, Jeffrey S. Kargel","doi":"10.1016/j.fluid.2025.114642","DOIUrl":"10.1016/j.fluid.2025.114642","url":null,"abstract":"<div><div>This is an erratum to the paper by Tan & Kargel published in Fluid Phase Equilibria 458 (2018) 153–169. Two typos originating from misprints in the manuscript that went undetected through proofreading were found in equations that appeared in Appendix A of the paper.</div></div>","PeriodicalId":12170,"journal":{"name":"Fluid Phase Equilibria","volume":"603 ","pages":"Article 114642"},"PeriodicalIF":2.7,"publicationDate":"2025-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145616576","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号