Solubility In Mixed Solvents Research Articles

Prediction of organic sulfur solubility in mixed solvent using feature-based transfer learning and a hybrid Henry's law constant calculation method

Prediction of organic sulfur solubility in mixed solvent using feature-based transfer learning and a hybrid Henry's law constant calculation method

Using Molecular Conformers in COSMO-RS to Predict Drug Solubility in Mixed Solvents

Using Molecular Conformers in COSMO-RS to Predict Drug Solubility in Mixed Solvents

Solubility of Gallic Acid in Single and Mixed Solvents

Gallic acid, known for its biological activity contributing to human health, including antioxidant, anti-inflammatory, anticancer, and antimutagenic properties, was the focus of this study. The solubility of gallic acid was experimentally measured in pure and mixed solvents of water, ethanol, and acetic acid and predicted using the COSMO-SAC model and the Hansen solubility parameter. The Hansen solubility parameter method predicted a higher solubility of gallic acid in pure water than in pure ethanol, and in a mixed solvent, it predicted the maximum solubility at 80% water content, showing different results from the experimental data trends. However, using the molar volume obtained from COSMO calculations resulted in a tendency that matched the experimental results. The results revealed higher solubility in ethanol compared to water, with the solubility in mixed solvent falling within the range between them. Using the same method, the Hansen solubility parameter obtained was applied to acetic acid/water and acetic acid/ethanol mixtures, and similar trends were observed compared to experimental data. In particular, gallic acid in the acetic acid/water mixture solvent exhibited maximum solubility, and this phenomenon was well-predicted. As the temperature increased, solubility in both pure and mixed solvents also increased. While the COSMO-SAC model effectively captured this trend, the predicted solubility values were slightly lower than the experimental data. The solubility trends depending on solvent types were confirmed by comparing the σ-profiles of each compound. The σ-profile of gallic acid closely resembled that of ethanol, and this result led to higher solubility than water and acetic acid. The maximum solubility in ethanol/water and acetic acid/water mixed solvents could be anticipated when two solvents with significant differences in their σ-profiles are mixed in an appropriate ratio.

Local composition-regular solution theory for analysis of pharmaceutical solubility in mixed-solvents

Local composition regular solution theory (LC-RST) model was developed by introducing non-equal i–j and j–i parameters into regular solution theory (RST) to account for specific molecular interactions. The LC-RST model was applied to the correlation of solubilities of active pharmaceutical ingredients (API) in hydrogen bond donor (HBD) – hydrogen bond acceptor (HBA) mixed-solvent systems. In the assessment of 32 API-mixed-solvent systems that contained 11 different APIs and 1947 data, it was determined that the LC-RST model had a logarithmic average relative deviation (ARDln) in solubility of 0.028, the Wilson model had an ARDln value of 0.24 and the purely predictive RST model had an ARDln value of 1.29. Partial molar excess enthalpies and entropies of the API in the mixed-solvents determined from the LC-RST model allowed estimation of API solubility regions where specific interactions occur between API and HBD-HBA complex molecules. In the correlation scheme, component numbers were designated as API molecule (1) – HBD molecule (2) - HBA molecule (3) and systems were grouped according to HBD molecule interactions with the API (pro-solvent, anti-solvent, synergistic with HBA molecule). Phenomenological parameter sensitivity analysis of the LC-RST model showed that dominant specific interactions tended to be 1–2 or 2–1 molecules for pro-solvent groupings, 2–3 or 2–3 molecules for anti-solvent groupings and all i–j and j-–i molecular pairs for synergistic HBD-HBA molecule groupings. Based on parameters determined from a given API-mixed-solvent system, the LC-RST model allows qualitative prediction of solubilities for other systems. The LC-RST model is simple to apply and it retains inherent predictive characteristics of regular solution theory.

Predicting sulfanilamide solubility in mixed solvents: A comparative analysis of computational models

The Jouyban-Acree model is the most common in the solubility prediction area. This work focused on providing the various mathematical models derived from the Jouyban-Acree equation for the solubility prediction of sulfanilamide in nine binary solvent mixtures in the 278.15 to 318.15 K temperature range. 24 various computational models are reported in this work, which are derived from the Jouyban-Acree equation and include the terms Abraham solute, solvent, Hansen solubility parameters, QSPR, and Yalkowsky models with the Van't Hoff and Apelblat parameters. This study reveals the reliability of the predicted solubility values from these 24 different computational models compared with one another's equations. By calculating the %ARD, it was validated that any variation to the Jouyban-Acree equation's parameters, including those that affect the solute, solvent, and Hansen solubility properties, in addition to the Van't Hoff, Apelblat parameters, and Yalkowsky model, made any difference in the solubility values as compared to experimentally obtained solubility data that had previously been published. Using mono-solvent solubility data to calculate binary solubility with the Yalkowsky model can be a useful and efficient approach, and it could be especially important in situations where experimental binary solubility data is limited or unavailable. However, the Yalkowsky model showed a low overall percentage relative deviation of 60.930 % compared to other computed models. In the binary solvent systems, the Yalkowsky model had the least %ARD in methanol (w) + ethanol (1-w) (3.77 %), followed by acetone (w) + methanol (1-w) (9.34 %) and acetone (w) + ethanol (1-w) (23.57 %). The Modified Yalkowsky-Jouyban-Acree Model, in combination with the Abraham Solute Parameter Model, demonstrated the lowest %ARD of 3.81 % in the methanol (w) + ethanol (1-w) binary mixture. Similarly, in most combined versions of the Jouyban-Acree models incorporating various Apelblat and Van't Hoff parameters, the lowest %ARD observed was 10.16 % in the methanol (w) + ethanol (1-w) binary mixture. The Yalkowsky model also showed a lower average percentage deviation than other models in acetone, methanol, and ethanol as the first solvent, with values of 50.535 %, 58.943 %, and 84.7 %, respectively. The Yalkowsky model provided a good fit to the experimental data, with regression coefficients of 0.87895, 0.96911, and 0.98164 for acetone, ethanol, and methanol as the first solvents in binary solvent systems.

Salt solubility modeling in mixed solvents with a modified Pitzer approach: Application in water + MEG + electrolyte systems

AbstractMonoethylene glycol (MEG) has been applied as an inhibitor of hydrate formation in natural gas exploitation. The recovery unit of MEG presents salt scale issues that promote operational problems. A literature survey and analysis of salt solubility data in water–MEG mixtures have been performed, revealing that experimental studies are available for eight electrolyte species (NaCl, NaHCO3, KCl, Na2SO4, KI, KBr, K2SO4, and CaSO4), in total 616 data points. This work aimed at the development of a calculation methodology using Pitzer–Lorimer approach for salt solubility in mixed solvents. Python language was properly used to design the computational tool, which is freely available. The results demonstrated the feasibility of the salt solubility description in mixed solvents, and relative deviations less than 5% were obtained for the eight studied salts. Furthermore, the developed tool can be easily applied for parameter estimation and salt solubility prediction of other electrolyte systems.

Retraction of "Challenges in Predicting Aqueous Solubility of Organic Molecules Using the COSMO-RS Model".

The conductor-like screening model for real solvents (COSMO-RS) is an emerging tool for predicting the thermodynamic properties of small molecules, such as activity coefficients, vapor pressure, and solubility in mixed solvents. We have evaluated the validity and performance of the COSMO-RS model in predicting aqueous solubility using a high-fidelity experimental data set containing 1852 small organic molecules. Our analysis indicates that overreliance on distributions of the screening charge densities on the outer surface of the molecule, known as the σ-profile, leads to frequent overestimation or underestimation, rendering the predicted aqueous solubility of organic molecules unreliable. Despite our efforts to refine the inherent parameters of the COSMO-RS model using a comprehensive experimental data set, the prediction accuracy remained limited (R2 ∼ 0.5). We analyzed the correlation between prediction accuracy and common molecular descriptors, such as dipole moment, molar mass, symmetry of charges, and types of charge centers. A notable challenge of COSMO-RS prediction arises from the electronegative atoms, which likely generate uneven charge distributions in the σ-profile, leading to overtly strong hydrogen bond corrections and thereby causing overestimation of aqueous solubilities.

Application of Polarisable Continuum Modelling to assess Minoxidil solubility in mixed solvents

ABSTRACT Experimental solubility of minoxidil was obtained in binary aqueous mixtures of ethanol, methanol and acetonitrile at 298.2 K. The effects of pH and micellisation were investigated on the solubility of the drug. Experimental solubility data was correlated by three co-solvency models, i.e. Yalkowsky, CNIBS/R-K, and the modified Wilson model. Density functional theory and hybrid functional B3LYP method was used with Polarisable Continuum Model (PCM) to correlate the experimental solubility data in various mass fractions of the binary solvent mixtures. We report the results of solute–solvent interaction energy as free energy of solvation (∆G sol ) of minoxidil in various mass fractions of ethanol + water mixtures. Total electrostatic energy, dipole moment, total Gibbs free energy and dielectric constant (ε) of minoxidil were also investigated.

Comments on “Stearic acid solubility in mixed solvents of (water + ethanol) and (ethanol + ethyl acetate): Experimental data and comparison among different thermodynamic models”

A polemic is given regarding the author's calculated values pertaining to the “goodness-of-fit” used to indicate the better of the predictive expressions considered to describe the observed solubility data of stearic acid. In the case of the van't Hoff model the author averaged the individual relative deviations for the solubility of stearic acid in both the ethanol and ethyl acetate mono-solvents. Positive and negative relative individual deviations would thus cancel. For the modified Apelblat model the author averaged the absolute values of the individual relative deviations for the solubility of stearic acid in the two organic mono-solvents.

Correlation and prediction of small to large sized pharmaceuticals solubility, and crystallization in binary and ternary mixed solvents using the UNIQUAC-SAC model

The recently reported UNIversal QUAsiChemical Segment Activity coefficient (UNIQUAC-SAC) model [developed by Haghtalab and Yousefi Seyf Ind. Eng. Chem. Res. 2015, 54, 8611] provides a practical thermodynamic framework to be used in VLE, LLE, and SLE calculations. The UNIQUAC-SAC model has the advantage of being independent of area (q) and volume (r) structural parameters used in the combinatorial part. While the UNIFAC or UNIFAC-DMD could not apply to the 47% (44 of 94) of the studied molecules because of the undefined groups. Here, the numbers of solvents with identified segment numbers were extended from 82 to 130 with a slight refinement to the previous values. The model parameters obtained via a large consistent set of VLE (isobaric and isothermal), and LLE experimental data. Also, the model was tested with pharmaceutical solubility experimental data in pure (94 solutes with 6210 solubility data), binary, and ternary solvents. The model provides a robust correlation of pharmaceuticals solubility in a mono-solvent and robust prediction of pharmaceuticals solubility in mixed solvents. The UNIQUAC-SAC model was successfully evaluated in solvent screening in cooling crystallization of ibuprofen and valsartan in both pure and binary solvents mixture (combined cooling and anti-solvent crystallization). The model with 130 solvents as a robust database provides a useful and practical thermodynamic tool in the conceptual design of pharmaceutical processes. A MATLAB based graphical user interface (GUI), finally, was developed to be used in the conceptual segment numbers regression procedure.

Further computation and some comments on “Stearic acid solubility in mixed solvents of (water + ethanol) and (ethanol + ethyl acetate): Experimental data and comparison among different thermodynamic models”

The experimental solubility data of stearic acid in the neat solvents, ethanol + water and ethyl acetate + ethanol binary solvent systems at the temperature range from 293.15 to 323.15 K have carefully been reanalyzed and some recommendations were proposed. Further computations are performed on the stearic acid solubility data by using some cosolvency models including CNIBS/R-K, modified Wilson, NRTL, Wilson, UNIQUAC, Yalkowsky, Jouyban-Acree, and Jouyban-Acree-van't Hoff models.

Stearic acid solubility in mixed solvents of (water + ethanol) and (ethanol + ethyl acetate): Experimental data and comparison among different thermodynamic models

In this study, the mole fraction solubility of stearic acid (SA) in ethanol, ethyl acetate and (water + ethanol) and (ethanol + ethyl acetate) solvents mixtures at atmospheric pressure from T = (298.15 to 330.15) K was estimated via a gravimetric method. The correlation of solubility data was performed with the van't Hoff, modified Apelblat and λh models in two binary solvent mixtures. While in mono-solvents, the van't Hoff, modified Apelblat, λh, Wilson, and NRTL models were employed. By comparing the models, we found that modified Apelblat equation and the van't Hoff equation fit the experimental data best. 5.94 × 10−2and 8.50 × 10−3 are the limit of the relative average deviation (RAD) as well as the root-mean-square deviation (RMSD) for the experimental and calculated solubility for the two binary solvent mixtures. Likewise, the dissolution enthalpy and dissolution entropy of stearic acid in the corresponding systems were predicted and the results demonstrated that the procedure for the dissolution in the mixed solvents is endothermic. The experimental solubility and mathematically calculated solubility can be usefully applied to the final crystallization and/or purification process of SA synthesis and to pharmaceutical formulation development of SA.

Carbon dioxide solubility in aqueous sulfolane solution

New experimental data are presented for the solubility of carbon dioxide in aqueous sulfolane (SFL) solution, an important solvent with applications in a variety of chemical industries. Solvent mixtures of (SFL + H2O) studied in this work contain the composition range xSFL = (0.0999–0.9313), temperatures from (303.15 to 353.15) K, with step 10 K, and total pressures up to 1.3 MPa. Solubility data and Henry’s law constants of CO2 solubility in pure SFL, pure H2O and mixed (SFL + H2O) are reported on mole fraction base. Results show that, CO2 solubility in mixed solvents increases with increasing pressure and gas-free mole fraction of SFL and also decrease with increasing temperature. A model based on the Gibbs excess energy of mixed solvent was used to correlate the obtained experimental data. The correlated values, with an average and maximum relative percent deviation of 1.78% and 7.55% from the obtained experimental data, show that the model is able to represent the experimental data with good correlative accuracy.

Solubility prediction in mixed solvents: A combined molecular simulation and experimental approach

We propose a simple method to predict solubility in mixed solvents, combining experimental data for solubility in pure solvents with non-ideal contributions obtained from molecular simulation. By evoking a well-established thermodynamic model for mixed solvent, we provide rationale and justification for this hybrid approach. We test the accuracy of the method for solubility prediction in two highly non-ideal mixtures, CO2 in ethanol + water, and CH4 in MDEA + water. Non-ideal behaviour in each system is characterised using molecular simulation across the full range of compositions and for temperatures covering the range 273.15 K–373.15 K. Comparison against experimental Henry's law constants for CO2 and CH4 gives mean absolute errors of 6.9% and 27%, respectively. We investigate the origin of non-ideal Henry's law behaviour in each of the mixed solvents by interrogating radial distribution functions from molecular simulation, finding increased association of CO2 and CH4 with the organic solvents at elevated temperatures.

Solubility of Dihydroxylammonium 5,5′-Bistetrazole-1,1′-diolate in (formic acid, water) and their binary solvents from 298.15 K to 333.15 K at 101.1 kPa

In this study, the solubility of Dihydroxylammonium 5,5′-Bistetrazole-1,1′-diolate (TKX-50) in formic acid, water and mixed solvents (formic acid + water) were measured with the temperature ranging from 298.15 K to 333.15 K by using a laser dynamic monitoring technique at 101.1 kPa. The experimental data showed that the solubility of TKX-50 increase with the increasing temperature in all solvents and decrease with the decreasing proportion of formic acid in mixed solvents (formic acid + water) at constant temperature. The mole fraction solubility of TKX-50 in pure formic acid and pure water were correlated with the modified Apelblat equation and the van’t Hoff equation, the mole fraction solubility in mixed solvents (formic acid + water) were correlated with the (CNIBS)/R-K equation, the Jouyban-Acree model, the Apelblat-Jouyban-Acree model, the van’t Hoff-Jouyban-Acree model, the Sun model and the Ma model. The result showed that calculated values correlated by these models are consistent with the experimental data. The solubility of TKX-50 in mixed solvents (formic acid + water) will provide essential support for the further research of crystallization and spheroidization of TKX-50.

Solubility and mixing thermodynamic properties of (2,4,6-trimethylbenzoyl) diphenylphosphine oxide in pure and binary solvents

The solubility of (2,4,6-trimethylbenzoyl) diphenylphosphine oxide (Lucirin TPO) in ten pure solvents and two binary solvent mixtures was measured from 273.15 K to 308.15 K by gravimetric method. The solubility of Lucirin TPO increased non-linearly with rising temperature in all the studied solvents. Furthermore, the solubility in mixed solvents presented a maximum-solubility effect. The modified Apelblat model, λh model, CNIBS/R-K model, NRTL model and Jouyban-Acree model were employed to correlate the experimental solubility data. The modified Apelblat model provided the best agreement in pure solvents, while the modified Apelblat model and CNIBS/R-K model gave better correlation results in binary solvent mixtures. Moreover, the thermodynamic properties of the mixing process, including the mixing Gibbs free energy, mixing enthalpy and mixing entropy were calculated from the solubility data using the NRTL model.

Remarkes on “solubility and solution thermodynamics of meldrum's adduct in ethyl acetate plus (acetonitrile, ethanol, hexane) binary solvent mixtures”

Combined Nearly Ideal Binary Solvent/Redlich-Kister (CNIBS/R-K) equations and van't Hoff-Jouyban-Acree model reported by Liu and coworkers [J. Mol. Liq. 220 (2016) 354–363] for calculating the solubility of meldrum's adduct in ethyl acetate plus (acetonitrile, ethanol, hexane) binary solvent mixtures at various temperatures. The authors of cited reference erroneously curve-fit the solubility data with multiplying the solubility value by ten times in the case of published equation coefficients for the CNIBS/R-K model. In addition, the authors of cited reference applied original van't Hoff-Jouyban-Acree model to predict meldrum's adduct solubility in mixed solvents at different temperatures, but that overlooked the fact that they did not measure the solubility data for meldrum's adduct in neat solvents. Furthermore, in determining the numerical values of the equation coefficients, no pure solvent solubility data were included in the regression analyses.

Solubility of pharmaceuticals: A comparison between SciPharma, a PC-SAFT-based approach, and NRTL-SAC

The solubility of seven pharmaceutical compounds (paracetamol, benzoic acid, 4-aminobenzoic acid, salicylic acid, ibuprofen, naproxen and temazepam) in pure and mixed solvents as a function of temperature is calculated with SciPharma, a semi-empirical approach based on PC-SAFT, and the NRTL-SAC model. To conduct a fair comparison between the approaches, the parameters of the compounds were regressed against the same solubility data, chosen to account for hydrophilic, polar and hydrophobic interactions. Only these solubility data were used by both models for predicting solubility in other pure and mixed solvents for which experimental data were available for comparison. A total of 386 pure solvent data points were used for the comparison comprising one or more temperatures per solvent. SciPharma is found to be more accurate than NRTL-SAC on the pure solvent data used especially in the description of the temperature dependence. This is due to the appropriate parameterization of the pharmaceuticals and the temperature-dependent description of the activity coefficient in PC-SAFT. The solubility in mixed solvents is predicted satisfactorily with SciPharma. NRTL-SAC tends to overestimate the solubility in aqueous solutions of alcohols or shows invariable solubility with composition in other cases.

Refinement of the theoretical solubility model and prediction of solute solubility in mixed solvent systems

We have refined the non-polar (van der Waalls) and hydrogen bonding interaction term of STMIPE-AC (statistical thermodynamic molar intermolecular potential energy activity coefficient) model, which reduces the average error of the model in correlating the molar internal energy change on vaporization of pure liquids (62 solvents, 10 temperature points for each solvent) from 6.6% to 3.1%. The binary interaction parameters kij are fitted to the solubility data of solid solutes in pure solvents or estimated from a semi-empirical equation using a similar local composition method, and then they are used to predict the solute solubility in mixed solvents. The overall average relative error (ARE) in predicted solubility (15 solutes, 64 systems, 1091 data points) is 11.24%, which makes this model a practical approach for crystallization process development.

Solubility of clonazepam and diazepam in binary and ternary mixtures of polyethylene glycols 400 or 600, propylene glycol and water at 298.2K - experimental data and modeling

Experimental molar solubilities of clonazepam and diazepam in binary and ternary mixtures of polyethylene glycols (PEGs) 400 or 600, propylene glycol (PG) and water (138 data points) along with the density of the saturated solutions at 298.2K were reported. The Jouyban-Acree model was used to fit to the measurements for providing a computational method. Employing the solubilities in the mono-solvents, the measured solubilities in mixed solvents were back-calculated and the overall mean percentage deviations (OMPDs) of the model were 16.0 % and 19.2% for diazepam and clonazepam, respectively. Addition of the Hansen solubility parameters to the model helps us to train all the data sets (clonazepam and diazepam) at once and the back-calculated OMPD for this analysis was 19.3%.