Model For Prediction Of Solubility Research Articles

Solubility of paracetamol in the ternary solvent mixtures of water + ethanol + glycerol at 298.2 and 303.2 K

ABSTRACT The solubility of paracetamol in the ternary mixtures of water + ethanol + glycerol at 298.2 and 303.2 K is determined via the shake-flask method. The reported data extend the solubility database for paracetamol in various solvent mixtures and also is employed to investigate the prediction capability of the cosolvency models for solubility prediction in the sub-binary and the ternary solvent systems. The solubility data are correlated with some cosolvency models and their accuracies are evaluated by the mean relative deviation (MRD) computed for back-calculated solubilities. The results show that the investigated models fit well and can be applied to the ternary solvent system, and the experimental data is relatively more consistent with the calculated values. The prediction ability of the trained Jouyban-Acree model for predicting drug solubility behaviour in the corresponding sub-binary systems is also investigated.

Do you know your r2?

The prediction of solubility of drugs usually calls on the use of several open-source/commercially-available computer programs in the various calculation steps. Popular statistics to indicate the strength of the prediction model include the coefficient of determination (r2), Pearson’s linear correlation coefficient (rPearson), and the root-mean-square error (RMSE), among many others. When a program calculates these statistics, slightly different definitions may be used. This commentary briefly reviews the definitions of three types of r2 and RMSE statistics (model validation, bias compensation, and Pearson) and how systematic errors due to shortcomings in solubility prediction models can be differently indicated by the choice of statistical indices. The indices we have employed in recently published papers on the prediction of solubility of druglike molecules were unclear, especially in cases of drugs from ‘beyond the Rule of 5’ chemical space, as simple prediction models showed distinctive ‘bias-tilt’ systematic type scatter.

Dissolving Cellulose in 1,2,3-Triazolium- and Imidazolium-Based Ionic Liquids with Aromatic Anions.

We present 1,2,3-triazolium- and imidazolium-based ionic liquids (ILs) with aromatic anions as a new class of cellulose solvents. The two anions in our study, benzoate and salicylate, possess a lower basicity when compared to acetate and therefore should lead to a lower amount of N-heterocyclic carbenes (NHCs) in the ILs. We characterize their physicochemical properties and find that all of them are liquids at room temperature. By applying force field molecular dynamics (MD) simulations, we investigate the structure and dynamics of the liquids and find strong and long-lived hydrogen bonds, as well as significant – stacking between the aromatic anion and cation. Our ILs dissolve up to 8.5 wt.-% cellulose. Via NMR spectroscopy of the solution, we rule out chain degradation or derivatization, even after several weeks at elevated temperature. Based on our MD simulations, we estimate the enthalpy of solvation and derive a simple model for semi-quantitative prediction of cellulose solubility in ILs. With the help of Sankey diagrams, we illustrate the hydrogen bond network topology of the solutions, which is characterized by competing hydrogen bond donors and acceptors. The hydrogen bonds between cellulose and the anions possess average lifetimes in the nanosecond range, which is longer than found in common pure ILs.

A constitutive model for predicting the solubility of gases in water at high temperature and pressure

In the drilling and development of gas fields, gas exhibits good solubility in drilling fluids and in the formation water under high temperatures and pressures. Therefore, the prediction of the amount of dissolved gas in water plays a vital role in the development of natural gas. In this work, the principle of equal fugacity in a gas-liquid equilibrium, a non-iterative gas solubility prediction model was established. The gas-phase fugacity coefficient was calculated by the Peng-Robinson equation of state combined with the non-random mixing rule, and the liquid phase was calculated by Henry's law. The model can predict the solubility of CO2, CH4 and C2H6 in pure water at temperatures of 273–448 K and pressures of 0–100 MPa, with an error of less than 5.5%; however, the calculation accuracy for H2S was low. It is found that the effect of moisture in the gas phase of gas-liquid equilibrium cannot be ignored. Compared with the existing mixed gas solubility experimental data, this model could accurately calculate the solubility of non-polar mixed gases, such as CO2 + CH4, and CH4 + C2H6 mixture. For the mixtures containing a polar gas, the lower the polar gas content, the smaller was the prediction error of the model for the non-polar gas in the mixture. On the contrary, the higher the content of polar gas, the larger was the deviation in the prediction of non-polar gas in the mixture, but lower was the deviation in the prediction of polar gas. Finally, the ability of Krichevsky-Kasarnovsky equation to predict the solubility of pure gas was evaluated.

Open Innovation Platform using Cloud-based Applications and Collaborative Space: A Case Study of Solubility Prediction Model Development

Open Innovation Platform using Cloud-based Applications and Collaborative Space: A Case Study of Solubility Prediction Model Development

Evolving an Accurate Decision Tree‐Based Model for Predicting Carbon Dioxide Solubility in Polymers

AbstractSolubility is one of the most indispensable physicochemical properties determining the compatibility of components of a blending system. Research has been focused on the solubility of carbon dioxide in polymers as a significant application of green chemistry. To replace costly and time‐consuming experiments, a novel solubility prediction model based on a decision tree, called the stochastic gradient boosting algorithm, was proposed to predict CO2 solubility in 13 different polymers, based on 515 published experimental data lines. The results indicate that the proposed ensemble model is an effective method for predicting the CO2 solubility in various polymers, with highly satisfactory performance and high efficiency. It produces more accurate outputs than other methods such as machine learning schemes and an equation of state approach.

Neural network modeling based double-population chaotic accelerated particle swarm optimization and diffusion theory for solubility prediction

Solubility is as a key chemical and physical property. Solubility prediction methods are applied in diverse fields including preparation synthesis and modifications of materials. To overcome the shortcomings of existing solubility prediction methods, taking the mass transfer of two-phase system as an example, a solubility prediction model based on the diffusion theory and hybrid artificial intelligence method was proposed in this paper. An improved double-population chaotic accelerated particle swarm optimization (APSO) algorithm combined diffusion theory was developed according to the particle evolution utilizing diffusion energy. The developed algorithm was applied in the training of parameters of the radial basis function artificial neural network and then a model for predicting solubility was developed. The experimental results of supercritical carbon dioxide solubility in 8 polymers were consistent with the predicted values by the model, indicating the high prediction accuracy. The average relative deviation, squared correlation coefficient, and root mean square error were respectively 0.0036, 0.9970, and 0.0152, displaying its higher comprehensive performance. The model may also be applied in other physicochemical fields.

QSPR Models for Prediction of Aqueous Solubility: Exploring the Potency of Randić-type Indices

QSPR Models for Prediction of Aqueous Solubility: Exploring the Potency of Randić-type Indices

A general model for prediction of BaSO4 and SrSO4 solubility in aqueous electrolyte solutions over a wide range of temperatures and pressures

The aqueous solubility of BaSO4 and SrSO4 in salt-containing solutions plays a crucial role in process industries and oilfield operations. Small variations in thermodynamic conditions result in precipitation and co-precipitation of these salts, mainly owing to their limited solubility in aqueous solutions. Understanding the environmental chemistry of BaSO4 and SrSO4 requires accurate knowledge of their solubility and thermodynamic behaviors. In this study, a general solubility prediction model based on a least square support vector machine and differential evolution algorithm was developed that is able to handle solubility predictions for BaSO4 in single salt solutions containing NaCl, CaCl2, MgCl2, KCl, KBr, Na2SO4, and Na2B4O7, and for SrSO4 solubility, it can handle predictions for single, binary, and ternary salt solutions containing NaCl, CaCl2, and MgCl2 over a wide range of temperatures and pressures. Model predictions agree excellently with experimental measurements, yielding an overall correlation coefficient (R2) of 0.9911 and an Average Absolute Error of 6.24%. A comparison of the general solubility prediction model with the Pitzer ion-interaction model revealed that both the models perform well in single salt solutions, while the Pitzer model fails to produce accurate solubility predictions when the aqueous system is extended to binary and ternary electrolyte mixtures.

Density-based UNIFAC model for solubility prediction of solid solutes in supercritical fluids

A new theoretical expression of group interaction parameter as a function of fluid density was put forward in this work. The group interaction parameters between the supercritical solvents (carbon dioxide, ethane, fluoroform and chlorotrifluoromethane) and the functional groups of solutes at reference fluid density were presented, which exhibit good compatibility with the UNIFAC Vapor-Liquid Equilibrium (VLE) group interaction parameters. For the solid Solute + C2H4 systems, satisfactory solubility data representation results can be obtained using the UNIFAC VLE group interaction parameters of “CC”. This density-based UNIFAC model was examined for the solubility of 20 solid solutes in 5 supercritical solvents (51 systems, 1111 data points); the Average Relative Errors (ARE) in solubility prediction are 19.1%, 18.7%, 21.6%, 20.4% and 12.8% for CO2, C2H6, C2H4, CHF3 and CClF3 correspondingly. In the case of each UNIFAC group being one molecule, the density dependent expression of group interaction parameter derived in this model may be applicable to other activity coefficient models (UNIQUAC, Wilson, NRTL) containing binary interaction parameters for solubility prediction of solids in supercritical fluids.

Trained models for solubility prediction of drugs in acetonitrile + water mixtures at various temperatures

ABSTRACT The trained versions of Jouyban–Acree and modified Wilson models are proposed to predict the drug's solubility in {acetonitrile + water} solutions at different temperatures. To obtain a full predictive model, the Abraham solvation and Hansen solubility parameters of solutes representing the interaction between components are combined with the proposed models. The solubility data of 12 drugs are fitted by these models. The overall average relative deviation percent (%ARD) values for the back-calculated solubility of drugs in aqueous acetonitrile solutions are 18.3–52.3 for the proposed models. The generally trained models provide acceptable estimation of the solubility of drugs and can be helpful in the pharmaceutical sciences.

A new computational method for drug solubility prediction in methanol + water mixtures

General solubility prediction models were developed using both classic least square and a novel method, in order to predict the solubility of the solutes in methanol + water binary solvent system. The novel approach to the regression analysis was investigated using an error minimization method. This aim was achieved by using a user-defined loss function regression, instead of the classic least square regression approach. To examine the results of the novel methodology, previous solubility data of 41 solutes were used for comparison. Both least square and novel methods were applied to the Jouyban-Acree model, Jouyban-Acree model in combination with Abraham parameters, and the modified Wilson model. The generally trained versions of the mentioned models produced more accurate predictions using the novel method than the least square method that has been confirmed by t-test analyses. The Jouyban-Acree model was the most accurate model among other generally trained models. Finally, the results were validated using a cross-validation analysis which produced the acceptable prediction accuracy of 24.6% mean percentage deviation (MPD) for the new methodology against 32.1% of the least square method. Also a new arithmetically transformed version of aforementioned models was introduced in this study to make the calculations easier to execute.

Micro-scale solubility assessments and prediction models for active pharmaceutical ingredients in polymeric matrices

The number of models for assessing the solubility of active pharmaceutical ingredients (APIs) in polymeric matrices on the one hand and the extent of available associated data on the other hand has been rising steadily in the past few years. However, according to our knowledge an overview on the methods used for prediction and the respective experimental data is missing. Therefore, we compiled experimental data, the techniques used for their determination and the models used for estimating the solubility. Our focus was on polymers commonly used in spray drying and hot-melt extrusion to form amorphous solid dispersions (ASDs), namely polyvinylpyrrolidone grades (PVP), polyvinyl acetate (PVAc), vinylpyrrolidone-vinyl acetate copolymer (copovidone, COP), polyvinyl caprolactam-polyvinyl acetate-polyethylene glycol graft polymer (Soluplus®, SOL), different types of methacrylate copolymers (PMMA), polyethylene glycol grades (PEG) and hydroxypropyl-methylcellulose grades (HPMC). The literature data were further supplemented by our own results. The final data set included 37 APIs and two sugar derivatives. The majority of the prediction models was constituted by the melting point depression method, dissolution endpoint measurements, indirect solubility determination by Tg and the use of low molecular weight analogues. We observed that the API solubility depended more on the working group which conducted the experiments than on the measuring technique used. Furthermore, this compilation should assist researchers in choosing a prediction method suited for their investigations.Furthermore, a statistical assessment using recursive feature elimination was performed to identify descriptors of molecules, which are connected to the API solubility in polymeric matrices. It is capable of predicting the criterium 20% API soluble at 100 °C (Yes/No) for an unknown compound with a balanced accuracy of 71%. The identified 8 descriptors to be connected to API solubility in polymeric matrices were the number of hydrogen bonding donors, three descriptors related to the hydrophobicity of the molecule, glass transition temperature, fractional negative polar van der Waals surface area, out-of-plane potential energy and the fraction of rotatable bonds. Finally, in addition to our own model, the data set should help researchers in training their own solubility prediction models.

Predicting the Solubility of Potential Redox Flow Battery Materials Using a Quantitative Structure-Property Relationship Models

Predicting the solubility and stability of redox-active organics across all states of charge is a challenging task. A trained chemist can often predict trends, but when it comes to numerical values, even the best computer software cannot be counted upon. In screening active materials for redox flow batteries (RFBs) – in which both properties are important – we are slave to trial and error, modifying organic molecules in attempt to optimize (increase) solubility and stability without compromising other important properties such as redox potential.1,2We hope to change our approach to materials development by better predicting properties in advance. Using quantitative structure-property relationship (QSPR), we have shown that properties of unknown organic molecules can be predicted when experimental properties of a training set of related compounds is provided.3Here we sought to expand this approach to a different redox core, phenothiazines, which we have evaluated as posolyte candidates for RFBs. We measured the solubility of about a dozen phenothiazine derivatives – in both relevant states of charge (neutral and radical cation) – in a nonaqueous electrolyte. The structure of these derivatives included variations in the number, type, and position of substituents to offer a diverse training set. We used our experimental values and the results of density functional theory calculations to develop a QSPR model for the phenothiazine core, following which we synthesized and measured the solubility of a new set of phenothiazines. Here we will report on the degree of success of our model for solubility prediction. References Kowalski, J. A.; Casselman, M. D.; Kaur, A. P.; Milshtein, J. D.; Elliott, C. F.; Modekrutti, S.; Attanayake, N. H.; Zhang, N.; Parkin, S. R.; Risko, C.; Brushett, F. R.; Odom, S. A., “A Stable Two-Electron-Donating Phenothiazine for Application in Nonaqueous Redox Flow Batteries.” J. Mater. Chem. A 2017, 5, 24371.Milshtein, J. D.; Kaur, A. P.; Casselman, M. D.; Kowalski, J. A.; Modekrutti, S.; Zhang, P. L.; Attanayake, N. H.; Elliott, C. F.; Parkin, S. R.; Risko, C.; Brushett, F. R.; Odom, S. A., “High Current Density, Long Duration Cycling of Soluble Organic Active Species for Non-Aqueous Redox Flow Batteries.” Energy Environ. Sci. 2016, 9, 3531.Sevov, C. S.; Hickey, D. P.; Cook, M. E.; Robinson, S. G.; Barnett, S.; Minteer, S. D.*; Sigman, M. S.; Sanford, M. S. "Physical Organic Approach to Persistent, Cyclable, Low-Potential Electrolytes for Flow Battery Applications." J. Am. Chem. Soc. 2017, 139, 2924.

The Development of Mode for Solubility of Oxygen in High Pressure Condition

Previous investigation of solubility of oxygen in high pressure is summarized in the paper. Summing-up shows that compared with various methods for such measurement, solubility theory is developed slowly. Models for prediction of oxygen solubility in aqueous acid and/or salt solution have been developed. However, no model can reproduce the existing experimental data, and extrapolate beyond the data range from simple aqueous salt system to complicated brine systems including seawater, simultaneously. A model to cover much wider temperature and pressure space in variable composition brine systems needs much more efforts. At last, the main consensuses and differences as well as the further work is reviewed.

Solubility of hydrocarbon and non-hydrocarbon gases in aqueous electrolyte solutions: A reliable computational strategy

Determining solubility of hydrocarbon and non-hydrocarbon components of natural gas is crucial for theoretical studies and engineering design. In this study, new solubility prediction models were developed for both hydrocarbon gases (methane, ethane, propane, and butane) and non-hydrocarbon gases (CO2 and N2) in aqueous solutions of strong electrolytes using a hybrid modeling strategy, which links the Coupled Simulated Annealing (CSA) to the Least-Squares Support Vector Machine (LSSVM) technique. Comparing the models’ predictions with experimentally determined solubility values, a very good agreement was noticed, leading to the overall correlation coefficients of 0.9880 and 0.9907 for the hydrocarbon and non-hydrocarbon gases, respectively. These models were also found to succeed in capturing the physical trends among experimental datasets through performing sensitivity analysis between the dependent and independent parameters. Developed models can be utilized to predict the solubility of pure and/or a mixture of hydrocarbon and non-hydrocarbon gases in aqueous electrolyte solutions, covering wide ranges of ionic strength, pressures, and temperatures up to supercritical conditions. Such a reliable predictive tool helps researchers and engineers to successfully obtain the key thermodynamic properties (e.g., solubility, vapor pressure, and compressibility factor), which are central to properly design and operate the corresponding units in a variety of chemical plants such as petrochemical plants, natural gas processing plants, and refineries.

A global version of modified Wilson model for solubility prediction of drugs in methanol + water mixtures

The trained version of modified Wilson model is proposed employing solubility data of 41 drugs and/or drug-like compounds in methanol + water mixtures at various temperatures. For covering the physicochemical properties of the solute to provide a QSPR model, the Abraham solvation parameters of solutes are combined with the modified Wilson model. The mean relative deviations for the correlated data are 32.5% and 27.1%, respectively for the modified Wilson model and its combined version with Abraham solvation parameters. Furthermore, an attempt is also made to correlate the solubility data with the modified Wilson model combined with van't Hoff equation and the Abraham general solvation parameters to provide a full predictive model for drug solubility prediction in methanol + water mixtures at various temperatures.

Model evaluation for the prediction of solubility of active pharmaceutical ingredients (APIs) to guide solid–liquid separator design

The assumptions and models for solubility modelling or prediction in systems using non-polar solvents, or water and complex triterpene and other active pharmaceutical ingredients as solutes aren't well studied. Furthermore, the assumptions concerning heat capacity effects (negligibility, experimental values or approximations) are explored, using non-polar solvents (benzene), or water as reference solvents, for systems with solute melting points in the range of 306–528 K and molecular weights in the range of 90–442 g/mol. New empirical estimation methods for the ΔfusCpi of APIs are presented which correlate the solute molecular masses and van der Waals surface areas with ΔfusCpi. Separate empirical parameters were required for oxygenated and non-oxygenated solutes. Subsequently, the predictive capabilities of the various approaches to solubility modelling for complex pharmaceuticals, for which data is limited, are analysed. The solute selection is based on a principal component analysis, considering molecular weights, fusion temperatures, and solubilities in a non-polar solvent, alcohol, and water, where data was available. New NRTL-SAC parameters were determined for selected steroids, by regression. The original UNIFAC, modified UNIFAC (Dortmund), COSMO-RS (OL), and COSMO-SAC activity coefficient predictions are then conducted, based on the availability of group constants and sigma profiles. These are undertaken to assess the predictive capabilities of these models when each assumption concerning heat capacity is employed. The predictive qualities of the models are assessed, based on the mean square deviation and provide guidelines for model selection, and assumptions concerning phase equilibrium, when designing solid–liquid separators for the pharmaceutical industry on process simulation software. The most suitable assumption regarding ΔfusCpi was found to be system specific, with modified UNIFAC (Dortmund) performing well in benzene as a solvent system, while original UNIFAC performs better in aqueous systems. Original UNIFAC outperforms other predictive models tested in the triterpene/steroidal systems, with no significant influence from the assumptions regarding ΔfusCpi.

Commentary On “Extended Hildebrand Approach: An Empirical Model for Solubility Prediction of Etodolac in 1,4-Dioxane and Water Mixtures”

Several mathematical errors in the published paper by Rathi and Deshpande (J Solution Chem 43:1886–1903, 2014) are identified. The errors concern the incorrect conversion of mass fraction to volume fraction concentrations of 1,4-dioxane, the incorrect conversion of mole fraction solubilities to molar solubilities of Etodolac, and the incorrect calculation of the ideal mole fraction solubility of Etodolac.

Solubility predictions of acetanilide derivatives in water: Combining thermochemistry and thermodynamic modeling

Knowledge about solubility in water is required for crystallization processes, for the development of structure-property relationships, for the establishment of solubility scales, assessing environmental contamination, and for validating thermodynamic models. Approaches are desired that allow predicting solubility without the use of any experimental solubility data. Most methods that have been proposed to predict aqueous solubility of organic compounds face low prediction reliability and the lack of model interpretability. This work proposes the use of a thermodynamic approach for the prediction of solubility of acetanilide and its derivatives in water. This approach requires fusion enthalpy and fusion temperature as well as the activity coefficient of the respective acetanilide derivative. The latter was obtained by the equation of state PC-SAFT, which uses thermochemistry data as input for model parametrization. The thermochemical data on acetanilide and its derivatives (vapor and sublimation pressures, sublimation and fusion enthalpies) were collected from the literature and evaluated for internal consistency. In order to validate the final solubility prediction model, aqueous solubility of acetanilide and 17 derivatives were predicted and compared to experimental solubility data from literature at 298.15 K as well as to an ideal solubility model, which assumes ideal mixture behavior. The results showed that mixtures of acetanilides + water are highly non-ideal, and the average deviations between solubility data and ideal solubility model could be reduced by two orders of magnitude by using PC-SAFT for the solubility predictions. More promising, PC-SAFT was found to allow predicting the temperature dependence of the aqueous solubility accurately, while ideal solubility model failed to quantitatively describe temperature-dependent solubility.