The solvation parameter model was observed to overpredict gas chromatographic retention factors (log k) for 1,4-benzodioxin on poly(siloxane) stationary phases containing 3,3,3-trifluoropropyl groups. Since this might be the result of a previously unidentified specific compound type-stationary phase retention mechanism not parameterized in the solvation parameter model, its origins were investigated. It is shown that the anomalous retention behavior is the result of a poor assignment of the excess molar refraction descriptor when calculated from its literature source refractive index. Reassignment of the excess molar refraction descriptor using experimental gas chromatographic retention factors and liquid-liquid partition constants for a varied and extensive experimental database provided an adequate explanation for the anomalous retention behavior for 1,4-benzodioxin.

The solvation parameter model employs six fixed descriptors (seven for compounds that exhibit variable hydrogen-bond basicity) to characterize the contribution of intermolecular interactions to equilibrium distribution properties. McGowan's characteristic volume (V) and excess molar refraction (E) for liquids at 20 °C can be calculated from structure and a refractive index value in the case of E. The remaining descriptors identified as dipolarity/polarizability (S), overall hydrogen-bond acidity (A), overall hydrogen-bond basicity (B or B°), the gas-liquid partition constant at 25 °C (L), and E for solids at 20 °C are experimental quantities more likely to vary when taken from different sources. There are two large, curated descriptor databases, known as the Abraham and Wayne State University (WSU-2025) databases that facilitate applications of the solvation parameter model to separation systems. These databases were developed independently using different experimental data. In some cases, this has led to significantly different descriptor assignments for the same compound. In this report, descriptor quality for over 300 compounds sourced from the above curated databases are compared using a generalized compound classification scheme and filters for descriptor difference values. For n-alkanes and monofunctional n-alkanes there are only small differences in descriptor values. Significantly larger differences are observed for multifunctional compounds including systematic differences in at least one descriptor for polycyclic aromatic hydrocarbons, phthalate esters, phenols, amides, and compounds of variable hydrogen-bond basicity. Partition constants for octanol-water, n-heptane-2,2,2-trifluoroethanol, and n-heptane-formamide biphasic systems, and gas chromatographic retention factors on a poly(3-cyanopropylphenyldimethylsiloxane) stationary phase are used to verify descriptor quality. The WSU-2025 database shows a significant improvement in model quality with better precision than the Abraham database descriptors as well as facilitating the identification of compounds likely to have misassigned descriptors in the Abraham database.

Abraham's solvation parameter model has been widely used to model retention in capillary micellar and microemulsion electrokinetic chromatography systems. To fit or predict retention factors in separation systems experimentally determined compound descriptors are required. These are typically obtained from two sources: the Abraham and Wayne State University (WSU) compound descriptor databases. Alternatively, compound descriptors can be estimated from molecular structures using computational, machine learning, and group contribution approaches without having to resort to experimental methods for compounds without assigned values in either database. Here, we compare descriptor quality for the Abraham and WSU experimental compound descriptor databases as well as computational approaches using group contribution and machine learning approaches for modelling retention in capillary micellar and microemulsion electrokinetic chromatography systems. It is shown that the WSU compound descriptor database affords a more accurate description for the retention of varied compounds while descriptors based on a machine learning approach more accurately modeled retention factors than a group contribution approach. Model standard errors and coefficients of determination for WSU descriptors varied from 0.046 to 0.116 and 0.976 to 0.996, respectively, while Abraham descriptors varied from 0.048 to 0.166 and 0.953 to 0.995, respectively. Model standard errors and coefficients of determination for machine learning descriptors ranged from 0.086 to 0.116 and 0.979 to 0.981, respectively, whereas group contribution descriptors ranged from 0.090 to 0.181 and 0.942 to 0.979, respectively. Descriptor quality is important when modeling retention in capillary micellar and microemulsion electrokinetic chromatography systems.

The solvation parameter model is used to characterize the retention properties of silica-based, reversed-phase liquid chromatography (RPLC) columns. System constants are obtained for the range 10-70 % (v/v) organic solvent at 10 % (v/v) increments for methanol-water (27 columns), acetonitrile-water (25 columns), acetonewater (7 columns), tetrahydrofuran-water (3 columns), and 2-propanol-water (1 column). The Wayne State University 2025 (WSU-2025) compound descriptor database is used as the sole source for compound descriptors to unify the system constants with a verified descriptor database. Column properties are described by system maps, a general feature of which is the dominant role of water in the retention mechanism. Water is more cohesive, hydrogen-bond acidic, and dipolar than common water-miscible organic solvents. The organic solvent moderates the controlling hand of water in the mobile phase while its selective absorption by the stationary phase is responsible for variation in the column phase ratio and contributes to the solvation properties of the stationary phase. A generic ligand designation is employed for the classification of octadecylsiloxane-type and phenylalkylsiloxane-type stationary phases for which the within the group variation of the system constants (selectivity) are comparatively small compared with those associated with changes of mobile phase composition. Changes in retention cannot simply be explained by differences in selectivity without also considering changes in the phase ratio with mobile phase type and composition. Hierarchical cluster analysis with the system constants as variables for a range of mobile phase compositions affords a useful classification tool to organize columns into groups with similar selectivity and to identify columns of different selectivity for method development. Octadecylsiloxane-type stationary phases with embedded polar functional groups, octylsiloxane-bonded, pentafluorophenylpropylsiloxane-bonded, and a perfluorohexylpropylsiloxane-bonded silica stationary phase are shown to have complimentary retention properties to octadecylsiloxane- and phenylalkylsiloxane-type stationary phases for both methanol- and acetonitrile-water binary mobile phase compositions. System constant correlation plots provide a useful tool for the identification of columns of equivalent selectivity. Ternary solvent systems are shown to provide a useful approach for method development.

Descriptors for fourteen semivolatile organic compounds associated with the authenticity, botanical origin, and flavor potential of the cinnamons of commerce were determined using the Solver method and experimental retention factors determined by gas chromatography at several temperatures on a minimum of seven selectivityselected, open-tubular columns and liquid-liquid partition constants in up to twenty totally organic biphasic systems. The six descriptors that encode the solvation properties of the compounds were used to predict watergas, octanol-gas, and octanol-water partition constants commonly employed to assess environmental distribution properties. For octanol-water partition constants, log KOW, the predicted partition constants exhibited an average absolute deviation of 0.12 for log KOW experimental - log KOW predicted (n = 14). Soil-water, soil-air, urban aerosol-air, skin-water permeation, and non-specific toxicity to the fathead minnow were predicted for the same compounds to assess their potential environmental impact. The product terms of the solvation parameter model provide a useful insight into the contribution of individual intermolecular interactions to the distribution properties of the cinnamon compounds and their environmental impact.

The solvation parameter model uses five system independent descriptors to characterize compound properties defined as excess molar refraction, E, dipolarity/polarizability, S, hydrogen-bond acidity, A, hydrogen-bond basicity, B, and the gas-liquid partition constant at 25 degrees C on n-hexadecane, L, to model transfer properties in gas-condensed phase biphasic systems. The E descriptor for compounds liquid at 20 degrees C is available by calculation using a refractive index value while E for solid compounds at 20 degrees C and the S, A, B, and L descriptors are determined by experiment. As a single-technique approach, it is shown that with up to 20 retention factor measurements on four columns comprising a poly(siloxane) containing methyloctyl or dimethyldiphenylsiloxane monomers (SPB-Octyl or HP-5), a poly(siloxane) containing methyltrifluoropropylsiloxane monomers (Rtx-OPP or DB-210), a poly(siloxane) containing bis(cyanopropylsiloxane) monomers (HP-88 or SGE BPX-90), and a poly(ethylene glycol) stationary phase (DB-WAXetr or HP-INNOWAX) are suitable for assigning the S, A, and L descriptors. Using the descriptors in the updated WSU compound descriptor database as target values the average absolute error in the descriptor assignments for 52 varied compounds in the temperature range 60-140 degrees C was 0.072 for E, 0.016 for S, 0.008 for A, and 0.022 for L corresponding to 30 %, 3.5 %, and 0.6 % as a relative average absolute error for E, S, and L, respectively. For the higher temperature range of 160-240 degrees C and 34 varied compounds that are liquid at 20 degrees C the average absolute error for the S, A and L descriptors was 0.026, 0.020, and 0.031, respectively, with the largest relative average absolute error for S of 3.2 % (< 1 % for the L descriptor). For 35 varied compounds that are solid at 20 degrees C the relative absolute error for the E, S, A, and L descriptors in the higher temperature range was 0.068, 0.035, 0.020, and 0.020, respectively, with a relative average absolute error for E (6.5 %), S (3.5 %) and L (0.88 %). The S, A, and L descriptor can be accurately assigned on the four-column system over a wide temperature range. The E descriptor for solid compounds at 20 degrees C exhibits greater variability than desirable. The B descriptor cannot be assigned by the four-column system, which lack hydrogen-bond acid functional groups, and is only poorly assigned on the weak hydrogen-bond acid ionic liquid column SLB-IL100.

Reversed-phase liquid chromatography (RPLC) offers significant advantages over traditional methods for estimating octanol-water partition constants, which are a critical parameter in drug discovery. In contrast to classical methods for determining the octanol-water partition constant, such as shake-flask techniques, RPLC is less time-consuming and easier to automate. In this study, we explored three alternative organic solvent modifiers: acetone, 2-propanol, and tetrahydrofuran for the indirect determination of the octanol-water partition constant for neutral compounds by RPLC using either isocratic retention factors or retention factors extrapolated to 100% water for several stationary phases. The Kinetex XB C18 column with acetone-water mobile phase compositions gave the best results for the construction of correlation models for the prediction of the octanol-water partition constant of varied compounds employing isocratic retention factors for 20 % (v/v) acetone-water and retention factors extrapolated to 100% water. Retention factors extrapolated to 100% water were nearly as good as the isocratic retention factors with a standard error of 0.136 and Fisher statistic of 1446 for the correlation model compared with a standard error of 0.122 and Fisher statistic of 1456 for the 20 % (v/v) acetone-water model. An XTerra MS C18 column with 10 % (v/v) tetrahydrofuran-water gave the best model for predicting the octanol-water partition constants with a standard error of 0.106 and Fisher statistic of 2806.

The solvation parameter model uses five system independent descriptors to characterize compound properties defined as excess molar refraction, E, dipolarity/polarizability, S, hydrogen-bond acidity, A, hydrogen-bond basicity, B, and McGowan's characteristic volume, V, to model transfer properties between condensed phases. The V descriptor is assigned from structure. For compounds liquid at 20°C the E descriptor can be assigned from the characteristic volume and its refractive index. The E descriptor for compounds solid at 20°C and the S, A, and B descriptors are experimental properties traditionally assigned from chromatographic, liquid-liquid partition, and solubility measurements. In this report liquid-liquid partition constants in totally organic and aqueous biphasic systems are evaluated as a standalone technique for descriptor assignments. Using six totally organic biphasic systems the S, A, and B descriptors were assigned with an average absolute deviation (AAD) of about 0.04, 0.03, and 0.04, respectively, compared with the best estimate of the true descriptor values for 65 compounds. The E descriptor for compounds solid at 20°C can only be estimated with an AAD of approximately 0.1. For six aqueous biphasic systems the B descriptor is assigned with a lower AAD of 0.028 and higher AAD of 0.08 and 0.05 for the S and A descriptors, respectively, than for the totally organic biphasic systems for compounds with a reliable value for the E descriptor. The preferred system for descriptor assignments utilizes both totally organic biphasic systems (heptane-1,1,1-trifluoroethanol, isopentyl ether-propylene carbonate, isopentyl ether-ethanolamine, heptane-ethylene glycol, heptane-formamide, and 1,2-dichloroethane-ethylene glycol-dichloroethane) and aqueous biphasic systems (octanol-water, cyclohexane-water) with the possible substitution of some systems with alternative systems of similar selectivity. For 55 varied compounds this combination of eight organic and aqueous biphasic systems resulted in an AAD of approximately 0.03, 0.02, and 0.02 for the S, A, and B descriptors compared to the best estimate of the true descriptor value. For 30 compounds solid at 20°C the AAD for the E descriptor of 0.11 is poorly assigned. The relative average absolute deviation in percent (RAAD) corresponds to 9.7%, 3.1%. 4.0% and 8.3% for E, S, A, and B, respectively, for the eight biphasic systems. Liquid-liquid partition is compared to reversed-phase liquid and gas chromatography as a standalone method for descriptor assignments.

Except for alkanes, most organic compounds are hydrogen-bond bases. The B° descriptor of the solvation parameter model provides a convenient measure of the effective (or summation) hydrogen-bond basicity of organic compounds. A fast and convenient method to assign the B° descriptor is required to support studies of hydrogen-bonding in separation systems. A two-column system with acetonitrile-water mobile phase compositions and the measurement of up to eleven isocratic retention factors is proposed for this purpose. Several reversed-phase column chemistries and mobile phases were evaluated with the two-column system consisting of a pentafluorophenylpropylsiloxane-bonded and octadecylsiloxane-bonded silica columns recommended for this purpose. To assess the accuracy of the method values for B° were taken from the Wayne State University (WSU) compound descriptor database, which were assigned using conventional multi-technique methods and large datasets. The two-column systems provided an unbiased assignment of B° with an average deviation of 0.008 and an average absolute deviation of 0.021 compared with the target value for 55 varied compounds. The two-column system is unsuitable for assigning the other descriptors used in the solvation parameter model and results in erroneous assignments of B° for nitrogen-containing compounds capable of electrostatic interactions on silica-based reversed-phase columns.

Solid-phase extraction is a sample preparation technique suitable for trace enrichment, matrix simplification, and media transfer for gas and solution phase samples. This technique employs adsorbents in cartridge, disk, or membrane format for sampling and thermal desorption or solvent elution for sample recovery. Typical adsorbents for sampling include inorganic oxides, low-specificity sorbents (e.g., carbon, porous organic polymers, etc.), organosiloxane-bonded silica materials, and class-specific sorbents (e.g., molecularly imprinted polymers, immunosorbents, surface-bound macrocyclic ligands, restricted access materials). To assist method development a fairly good understanding of the theory of operation (e.g., prediction of breakthrough volumes, sorbent selectivity, prediction of elution volumes) based on the plate height model for short sorbent beds and the solvation parameter model for retention is available for both cartridge and disk sampling devices. Empirical rules for sample processing are summarized for conditions not covered by theoretical models, often as a result of inadequate information regarding sample properties. Various approaches to partial or full automation of the sampling process are described for off-line and coupled-column systems.

The solvation parameter model uses six descriptors identified as excess molar refraction, E, dipolarity/polarizability, S, overall hydrogen-bond acidity, A, overall hydrogen-bond basicity, B, McGowan's characteristic volume, V, and the gas-liquid partition constant on hexadecane at 25°C, L to model the distribution of neutral compounds in biphasic systems. Abraham's version of this model uses all six descriptors with two separate linear free energy relationship models for the transfer of compounds from a gas phase to a condensed phase and between condensed phases. Goss proposed a modification to this model that uses a single calibration model regardless of the physical state for each phase and five of the descriptors employed in Abraham's model (E descriptor is eliminated). The capability of Abraham's model and the Goss-modified model to characterize the contribution of intermolecular interaction to retention for gas and reversed-phase liquid chromatographic systems and distribution in liquid-liquid partition systems is evaluated using the WSU compound descriptor database. These more accurate values for the Abraham descriptors have not been utilized previously for the evaluation of the Goss-modified model and should be more capable of discerning subtle differences in model performance. It is shown that model quality defined by statistical parameters favors Abraham's model over the Goss-modified model with differences in model quality greater for systems in which Abraham's model indicates a significant contribution from electron lone pair interactions and for systems in which one phase is a solvent containing perfluoroalkyl substituents. There is a small systematic difference for the terms describing the combined contributions of cavity formation and dispersion interactions and for interactions of a dipole-type. The contribution of hydrogen-bonding interactions is virtually identical for the two models. The model intercepts are generally different and potentially assigned to a larger contribution from lack-of-fit for the Goss-modified model. Although the Abraham model descriptors have been routinely employed for applications using the Goss-modified model the possibility that Goss-model specific descriptors should be employed was evaluated. Using the Solver method and Goss-model specific calibration models for chromatographic and liquid-liquid partition systems a new set of Goss-specific descriptors was calculated for 28 varied compounds. These descriptors show good statistical agreement with the Abraham descriptor values with an average deviation of 0.009, -0.003, -0.004, and -0.023, respectively, for the S, A, B, and L descriptors, corresponding to a relative absolute deviation in percent of 2.2%, 3.9%, 4.3%, and 1.2%, respectively.

The solvation parameter model uses six compound descriptors to model equilibrium properties in biphasic systems formally defined as excess molar refraction, E, dipolarity/polarizability, S, overall hydrogen-bond acidity, A, overall hydrogen-bond basicity, B, McGowan's characteristic volume, V, and the gas-liquid partition constant on hexadecane at 25°C, L. The V descriptor can be assigned from structure and the E descriptor for compounds liquid at 20°C can be calculated from its refractive index and characteristic volume. The E descriptor for compounds solid at 20°C and the S, A, B, and L descriptors are assigned from experimental properties traditionally obtained by chromatographic, liquid-liquid partition, and solubility measurements. Here I report an efficient experimental design using the Solver method for the accurate assignment of descriptors for neutral compounds that simultaneously minimizes laboratory resources. This multi-technique approach requires 3 retention factor measurements in a 60°C temperature range per compound on four columns by gas chromatography, 3 retention factor measurements in a 30 % (v/v) acetonitrile composition range per compound on two columns by reversed-phase liquid chromatography, and eight partition constant measurements by liquid-liquid partition in totally organic and aqueous biphasic systems for a total of 26 experimental measurements. The accuracy of the descriptor assignments was validated by comparison with the values in the Wayne State University (WSU) descriptor database taken as the best estimate of the true descriptor values. The E, S, A, B and L descriptors were assigned simultaneously by the Solver method using the above approach without significant bias and with an average absolute deviation (AAD) of 0.054, 0.018, 0.015, 0.013, and 0.040, respectively, compared with the WSU database values, corresponding to a relative absolute average deviation in percent (RAAD) of 7.2, 1.9, 3.6, 5.1, and 0.84 %, respectively, for 32 varied compounds. This streamlined approach represents a significant improvement on earlier single-technique approaches used as the starting point for the development of the multi-technique approach. For compounds of variable hydrogen-bond basicity modifications to the multi-technique approach were implemented while maintaining the same number of experimental measurements. Acceptable descriptor assignments for B/B° were obtained for compounds liquid at 20°C for which the E descriptor was available by calculation. For solid compounds at 20°C the E and B/B° descriptors are restricted to qualitative application where approximate values may be acceptable.

Biopartitioning processes are challenging to study and often require the sacrifice of multiple animals. Therefore, it is more practical and cost-effective to correlate these processes with easily determined properties, such as chromatographic retention data, or to make predictions based on structural descriptors such as quantitative structure-property relationships or linear free energy relationships. Abraham's solvation parameter model uses six solute properties to characterize the interactions responsible for the transfer of neutral compounds between immiscible phases in chromatographic or biological systems. This review discusses the prediction of biological properties of small molecules from chromatographic measurements and the solvation parameter model. It covers the characteristics of solute descriptors in the solvation parameter model, as well as experimental approaches for their determination. Additionally, it explores recent applications of the solvation parameter model in characterizing biological systems and its use in identifying surrogate chromatographic models for predicting biological properties.

Revised descriptors are determined for fifteen phthalate esters for use in the solvation parameter model and form part of the Wayne State University (WSU) compound descriptor database. For thirteen phthalate esters a comparison is made with the same compounds in the Abraham descriptor database. Gas chromatographic retention factors on poly(methyloctylsiloxane), SPB-Octyl, and poly(cyanopropylphenyldimethylsiloxane), DB-225, stationary phases are used to facilitate an assessment of the contribution of cavity formation and dispersion interactions, L descriptor, and dipole-type interactions, S descriptor, to the experimental retention factors (log k) for the phthalate esters with minimum interference from competing intermolecular interactions. The results indicate a systematic overprediction of the cavity and dispersion interaction term and underprediction of dipole-type interactions for the Abraham descriptors compared with the WSU descriptors for the phthalate esters. The average absolute deviation (AAD) for 13 phthalate esters on SPB-Octyl is 0.039 (WSU descriptors) compared with 0.252 (Abraham descriptors) and for 9 phthalate esters on DB-225 0.030 (WSU descriptors) compared with 0.167 (Abraham descriptors). The results for dipole-type interactions are confirmed and extended to include the hydrogen-bond basicity of the phthalate esters, B descriptor, by evaluation of partition constants in aqueous biphasic systems and the n-heptane-2,2,2-trifluoroethanol biphasic system. Differences in the contribution of the hydrogen-bond basicity of the phthalate esters to the experimental partition constants are largely random with respect to database selection but important for the accurate prediction of the partition constants. The AAD for the partition constant for 15 phthalate esters is 0.063 (WSU descriptors) compared with 0.320 (Abraham descriptors) for the heptane-2,2,2-trifluoroethanol biphasic system and 0.13 (WSU descriptors) compared with 0.25 (Abraham descriptors) for 9 phthalate esters in the octanol-water biphasic system. The WSU descriptors for the phthalate esters exhibit a better fit with the experimental data for separation systems and are free of the extreme values predicted for the Abraham descriptors for several phthalate esters.

Experimental RM values for a series of barbiturates were determined using reversed-phase high-performance liquid chromatography. Statistically significant linear relationships were found between these RM values and Hansch's π parameters. Relationships were also obtained between RM values and partition coefficients determined in the system diethyl ether-dimethylformamide-water (2:1:1) and partition coefficients determined using gas-liquid chromatography. Hansch's π parameters for anilines were correlated with RM values of some substituted urea derivatives.

High-performance column and thin-layer chromatography are both instrumental techniques but differ in that column chromatography requires a fully integrated instrument platform with high-pressure capability while for thin-layer chromatography separate devices are used for each unit operation, usually at or close to atmospheric pressure, and afford higher flexibility supporting online or offline operation. The unit operations of thin-layer chromatography are defined as layer pretreatment, sample application, development, and evaluation with derivatization as an optional step. The diversity of equipment for each operation contributes to the flexibility of analysis by thin-layer chromatography and supports manual, semiautomated, or full automation of the unit processes. Instrument platforms are more than a convenience as they affect performance, repeatability, sample detectability, and time management. The current trend in thin-layer chromatography is to make the unit operations independent of the user so that analysts can perform other tasks while each step is performed. In this chapter, we review contemporary instrumentation employed in thin-layer chromatography for sample application, development, derivatization, photodocumentation, densitometric evaluation, and hyphenation with spectroscopic detectors. Some recommendations for best practices are included.

The distribution of neutral compounds in biphasic separation systems can be described by the solvation parameter model using six solute properties, or descriptors. These descriptors characterize the size (McGowan’s characteristic volume), V, excess molar refraction, E, dipolarity/polarizability, S, hydrogen-bond acidity and basicity, A and B, and the gas-liquid partition constant on n-hexadecane at 298.15 K, L. McGowan’s characteristic volume and the excess molar refraction for liquids are available by calculation (E requires and experimental refractive index). The other descriptors and excess molar refraction for solids are experimental quantities and subject to greater variation or are estimated using computational or empirical models. Solute descriptors for several thousand compounds are available in the Abraham descriptor database and for several hundred compounds in the WSU descriptor database. These publicly accessible databases were developed independently using different approaches and for many compounds provide different descriptor values. In this report we evaluate the effect of mixing descriptors from the two databases on modeling chromatographic retention factors and liquid-liquid partition constants. It is shown that the two descriptor databases are not interchangeable. The WSU descriptor database consistently demonstrates improved model quality as determined by statistical parameters. Model system constants exhibit a general dependence on database selection with an approximately linear trend as a function of the fraction of compounds assigned descriptors from either database. There is no general model performance advantage to using mixed descriptor datasets and no real cause for concern for relatively large datasets containing < 15% of compounds with descriptors assigned from the other database. For small datasets, descriptor quality is an important variable for adequate model performance.

The distribution of neutral compounds in biphasic separation systems can be described by the solvation parameter model using six solute properties, or descriptors. These descriptors (McGowan's characteristic volume, excess molar refraction, dipolarity/polarizability, hydrogen-bond acidity and basicity, and the gas-liquid partition constant on n-hexadecane at 298.15 K) are curated in two publicly accessible databases for hundreds (WSU compound descriptor database) or thousands (Abraham compound descriptor database). These databases were developed independently using different approaches resulting in descriptor values that vary for many compounds. Previously, it was shown that the two descriptor databases are not interchangeable, and the WSU descriptor database consistently demonstrated improved model performance for chromatographic systems where the uncertainty in the dependent variable was minimized by suitable quality control and calibration procedures. In this report we wish to evaluate whether the same conclusions are true for models with a dependent variable containing significant measurement uncertainty. To evaluate this hypothesis, we assembled databases for water-air, octanol-air, and octanol-water partition constants reported by multiple laboratories using various measurement methods. It was found that database selection has little effect on model quality or model predictive capability but significantly affects the assignment of the contribution of individual intermolecular interactions to the dependent variable. The latter information is database specific, and a quantitative comparison of system constants should be restricted to models using the same compound descriptor database.

Revised descriptors for 74 varied compounds recommended for the characterization of reversed-phase silica columns using the solvation parameter model are assigned by the Solver method from experimental retention factors for calibrated gas-liquid and reversed-phase liquid chromatographic (RPLC) and biphasic liquid-liquid partition systems. These descriptors are taken as the best estimate of the true descriptor values and used to evaluate several RPLC systems as a single-technique approach for descriptor assignments. Various combinations of isocratic single column and multiple mobile phase compositions, multiple columns with a single mobile phase composition, and multiple columns and multiple mobile phase compositions are evaluated. A multiple column (Discovery HS C18 and HS F5, Fluophase-RP, and XBridge Shield RP18, Phenyl, and C-18) with acetonitrile- and methanol-water mobile phases provided the best results with an absolute average deviation (AAD) of 0.043 for the electron lone pair interaction descriptor E, 0.047 for the dipole-type interaction descriptor S, 0.020 for the hydrogen-bond acid descriptor A, and 0.010 for the hydrogen-bond base descriptor B degrees for 46 varied compounds. These values compare favorably with the larger all columns and mobile phase composition RPLC dataset with the best estimate of the true descriptors, or all data set, with AAD = 0.026 for E, 0.044 for S, 0.027 for A, and 0.011 for B degrees for 74 varied compounds. The preferred RPLC systems for descriptor assignments contain columns identified as belonging to different selectivity groups that maximize the relative magnitude of the dipole-type contribution, s system constant, hydrogen-bond basicity, a system constant, and electron lone pair interactions, e system constant. RPLC systems are well suited to assigning the B degrees descriptor and, with proper system selection, the A and S descriptors. The E descriptor for compounds unavailable by calculation is more problematic and a few extreme values with an AAD > +/- 0.1 were observed and can affect the reliability of the S descriptor assignments for these compounds. The likelihood of poor descriptor assignments using the Solver method can be identified by evaluating descriptor wells.

Extraction is the most common sample preparation technique prior to chromatographic analysis for samples which are too complex, too dilute, or contain matrix components incompatible with the further use of the separation system or interfere in the detection step. The most important extraction techniques are biphasic systems involving the transfer of target compounds from the sample to a different phase ideally accompanied by no more than a tolerable burden of co-extracted matrix compounds. The solvation parameter model affords a general framework to characterize biphasic extraction systems in terms of their relative capability for solute-phase intermolecular interactions (dispersion, dipole-type, hydrogen bonding) and within phase solvent-solvent interactions for cavity formation (cohesion). The approach is general and allows the comparison of liquid and solid extraction phases using the same terms and is used to explain the features important for the selective enrichment of target compounds by a specific extraction phase using solvent extraction, liquid-liquid extraction, and solid-phase extraction for samples in a gas, liquid, or solid phase. Hierarchical cluster analysis with the system constants of the solvation parameter model as variables facilitates the selection of solvents for extraction, the identification of liquid-liquid distribution systems with non-redundant selectivity, and evaluation of different approaches using liquids and solids for the isolation of target compounds from different matrices.