Liquid Partition Coefficients Research Articles

Physicochemical Properties and Activity Coefficients at Infinite Dilution for Organic Solutes and Water in the Ionic Liquid 1-Decyl-3-methylimidazolium Tetracyanoborate

The activity coefficients at infinite dilution γ(13)(∞) and gas−liquid partition coefficients K(L) for 43 solutes—alkanes, alkenes, alkynes, cycloalkanes, aromatic hydrocarbons, alcohols, carboxylic acids, thiophene, ethers, ketones, and water—in the ionic liquid 1-decyl-3-methylimidazolium tetracyanoborate ([dmim][TCB]) were determined by gas−liquid chromatography at temperatures from 318.15 to 378.15 K. The partial molar excess Gibbs energies ΔG(1)(E,∞), enthalpies ΔH(1)(E,∞), and entropies ΔS(1)(E,∞) at infinite dilution were calculated from the experimental γ(13)(∞) values obtained over the temperature range. Additionally, the densities for investigated ionic liquid over the temperature range were determined. The selectivities for the aliphatic/aromatic hydrocarbon separation problem were calculated from γ(13)(∞) and compared to the literature values for N-methyl-2-pyrrolidinone (NMP) and sulfolane. It was found that the investigated [dmim][TCB] ionic liquid shows much higher capacity than NMP and sulfolane and selectivity on the same level as NMP and lower than that for sulfolane for the separation of aliphatic hydrocarbons from aromatic hydrocarbons.

Activity coefficients at infinite dilution of organic solutes in the ionic liquid 1-ethyl-3-methylimidazolium methanesulfonate

Infinite dilution activity coefficients γ 1 ∞ and gas–liquid partition coefficients K L of 30 selected hydrocarbons, alcohols, ketones, ethers, esters, haloalkanes, nitrogen- and sulphur-containing compounds in the ionic liquid (IL) 1-ethyl-3-methylimidazolium methanesulfonate [EMIM][MeSO 3] were determined by gas–liquid chromatography at five temperatures in the range from 318.15 to 353.15 K. Relative contribution of adsorption at gas–liquid interphase to the overall solute retention, as examined by varying sample size and IL loading in the column, was found negligible. Partial molar excess enthalpies and entropies at infinite dilution were derived from the temperature dependence of the γ 1 ∞ values. The linear free energy relationship (LFER) solvation model was used to correlate successfully the K L values. The LFER correlation parameters and excess thermodynamic functions were analyzed to disclose molecular interactions operating between the IL and the individual solutes. Comparison to [EMIM][NO 3] studied previously revealed [MeSO 3] − anion to have still greater hydrogen bond basicity than that found for [NO 3] −. In addition, similarly to [EMIM][NO 3], promising potential of [EMIM][MeSO 3] for applications in solvent-aided separation processes was identified, the selectivities of [EMIM][MeSO 3] for separation of aromatic hydrocarbons and thiophene from saturated hydrocarbons ranking among the highest ever observed with ILs or molecular solvents.

Soret separation of highly siderophile elements in Fe–Ni–S melts: Implications for solid metal–liquid metal partitioning

Soret separation of melts consisting of Fe (+ 1–10 wt.% Ni) with added C, Si or S and trace highly siderophile elements (HSE; Ru, Rh, Pd, Re, Os, Ir, Pt and Au) was investigated in experiments done at 2 GPa, T hot ~ 2000 °C and T hot−T cold ~ 250 °C. Experiments with added C and Si produced homogeneous samples, whereas those with S resulted in significant compositional gradients, with S enriched at the hot end of the sample, and Fe and HSE enriched at the cold end. The magnitude of the separation of the HSE is not the same for each, but varies in the order (smallest to largest) Pd~Au<Rh<Ru~Pt<Ir~Re~Os. We show that the observed distribution of HSE within experiments is consistent with the variation in activity coefficients for these elements as predicted by the non-metal avoidance model of Jones and Malvin (1990). Thus, our results indicate that the melt composition effect on solid metal–liquid metal partitioning can be captured in a single thermal diffusion experiment. The magnitude of the melt composition effect, as judged by the Jones–Malvin interaction parameter, β, varies with metal size in a manner similar to solid metal–liquid metal partition coefficients, suggesting like origins of the HSE “selectivity” for Fe-rich solid or liquid structures.

Evaluation of the Affinity of Some Toxic Elements to Schwertmannite in Natural Streams Contaminated with Acid Mine Drainage

In order to evaluate fixation potential of schwertmannite for fluvial transport of various toxic elements, we examined bottom precipitates and stream waters collected from the rivers contaminated with acid mine drainage (AMD), which arose from the abandoned Nishinomaki mine (Shimonita, Gunma, Japan). Mineralogical and morphological observations revealed that schwertmannite was the main mineral of the precipitates. The affinity of various toxic ions to schwertmannite was evaluated on the basis of (1) apparent solid–liquid partition coefficients (Kd’s) between precipitates and stream waters, (2) coprecipitation behaviors during schwertmannite formation in a laboratory test, and (3) consideration on coprecipitation processes using partial charge model (PCM). As a result, oxyanions of V, As, Mo and Sb, Kd’s of which were relatively large (>104 (ml g−1)), were considered to be immobilized by schwertmannite precipitates. A laboratory test also demonstrated that these ions except Mo coprecipitated with schwertmannite. In addition, partial charges and average electronegativities predicted on the basis of PCM suggested that the oxyanions of V, As, Mo, and Sb could create stable inner sphere complexes with schwertmannite embryos, which results in their high affinity to schwertmannite. On the other hand, cationic ions of Mn, Cu, Zn, Sr, Cs, and U, Kd’s of which were relatively small (<104 (ml g−1)), were thought to have a tendency to flow downstream without uptake by schwertmannite precipitates. All these results suggested that schwertmannite has high fixation potential for fluvial transport of various toxic oxyanions in AMD-contaminated rivers.

Linear free energy relationship (LFER) correlations for the solubilising characterisation of room temperature ionic liquids containing triethylsulphonium and 1-butyl-1-methylpyrrolidinium cations

Gas-to-room temperature ionic liquid (RTIL) partition coefficients have been compiled from the published literature for solutes dissolved in triethylsulphonium bis(trifluoromethylsulphonyl)imide, {[E3S]+[(Tf)2N]−}, and in 1-butyl-1-methylpyrrolidinium trifluoromethanesulphonate, {[BMPyr]+[Trif]−}. These partition coefficients were converted into water-to-RTIL partition coefficients using the corresponding gas-to-water partition coefficients. Both sets of partition coefficients were analysed using the Abraham model with cation-specific and anion-specific equation coefficients. Equation coefficients are reported for the triethylsulphonium and 1-butyl-1-methylpyrrolidinium cations. The calculated cation coefficients can be combined with our previously determined eight sets of anion-specific equation coefficients to yield expressions capable of predicting the partition coefficients of solutes in 16 different RTILs.

Linear free energy relationship correlations for the solubilising characterisation of room temperature ionic liquids containing 1-hexyloxymethyl-3-methylimidazolium and 1,3-dihexyloxymethylimidazolium cations

Gas-to-room temperature ionic liquid (RTIL) partition coefficients have been compiled from the published literature for solutes dissolved in 1-hexyloxymethyl-3-methylimidazolium bis(trifluoromethylsulphonyl)imide, {[HxomMIm]+[(Tf)2N]−}, and in 1,3-dihexyloxymethylimidazolium bis(trifluoromethylsulphonyl)imide, {[(Hxom)2Im]+[(Tf)2N]−}. These partition coefficients are converted into water-to-RTIL partition coefficients using the corresponding gas-to-water partition coefficients. Both sets of partition coefficients are analysed using the Abraham model with cation-specific and anion-specific equation coefficients. The equation coefficients are reported for the 1-hexyloxymethyl-3-methylimidazolium and 1,3-dihexyloxymethylimidazolium cations. The calculated cation coefficients can be combined with our previously determined nine sets of anion-specific equation coefficients to yield expressions capable of predicting the partition coefficients of solutes in 18 different RTILs.

Prediction of Partition Coefficients of Organic Compounds in Ionic Liquids: Use of a Linear Solvation Energy Relationship with Parameters Calculated through a Group Contribution Method

A group contribution method is proposed to determine linear solvation energy relationship parameters (GC-LSER) in view of estimating the gas-to-ionic liquid partition coefficients and water-to-ionic liquid partition coefficients. Large sets of partition coefficients were analyzed using the Abraham solvation model to determine the contributions of 21 groups: 12 groups characterizing the cations and 9 groups for the anions. The derived equations correlate the experimental gas-to-ionic liquid and water-to-ionic liquid partition coefficient data to within 0.15 and 0.17 log units, respectively. The 21 group-parameters can be used to predict the partition coefficients of solutes in alkyl or functionalized ionic liquids with a good accuracy.

Determination of descriptors for semivolatile organosilicon compounds by gas chromatography and non-aqueous liquid–liquid partition

The measurement of retention factors by gas chromatography on up to 15 stationary phases at several temperatures in the range 60–240 °C for each stationary phase and liquid–liquid partition coefficients in three biphasic organic solvent systems ( n-hexane–acetonitrile, n-heptane– N, N-dimethylformamide, and n-heptane–2,2,2-trifluoroethanol) were used to calculate solute descriptors for 28 semivolatile linear and cyclic organosilicon compounds for use in the solvation parameter model. Regression analysis for oligomeric compounds allowed the descriptor values for the dimethylsiloxane, diethylsiloxane, methylvinylsiloxane, and methylhydrosiloxane monomer groups to be estimated. These monomer groups contribute significantly to the hydrogen-bond basicity ( B descriptor) and cavity formation and dispersion interactions ( L and V descriptors) of oligomeric organosilicon compounds, are non-hydrogen-bond acidic ( A descriptor = 0), are slightly less dipolar/polarizable than an n-alkane ( S descriptors are negative and close to zero), and bind electron lone pairs more tightly than an n-alkane ( E descriptor is small and negative). The semivolatile organosilicon compounds with polar functional groups in their side chain demonstrate a much wider range of selective intermolecular interactions than alkylsiloxanes and have solvation properties closer to those of polar organic compounds.

Activity Coefficients at Infinite Dilution of Organic Solutes in the Ionic Liquid 1-Ethyl-3-methyl-imidazolium Nitrate

Infinite dilution activity coefficients gamma(1)(infinity) and gas-liquid partition coefficients K(L) of 30 selected hydrocarbons, alcohols, ketones, ethers, esters, haloalkanes, nitrogen- and sulfur-containing compounds in the ionic liquid (IL) 1-ethyl-3-methylimidazolium nitrate [EMIM][NO(3)] were determined by gas-liquid chromatography at five temperatures in the range from 318.15 to 353.15 K. Relative contribution of adsorption at gas-liquid interphase to the overall solute retention, as examined by varying sample size and IL loading in the column, was found negligible. Partial molar excess enthalpies and entropies at infinite dilution were derived from the temperature dependence of the gamma(1)(infinity) values. The linear free energy relationship (LFER) solvation model was used to correlate successfully the KL values. The LFER correlation parameters and excess thermodynamic functions were analyzed to disclose molecular interactions operating between the IL and the individual solutes. In addition, the promising potential of [EMIM][NO(3)] for applications in solvent-aided separation processes was identified, the selectivities of [EMIM][NO(3)] for separation of aromatic hydrocarbons and thiophene from saturated hydrocarbons ranking among the highest ever observed with ILs or molecular solvents.

LFER correlations for the solubilising characterisation of room temperature ionic liquids containing trifluoromethanesulfonate and trifluoroacetate anions

Gas-to-RTIL (room temperature ionic liquid) partition coefficients have been compiled from the published literature for solutes dissolved in 1-methyl-3-butylimidazolium trifluoromethylsulfonate, {[MBIm]+[Trif]−}, and in 1-methyl-3-ethylimidazolium trifluoroacetate, {[MEIm]+[F3Ac]−}. These partition coefficients were converted into water-to-RTIL partition coefficients using the corresponding gas-to-water partition coefficients. Both sets of partition coefficients were analysed using the Abraham model and Goss modified Abraham models with cation-specific and anion-specific equation coefficients. Equation coefficients are reported for the trifluoromethylsulfonate and trifluoroacetate anions. The calculated anion coefficients can be combined with our previously determined eight sets of cation-specific equation coefficients to yield expressions capable of predicting the partition coefficients of solutes in 16 different RTILs.

Air−Liquid Partition Coefficient for a Diverse Set of Organic Compounds: Henry’s Law Constant in Water and Hexadecane

The SPARC vapor pressure and activity coefficient models were coupled to successfully estimate Henry's Law Constant (HLC) in water and in hexadecane for a wide range of organic compounds without modification to, or additional parametrization of, either SPARC model. The vapor pressure model quantifies the solute-solute intermolecular interactions in the pure liquid phase, whereas the activity coefficient model quantifies the solute-solvent and solvent-solvent (in addition to the solute-solute) interactions upon placing solute, i, in solvent, j. These intermolecular interactions are factored into dispersion, induction, dipole-dipole, and H-bonding components upon moving a solute molecule from the gas to the liquid phase. The SPARC HLC calculator so produced was tested and validated on the largest experimental HLC data set to date: 1356 organic solutes, spanning a wide range of functional groups, dipolarities and H-bonding capabilities, such as PAHs, PCBs,VOCs, amides, pesticides, and pharmaceuticals. The rms deviation errors for the calculated versus experimental log HLCs for 1222 compounds in water and 563 in hexadecane were 0.456 and 0.192 log [(mol/L)/(mol/L)] units, respectively, spanning a range of more than 13 and 20 log HLC dimensionless units for the compounds in water and hexadecane, respectively. The SPARC calculator web version is available for public use, free of charge, and can be accessed at http://sparc.chem.uga.edu.

Mass Transfer of Ozone in a Novel In Situ Ozone Generator and Reactor

A porous tubular reactor that also served as an electrode for ozone generation was studied in this research to determine the effects of in situ ozone generation on mass transfer and reaction rates. Experimental data over a range of gas flow rates and ozone generation rates gave KL a values in the range 0.77–1.14 min−1 . These values are more than double the values typically reported for bubble columns, and about 30% higher than that for packed beds. The specific power requirement for the laboratory-scale in situ reactor is an order of magnitude lower than that for bubble columns and stirred tank reactors that are used for ozone dissolution. A compartments-in-series fluid flow model was developed to describe the reactor system, and this model provides a good comparison to the experimental data for dissolved ozone and off-gas concentrations in the reactor. Sensitivity analyses indicate that the dissolved and off-gas ozone profiles are most sensitive to the gas–liquid partition coefficient and the overall ma...

Iridium, ruthenium and rhodium in komatiites: Evidence for iridium alloy saturation

A large body of new data (over 400 samples) on platinum-group element (PGE) concentrations in Yilgarn Craton komatiites is combined with all available high-quality literature data, to reveal trends in a spectrum of komatiite lithologies ranging from spinifex-textured rocks to pure olivine and olivine–chromite adcumulates. The data set is restricted to Carius Tube-isotope dilution analyses, and high-precision nickel sulfide fire-assay-ICP-MS analyses, and we demonstrate good agreement between the two methods. Only S-poor samples are considered, and the data set is filtered to eliminate the effects of addition of cumulus chromite, and addition or segregation of magmatic sulfide liquid. The residual trends indicate that Ir, Ru and Rh are decoupled from one another and from Pt and Pd. Rhodium behaves more like Pt and Pd, whereas Ru is intermediate between Ir and Rh. Rhodium and Ru both show indications of olivine control with partition coefficients less than unity. The maximum Ir whole-rock content is a roughly linear positive function of MgO, but the minimum is constant, sub-1 ppb and independent of MgO. Ruthenium shows some evidence for limited control by chromite, but anomalous Ru enrichment is also found in some olivine cumulate samples with no evidence for accumulated chromite or sulfide. Neither Rh nor Ir shows evidence for any detectable control by chromite. The distribution of Ir, and to a lesser degree Ru and Rh, cannot be adequately explained by olivine control, and a major influence of an Ir-rich phase is strongly indicated. The pattern of Ir variance in relation to MgO bears some important similarities to that of Cr. This similarity arises from the fact that both elements are controlled by a temperature-dependent solubility limit: the chromite saturation surface in the case of Cr, and saturation in Ir (or Ir–Os) alloy in the case of Ir. Combining this effect with moderate compatibility in olivine allows us to successfully model the entire range of Ir data. The upper limit of Ir concentration in olivine cumulates, in an exact analogy to that for Cr, is interpreted as the cotectic ratio of olivine to alloy, as determined by the average slope of the tangent to the alloy solubility curve. Sulfide-undersaturated mantle melts of a wide range of compositions are likely to be Ir alloy saturated. Experimental predictions of very low Ir solubilities in magmas are borne out, with the further implication that true thermodynamic sulfide–silicate liquid partition coefficients, at least for Ir and by inference for the other PGEs, are of the order of 10 7 or even higher.

Aroma release from gum arabic or egg yolk/xanthan-stabilized oil-in-water emulsions

The partitioning and release of limonene and trans-2-hexenal from oil-in-water emulsions was evaluated by applying static headspace gas chromatography. Experiments were carried out on gum arabic/xanthan (GA/X) and egg yolk/xanthan (EY/X)-stabilized emulsions. The dispersed phase (φ0 = 0.05 and 0.20) was composed of sunflower oil, while the oil droplet size (d32) ranged from 0.8 to 12.5 μm. The type of emulsifying/stabilizing system influenced the thermodynamic component, as expressed by the air–liquid partition coefficient (Ka/l) values at equilibrium (37 °C), of the release of limonene only. Increase in lipid fraction decreased the release of both aroma compounds. The oil droplet size had no significant impact on the partition coefficients of trans-2-hexenal, whereas for a lipid content of 20% (v/v) and GA/X stabilizing system, the partition coefficient of limonene was considerably reduced. Additionally, information on the aroma compound diffusion through the liquid–gas interface was obtained by calculating the initial slopes of the time–release curves. Both aroma compounds diffused more slowly through the GA/X-stabilized emulsions compared to EY/X-stabilized systems, while the release rate of limonene was significantly decreased when oil droplet size was increased. This behaviour was attributed to changes in viscosity as well as to reinforcement of xanthan network because of the presence of gum arabic. On the other hand, the nature of the interfacial film and the extent of oil droplet interactions in the EY/X-stabilized emulsions is considered to be responsible for their lower ability to retain the two aroma compounds. Odour assessment experiments were also conducted and the results were found to partially correlate with gas chromatographic results.

Determination of descriptors for organosilicon compounds by gas chromatography and non-aqueous liquid–liquid partitioning

Measurements of retention factors by gas chromatography on up to 10 complementary stationary phases at up to 5 temperatures for each stationary phase and liquid–liquid partition coefficients in three biphasic organic solvent systems ( n-hexane-acetonitrile, n-heptane- N, N-dimethylformamide and n-heptane-2,2,2-trifluoroethanol) were used to estimate solute descriptors for 54 organosilicon compounds for use in the solvation parameter model. Many of the E descriptor values (electron lone pair interactions) are negative for simple siloxanes and silanes indicating that these compound bind electron lone pairs more tightly than n-alkanes. Silanes and siloxanes with alkyl groups have near zero dipolarity/polarizability ( S descriptor). The S descriptor is only modest for simple phenylsilanes, silazanes, silanols, orthosilicates, and alkoxides. All organosilicon compounds with silicon-oxygen bonds are reasonably strong hydrogen-bond bases ( B descriptor) but only the silanol group is a reasonably strong hydrogen-bond acid ( A descriptor). Silanes (Si H) and silazanes (Si NH Si) are weak hydrogen-bond acids. Cavity formation and dispersion interactions ( V or L descriptor) are often the main component of solvation models for siloxanes and silanes that have simple alkyl and aromatic substituents. A number of physicochemical properties (vapor pressure, aqueous solubility, biphasic partition coefficients, sorption coefficients, etc.) for linear and cyclic dimethylsiloxanes can be reliably predicted from their descriptors in established models for organic compounds.

Chemical imaging with NanoSIMS: A window into deep-Earth geochemistry

We use a combination of nanometer-resolution secondary ion mass spectrometry (NanoSIMS) and analytical transmission electron microscopy (ATEM) for chemical imaging of material transformed in a laser-heated diamond anvil cell (LH-DAC), in the pressure and temperature range of Earth's lower mantle. MORB (mid-ocean ridge basalt), one of the components of subducted oceanic lithosphere, was transformed to an assemblage of Mg-perovskite, Ca-perovskite, stishovite and a calcium ferrite-structure phase at 55 GPa and 2100 °C in an LH-DAC. Elemental imaging spanning the entire range of concentrations, from major elements such as silicon (49.5 wt.% SiO 2) to trace elements such as strontium (118 ppm), scandium, and yttrium (both at 40 ppm) was obtained with a Cameca NanoSIMS 50. We observe a preferential partitioning of scandium, yttrium and strontium in the calcium silicate perovskite phase, and we compare this to recently measured solid–liquid partition coefficients and fractionation at lower pressures. This type of measurement demonstrates that even the most complex mineral assemblages can be probed using this combination of techniques and opens new pathways towards the characterization and quantification of geochemical interactions and processes occurring in the deep Earth.

Characterization of Room-Temperature Ionic Liquids by the Abraham Model with Cation-Specific and Anion-Specific Equation Coefficients

Gas-to-RTIL (room-temperature ionic liquid) partition coefficients have been compiled for 592 different solute-RTIL combinations. These partition coefficients were converted into water-to-RTIL partition coefficients using the corresponding gas-to-water partition coefficients. Both sets of partition coefficients were analyzed using the Abraham solvation parameter model with cation-specific and anion-specific equation coefficients. The derived equations correlated the experimental gas-to-RTIL and water-to-RTIL partition coefficient data to within 0.10 and 0.14 log units, respectively. The 8 sets of calculated cation-specific equation coefficients and 4 sets of calculated anion-specific equation coefficients can be combined to yield expressions capable of predicting the partition coefficients of solutes in 32 different RTILs.

Predicting radium availability and uptake from soil properties

The results of a potted soil experiment to determine the soil and plant factors ruling radium availability and uptake by ryegrass and clover are described. Nine soils with distinct soil characteristics were spiked with 226Ra. They were thoroughly characterized and the solid liquid partitioning coefficient, K d , was determined. K d ranged from 38 l kg −1 to 446 l kg −1 (average: 188 ± 156 l kg −1) and was linearly related to cation exchange capacity (CEC) and organic matter (OM) content. The soil-to-plant transfer factor (TF) was significantly affected by the chemical properties of the soils and ranged from 0.054 kg kg −1 to 0.719 kg kg −1 for ryegrass and from 0.034 kg kg −1 to 1.494 kg kg −1 for clover. Overall, no significant difference in TF between ryegrass and clover was observed. TF was related to K d , to CEC, OM (for ryegrass only when excluding one soil) and the calcium concentration in the soil solution (for both plants if excluding one soil). Radium flux were calculated from the radium concentration in the soil solution and the evapotranspiration, to predict total radium uptake derived from shoot radium concentration and biomass yield. It was found that radium uptake could be predicted from the radium flux ( R 2 = 0.61 and 0.83 for ryegrass and clover, respectively). Higher predictability ( R 2 = 0.70 and 0.91 for ryegrass and clover, respectively) was obtained when relating total radium uptake to a radium flow considering competition effects at the root surface by bivalent cations.

Gas–liquid partition coefficients and Henry's law constants of methyl mercaptan in aqueous solutions of Fe(II)–CDTA chelate complex

Utilization of ferric chelate complex of trans-1,2-cyclohexanediaminetetraacetic acid (CDTA) for the oxidative scrubbing of H 2S and CH 3SH in Kraft mill streams can be beneficial from the standpoints of iron protection against precipitation and oxygen-mediated regenerative oxidation of the ferrous chelate CDTA. The physical solubility of methyl mercaptan in CDTA–Fe(III) complex cannot be measured directly because of oxidation of the sulfur-bearing gaseous species with the ferric chelate. Therefore, this investigation was carried out to determine the gas–liquid partition coefficients and Henry's law constants of methyl mercaptan in aqueous iron-free CDTA solutions and non-reacting ferrous chelate solutions (CDTA–Fe(II) complex), using the static headspace method with an estimated accuracy of about 2%. Experiments with aqueous solutions of chelate concentrations varying between 38 and 300 mol m −3 were carried out at temperatures between 298 and 333 K and atmospheric pressure. It was shown that the methyl mercaptan solubility decreases with increasing temperature for all systems but was not much influenced, in the studied conditions, by the chelate concentration especially at larger temperatures.

A note on the temperature dependence of Henry's Law coefficients for methanol and ethanol

Measurements reported in the literature of gas–liquid partition coefficients for methanol and ethanol dissolved in water are compiled and critically evaluated to establish the temperature dependence. The data are linearly correlated in the ln( K H[mol dm −3 atm −1]) versus reciprocal absolute temperature coordinate frame and, when treated by a linear regression analysis, yield: ln( K H)=−(12.46±0.25)+(5312.4±76.0)/ T in the case of methanol (0–80 °C) and ln( K H)=−(15.87±0.82)+(6274.0±241.6)/ T in the case of ethanol (0–60 °C). The measurements at 25 °C average to K H(298.15)=(2.16±0.14)×10 2 mol dm −3 atm −1 ( n=8) and K H(298.15)=(1.94±0.13)×10 2 mol dm −3 atm −1 ( n=8), respectively. Enthalpies of solution derived from the temperature dependence are 44.17±0.63 kJ mol −1 for methanol and 52.16±2.01 kJ mol −1 for ethanol.