Gas Solubility Data Research Articles

Evaluation of Data on Solubility of Simple Apolar Gases in Light and Heavy Water at High Temperature

The solubility data of apolar gases in light and heavy water over the temperature range covered experimentally have been evaluated, laying particular emphasis to the region above the normal boiling points of the solvents. The systems that have been included in this work are the inert gases and CH4 in light water and heavy water, H2, O2, N2, and C2H6 in light water and D2 in heavy water. Data in the original sources have been brought to the same footing by calculating from the raw experimental data P, T, and x when they were not reported by the author. This step is considered necessary to assess critically the available sets of data. The temperature dependence of Henry’s constants for all the binary systems have been expressed in terms of two different polynomial equations. The formulations presented are discussed and the limits of application given.

Calculation Of Gas Solubility In Wabasca Bitumen

Abstract Calculations are described for the solubility of five gases, namely, N2, CO, CH4 CO2 and C2H6 in Wabasca bitumen at temperatures ranging from 23 °C to 110 °C. The gas solubilities were calculated using the Peng-Robinson equation of state, and these were compared with published data. The best results were obtained by modelling the bitumen as a mixture of three pseudocomponents which represent the distillable maltene, the undistillable maltene and the asphaltene fractions. Introduction In the past, we have published data for gas solubility, density and viscosity of Athabasca, Marguerite Lake, Peace River, Wabasca and Cold Lake bitumens(1–7). The gases whose effectson bitumen properties have been measured include N2, CO, CH4, CO2, CZH6, and mixtures containing two or more of these gases. Gas solubility and density data for Athabasca and Cold Lake bitumens were also reported by Fu et al.(8, 9). To make the data useful in simulating oil sand or heavy oil reservoirs and/or processes, empirical correlations for gas solubility and viscosity of bitumens were provided(2–3). On the other hand, bitumen characterizations were developed(10–12) for the calculation of bitumen gas-solubility using the Peng-Robinson (P-R)(13) equation of state (EOS). In these studies, bitumens were subdivided into several pseudocomponents and the Kesler-Lee(14) correlations were shown to provide a satisfactory estimation of the critical properties and acentric factor for each bitumen pseudocomponent(7, 11, 12). Other gas solubility calculation methods(8, 9, 15, 16) were based on lumping the bitumen as one pseudocomponent. To match the calculated gas solubilities with the data, temperature-dependency for the binary interaction parameter was proposed. The gas solubility results presented in this paper are based on the use of a three-pseudo-component ("3PC") characterization scheme that was developed for the calculation of the viscosity of gas-free Wabasca bitumen using the corresponding states method(17). For bitumen viscosity calculation, the use of a one-component characterization scheme has been shown to give unacceptable results(18). In recent studies(11, 12), we have attempted to demonstrate that the use of multicomponent bitumen characterization yields a better match of gas solubility data. Furthermore, multicomponent bitumen characterizations would be needed for simulating some very interesting reservoir phenomena such as "fingering" and "distillation" effects. In surface facilities calculations, particularly separation simulations, one-component bitumen/heavy-oil characterizations would obviously be useless. Similarly, vapour-liquid-solid (V-L-S) multiphase equilibrium for bitumen-diluent systems can only be described with muiticomponent bitumen characterizations(19). It is emphasized that the data for gas-bitumen phase equilibria of only the whole bitumens are presently available in the literature. That is, there are no reported data or properties for the fractions or ‘cuts’ of Alberta bitumens. Obviously, a more rigorous approach for predicting bitumen properties will be based on the characterization parameters and phase equilibrium data for each of the bitumen fractions. Generally, the larger the number of pseudocomponents the better one would expect to model the heavy-oil/bitumen thermodynamic and transport properties. However, increasing the number of pseudocomponents that the heavy-oi

Gas solubility and Henry's law near the solvent's critical point

AbstractIt has been experimentally observed, for water and nonaqueous solvents alike, that Henry's constant passes through a maximum and then declines as the temperature is raised from the triple point to the critical point. From classical and nonclassical models, we derive exact relations for the value of Henry's constant and its temperature dependence at the solvent's critical point, showing that the decline of this constant is a universal phenomenon. We demonstrate that the limiting temperature dependence of Henry's constant can be predicted from the thermodynamic properties of the pure solvent and the initial slope of the critical line. The validity of our prediction is tested by comparing it with experimental solubility data for several gases in high‐temperature water and benzene. Our predictive model appears valid over a temperature range of at least 15% in temperature below the critical point of the solvent.

Properties Of Peace River Bitumen Saturated With Field Gas Mixtures

Abstract This paper presents new data for the viscosity and gas solubility of Peace River bitumen saturated with CO2 and two field gases. The Peace River Produced Gas was rich in carbon dioxide while the Three Creeks Gas was rich in methane. The experimental results show that the field gas-saturated bitumen properties are between those of bitumen saturated with pure carbon dioxide and methane. These tests were performed on a new experimental apparatus for measuring the properties of gas-saturated bitumens. In addition to providing data on the viscosity and gas-solubility of bitumen, the apparatus is capable oj detecting the presence of solid (asphaltene) particles. Another unique feature of the experimental procedure allows for continuous sampling of released gas as the pressure is lowered. With each of the three gases, experiments were performed at four temperatures ranging from 15 ºC to 120 ºC and pressures up to 10 MPa. In experiments with field gas mixtures composition of the released gas showed a considerable variation with system pressure. The CO2 content in the released gas was found to gradually increase as the system pressure was reduced to atmospheric pressure. For methane and other light gases the trends were quite different. Introduction Reserves of conventional crude oils in Alberta have been for some time declining. However, Alberta has the world's largest known deposits, 200 billion m3, of bituminous oil sands(1). Additionally, there are significant deposits of heavy oil, natural gas and coal. The major obstacle to the economic recovery and processing of heavy oils and bitumens is their high viscosity. To facilitate the recovery of these heavy oils and bitumens a Viscosity reduction process – such as gas-miscible or gas immiscible flood, steam stimulation, or in situ combustion – is necessary. In addition, adequate process equipment design requires a knowledge of the viscosity of heavy hydrocarbon/gas systems. The viscosity data for the following gas-free Alberta bitumens have been provided in the past: Athabasca (GCOS) bitumen(2–5), Athabasca (ARC) bitumen from ARC Sample Bank(6), Peace River bitumen from a core sample(7), Peace River bitumen from PRISP facility(8), Marguerite Lake bitumen(9), Wabasca bitumen(10), and Cold Lake bitumen(11). Figure 1 shows the effect of temperature on the viscosity of gas-free Alberta bitumens. It is obvious that the bitumens from Athabasca and Peace River regions have high viscosities, from Marguerite Lake and Cold Lake regions have intermediate viscosities, and a sample of Wabasca bitumen is far less viscous. Table 1 presents a simple bulk analysis for the Alberta bitumens. Detailed analyses of Alberta bitumens are presented in the AOSTRA Data Book(12). The data in Table 1 and Figure 1 indicate that bitumen viscosity increases generally with an increase in the asphaltene content and the average molecular weight of the bitumen. The viscosity, density and gas-solubility data reported by the authors for several bitumen-gas systems are summarized in Table 2. Also shown in Table 2 are reference numbers to published papers in which the data appear. In general, the order of decreasing solubility of gases in bitumens is C2H6. CO2, CH4, CO and N2.

Characterization Of Athabasca Bitumen For Gas Solubility Calculations

Abstract The solubility of five pure gases, namely N2 CO, CH4, CO2 and C2H2, in Athabasca (GCOS) bitumen is calculated using the Peng-Robinson EOS and with the use of the literature bitumen characterizations and a new ‘3PC’ characterization. The calculated gas solubilities are compared with the published data over a temperature range of 20 °C to 110 °C and pressures up to 10 MPa. The best over-all match of the gas-solubility data is obtained with the ‘3PC’ scheme which models the bitumen as a mixture of three pseudocompoments. Optimized values of the binary interaction parameter for use in the Peng-Robinson EOS are also provided. Introduction The objective of this paper is to evaluate bitumen characterizations for calculating the solubility of five gases, namely N2, CO, CH4, CO, and C2 H6, in Athabasca bitumen by use of the Peng-Rpbinson(l) equation of state (EOS). Bitumen is a highly complex mixture of light, medium and heavy organic compounds. Hence, it is not reasonable to model the bitumen as a single pseudocomponent even for use in simple calculations. Three bitumen characterization schemes for Athabasca (GCOS) bitumen are considered in this paper. The first scheme of Mehrotra et al.(2), with the bitumen-split suggested originally by Bishnoi et al.(3), consisted of five pseudocomponents. The second scheme of Johnson et al.(4), consisting of four pseudocomponents, had been used for predicting the viscosity of gas-free and gas-saturated Athabasca bitumen by use of a corresponding states method. A third scheme consisting of three pseudocomponents (termed the ‘3PC’ scheme) was developed in the present study. For all pseudocomponents in each of the above mentioned schemes, the critical properties and acentric factor were estimated from the Kesler-Lee(5) correlations, which have been shown to produce satisfactory results for the solubility of CO2, in Athabasca and Peace River bitumens(2). The literature on the measurement and prediction of thermodynamic properties of light hydrocarbons/gas systems is extensive. However, the literature specifically related to the properties of Alberta bitumen/gas systems is still very limited.

Correlation and prediction of gas solubility in cold lake bitumen

AbstractThe solubility data for pure light gases, namely N2, CH4, CO2 and C2H6, in Cold Lake bitumen are correlated by use of the Peng‐Robinson equation of state. Modeling Cold Lake bitumen as a mixture of three pseudocomponents adequately matched two sets of the gas‐solubility data. The critical properties and acentric factor for the bitumen pseudocomponents were estimated using the Kesler‐Lee correlations. For each gas‐bitumen pair, a single constant value of the binary interaction parameter is shown to predict the gas‐solubility data with average deviations ranging from 2 to 8%. Subsequently, these calculations were extended to predict the solubility of CO2+CH4 gas‐mixtures in Cold Lake bitumen. It is shown that the solubility predictions for the gas mixtures data are higher by approximately 6 to 9%.

Relationship between gas separation properties and chemical structure in a series of aromatic polyimides

Gas permeabilities and permselectivities for various gas pairs [He/CH 4, CO 2/CH 4, N 2/CH 4, and O 2/N 2] are reported for a series of aromatic polyimides synthesized in our laboratory. Gas solubility and diffusivity data are also reported for some of the polyimides. Systematic variations in chemical structure were found to lead to significant changes in permeabilities and selectivities. Further analysis shows that the permeability and selectivity changes in these cases are primarily due to the alteration of diffusivity factors. Generally, increases in permeability are attended by losses in permselectivity; however, some of the polyimides in the present series of materials were found to deviate very favorably from this typical behavior. Specifically, materials were discovered having simultaneously higher permeabilities and permselectivities than commercial polymers currently being used as membrane materials. The relationship between gas separation properties and chemical structure is discussed in terms of intrasegmental mobility and intersegmental packing of the constituent polymers. The experimental data and their physical implications suggest the possibility that many new polymers having both high permeability and selectivity can be designed by tailoring the intrasegmental mobility and intersegmental packing of membrane polymers.

Gas solubility data bank for further development of phase equilibrium models

A data bank for gas solubilities in non-electrolyte systems was set up at the Chair of Industrial Chemistry, University of Dortmund in collaboration with the group of Professor Knapp (Berlin) as a part of the existing Dortmund Data Bank. This data bank contains experimental data taken from the original references, whereby the data are stored in the original units. A number of computer programs for further processing of the data has been developed: these include programs for data retrieval, unit conversion, data regression and analysis using different empirical and semi-empirical models and different possibilities for graphical representation. The main purpose was to establish a computerized set of reliable data for model development. One of the models used is the mod. UNIFAC method which has been extended for the prediction of gas solubilities.

Corresponding States Method For Calculating Bitumen Viscosity

Abstract A method based on the extended principle of corresponding states is presented for calculating the viscosity of Alberta bitumen's. Bitumen's from four regions in Alberta, namely Athabasca, Peace River, Marguerite Lake and Wabasca, are examined in this study. The bitumens are modelled as mixtures of two, three and four pseudocomponenis. The Kesler-Lee correlations and the method of Riedel are used to estimate the critical properties and acentric factor of each bitumen pseudocomponent. Using PSU No. 625 (1,2,3.4,5,6,7,8-octahrydrophenanthrene) as the reference fluid and the shape factors of Leach et al. as modified by Johnson, the calculated bitumen viscosities compare satisfactorily with the published viscosity data over a wide range of temperatures. The calculated viscosities are within about 10% of the data over a wide viscosity range of 0.5 to 271 Pa.s. Introduction An obstacle to the economic recovery and processing of heavy oils and bitumen's is their high viscosity. To facilitate the recovery of heavy oils or bitumen's, a viscosity reduction process such as gas-miscible flood, steam stimulation or in-situ combustion, is necessary. In addition, adequate process equipment design requires a knowledge of the viscosity of heavy oils or bitumen. Due to the time and costs involved in obtaining experimental data, a method to predict the viscosity of heavy oils and bitumen's is required. Viscosity data for gas-free and gas-saturated bitumens have been presented in the literature. Jacobs et al.(1) and Svrcek and Mehrotra(2) provided the viscosity data for gas-saturated Athabasca bitumen. The viscosity, density and gas-solubility data for Marguerite Lake, Peace River, Wabasca and Athabasca bitumens were provided by Mehrotra and Svrcek(2–7). The effect of temperature on the viscosity of gas-free bitumens has also been shown in these investigations(1–7). Based on the above experimental viscosity data, viscosity correlations have been presented in the literature. Correlations for the viscosity and gas-solubility of Athabasca bitumen were developed by Mehrotra and Svrcek(1). Khan et al. (8) provided modified Eyring and Hildebrand equations that correlated the viscosity of gas-free Athabasca bitumen. Two other empirical orrelations were also proposed for the effect of temperature on the viscosity of Athabasca bitumen(8), Sarkar(9) developed a graphical method for correlating the viscosity of gas-saturated bitumens. Johnson(10) presented a method based on the extended principle of corresponding states for the prediction of Athabasca bitumen viscosity. It was shown that the use of methane as the reference fluid gave inadequate viscosity predictions for Athabasca bitumen, However, an aromatic-naphthenic compound (1,2,3,4,5,6,7 ,8-octahydrophenanthrene) as the reference fluid gave viscosity predictions that were within 6% of the gas free bitumen viscosity data of Jacobs(11). This approach also gave satisfactory predictions for the viscosity of gas-saturated Athabasca bitumen(10). It should be noted that this method docs not require any viscosity data, hence the method is truly predictive. The bitumen viscosity predictions in Johnson's method were obtained by characterizing the bitumen as a mixture of pseudo (or hypothetical) components, This type of characterization approach has also been shown to give satisfactory prediction of vapour-liquid and vapour-liquid-liquid equilibrium for bitumen/gas systems as well as the surface tension of Athabasca bitumen(12,13).

The influence of pH on gas solubilities in aqueous solutions of sodium octanoate at 25°C

Measurements have been made to determine the solubilities of oxygen, methane, ethane, and propane in aqueous solutions of sodium octanoate at 25°C and at pH values ranging from 7 to 13. The solubility of each gas obeys Henry's law at all surfactant concentrations and pH values. The solubilities measured for each gas are effectively independent of sodium octanoate concentration as well as pH at concentrations well below the CMC, and, within experimental error, are the same as the respective gas solubilities in pure water at 25°C. Above the CMC, the solubility of each gas increases with surfactant concentration indicating micellar solubilization. The degree of solubilization is greatest for propane and decreases in the order: propane > ethane > methane > oxygen. However, unlike the situation at low surfactant concentrations, the individual gas solubilities vary with pH at higher surfactant concentrations. The degree of solubilization is found to be quite constant for each gas at pH values greater than 9, and intramicellar solubilities derived from gas solubility data taken at pH 12 are identical to those determined for a similar surfactant having a seven carbon alkyl chain, sodium 1-heptane sulfonate. However, the extent of solubilization for each gas increases rapidly as the solution pH decreases from 9 to 7, suggesting that mixed micelles composed of octanoic acid and sodium octanoate are more effective solubilizing agents than those of the pure soap.

The solubility of gases in aqueous solutions of sodium 1-heptanesulfonate and sodium perfluorooctanoate

Measurements have been made to determine the solubilities of oxygen, argon, methane, ethane, propane, and tetrafluoromethane at elevated pressures in aqueous solutions of sodium 1-heptanesulfonate and sodium perfluorooctanoate. The solubility of each gas follows Henry's law at all surfactant concentrations. In the case of sodium 1-heptanesulfonate, gas solubility data are obtained above and below the critical micelle concentration (CMC). Below the CMC, the solubility of each gas is approximately the same as that in pure water. Above the CMC, the solubility of each gas increases with surfactant concentration indicating micellar solubilization. Intramicellar solubilities of the various gases are calculated from this latter data for each surfactant. The intramicellar solubilities for the various gases dissolved in sodium 1-heptanesulfonate agree well with values predicted by the Laplace pressure model for micelles composed of surfactant ions having an alkyl chain length of seven carbon atoms. The 1-atm micellar solubilities of these same gases in sodium perfluorooctanoate are found to be considerably larger than those measured for sodium 1-heptanesulfonate, even though the alkyl chain lengths of the two surfactant anions are the same. The enhanced capacity for gas solubilization exhibited by the perfluorooctanoate micelles can be explained satisfactorily in terms of bulk properties of perfluorocarbon solvents using the Laplace pressure model.

Consideration of some dilute-solution phenomena based on an expression for the Gibbs free energy

Rigorous expressions based on the Lennard-Jones (6-12) potential, are presented for the Gibbs and Helmholtz free energy of a dilute mixture. These expressions give the free energy of the mixture in terms of the thermodynamic properties of the pure solvent, thereby providing a convenient means of correlating dilute mixture behavior with that of the pure solvent. Expressions for the following dilute binary solution properties are derived: Henry's constant, limiting activity coefficients with their derivatives, solid solubilities in supercritical gases, and mixed second virial coefficients. The Henry's constant expression suggests a linear temperature dependence; application to solubility data for various gases in methane and water shows a good agreement between theory and experiment. In the thermodynamic modeling of supercritical fluid extraction, we have demonstrated how to predict new solubility-pressure isotherms from a given isotherm, with encouraging results. The mixed second virial coefficient expression has also been applied to experimental data; the agreement with theory is good.

Temperature dependence of errors in parameters derived from Van't Hoff studies

Applying the method of Clarke and Glew to high-precision gas solubility data as an example that serves to illustrate how one can determine various thermodynamic parameters and their associated errors at various temperatures.

Thermodynamics of transfer of noble gases in hydrophobic solvents and in phospholipid membranes

A theory of gas solubility has been formulated and tested using solubility data of noble gases in several hydrophobic solvents and in a phospholipid membrane of dimyristoyl lecithin. The theory describes solubility in terms of two independent processes: hole creation in the solvent and solute adsorption in these holes. It gives the standard thermodynamic functions of solution in terms of four independent microscopic parameters. Two parameters reflect the nature of the pure solvent and the others the nature of the solute and its interactions with the solvent. Examination reveals good agreement between theory and experiment for the solvents tested, paving the way for comparisons between membranes and bulk solvents which are more significant than the existing correlations. The characteristics of the theory and its agreement with experiment suggest an interesting method for evaluating, from the solubility data of gases, the physical properties of hydrophobic regions of membranes and of other solvents which cannot be measured directly.

Prediction of gas solubilities by a modified UNIFAC equation

The design of absorption processes and of gas-liquid reactors requires the knowledge of gas solubility data. Because experimental data are often not available, an estimation method has to be used. Various methods for the prediction of gas solubilities have been published in the literature. It is shown that none of these methods provides satisfactory results. Therefore a new method - the modified UNIFAC method - is proposed. This modification of the UNIFAC method takes into account free volume effects. The method has been used successfully for the estimation of gas solubilities in aliphatic hydrocarbons, alcohols and ketones at 1 atm.

Prediction of Gas Solubilities by a Modified UNIFAC‐Equation

AbstractThe design of absorption processes requires the knowledge of gas solubility data. Because experimental data are often not available, an estimation method must be used. Various methods for the prediction of gas solubilities have been published in the literature. It is shown that none of the described methods provides satisfactory prediction results. Therefore a new method – the modified UNIFAC method – is proposed. This modification of the UNIFAC method takes into account free volume effects. The method has been used successfully for the estimation of carbon dioxide, methane, oxygen and ethylene solubilities in aliphatic hydrocarbons and alcohols at 1 atm.

Correlations For Properties of Bitumen Saturated With CO2 , CH4 And N2 , And Experiments With Combustion Gas Mixtures

Abstract The solubility in bitumen of a synthetic combustion gas (SCG) mixture of target composition, '17% CO2, 1% CH, and 82% N2 was experimentally measured. Along with the viscosity and density measurements of gas-saturated bitumen, the compositions of dissolved and equilibrium gases were also determined. Correlations are presented that predict solubilities of pure CO2CH4 and N2 gases, and the viscosity of saturated bitumen as functions of pressure and temperature. The experiments with SCG mixtures reveal that the mixture solubility is somewhat lower than that predicted by pure component partial solubilities. Although individual gas solubilities are not additive, the composition of dissolved gas can be predicted within a reasonable accuracy using the pure component solubility data. Introduction The solubility of carbon dioxide, methane and nitrogen gases in Athabasca bitumen along with the viscosity and density data for gas-saturated bitumen were presented elsewhere(l). The need for suitable correlations for the prediction of viscosity and solubilities of gases in bitumen cannot be overemphasized. Such correlations will find extensive applications in numerical simulation of the extraction of tar-sands by in-situ techniques such as CO2-injection, fire-flooding, etc. The quantitative description of the bitumen viscosity behaviour is a rather complicated matter. Due to a great range in the properties and composition of tar-sands bitumen, it may be of interest to introduce a means of oil characterization. However, the complete analysis of bitumen is virtually impossible at the present time. Hence, bitumen may be characterized either by the viscosity of gas-free oil (the dead oil), reflecting the sum total contribution of all the components present, or by partial, analysis in terms of several real arid hypothetical components: Bishnoi et al.(2) presented a computational scheme based on the latter approach for the prediction of C02 and CH4 solubilities in bitumen. Their calculation procedure used the Peng-Robinson equation of state and a Chao-Seader correlative method. The bitumen was characterized by 5 "cuts" of hypothetical components with assigned molecular weights, densities, mole fractions and boiling points. Their computed predictions were remark a by close to the experimental results(3). Most of the heavy hydrocarbon viscosity correlative methods in the literature are based on dead oil viscosities. However, there is no literature on the correlation of gas-solubility data that may be suitable for a bitumen system. The empirical correlations for viscosity prediction reflect the effects of temperature and/ or pressure. These correlations also depend on the kind of gas which is dissolved in the hydrocarbon phase. (Figure in full paper) comparative basis by Tsturyants and Muradov(4). These authors investigated the effects of methane, ethane and propane under saturation conditions. At the same mole fraction of a dissolved gas, the highest viscosity decrease was caused by propane followed by ethane and then methane. The change in viscosity became more pronounced as the total molar concentration in gas (i.e. pressure) was increased. Tumasyn et al.(5) studied the effect of CO2 on oil viscosity. Their experiments clearly showed that oil viscosity decreases with the amount of carbon dioxide dissolved. Furthermore, the drop in oil viscosity was higher for the oils that had a high dead oil viscosity.

Molecular dynamics simulation of silicon dioxide glass

Abstract A molecular dynamics study of the thermodynamic, structural and dynamical properties of a model silicon dioxide glass is described. The system consists of 3000 ions in a cubic volume. The results of the present simulation are compared with the results of our earlier simulation of a small system containing only 375 ions. The glass transition point is identified by a sudden change in the value of specific heat and pressure. Below this point diffusion also ceases. The structure of simulated glass is found to be a three-dimensional continuous network of SiO4 tetrahedra, connected at the oxygen nodes with a broad distribution of Si-O-Si angle centred around 153°. Other structural features, namely radial distribution functions and the interference function, are found to be in good agreement with the experimentally observed values. The holes in this open network system are interconnected, and form a continuous network. The existing inert gas solubility data in SiO2 glass are explained by using the hole ...

Distribution of holes in simulated silicon dioxide glass

The configuration of silicon dioxide glass simulated by molecular dynamics has been analysed to understand the free volume (holes or cavities) distribution. The hole size distribution is used to explain the inert gas solubility data of silicon dioxide glass.

Model of sorption of simple molecules in polymers

AbstractA model of simple molecule sorption in polymers is proposed which embraces both the glassy and rubbery regions, and incorporates the successful dual‐mode model below the glass‐transition temperature. Hole filling is shown to be an important sorption mechanism both above and below Tg, although saturation effects do not occur in the rubbery polymer. The model interprets the “dual‐mode” Langmuir and Henry's law parameters at the molecular level, and a simple statistical mechanical analysis allows estimation of the parameter values, as well as specifying certain interrelationships between the parameters. Applications of the model to gas solubility data in five polymers are considered [polyethylene (PE), poly(ethylene terephthalate) (PET), polystyrene (PS), polymethacrylate (PMA), poly(vinyl acetate) (PVAc)] and semiquantitative agreement is obtained for PE, PET, and to a lesser extent, PS. For PMA and PVAc, the agreement is qualitative only.