An improved group contribution method for PC-SAFT applied to branched alkanes: Data analysis and parameterization (original) (raw)
Related papers
The Journal of Supercritical Fluids, 2015
The current study aims at assessing universality of the recently proposed generalized predictive critical point-based PC-SAFT EoS (CP-PC-SAFT). Phase equilibria and the single phase thermodynamic properties of aromatic compounds and their mixtures are compared with the available experimental information in a wide range reaching at times the pressures of 6000 bars and the super-critical temperatures, while covering nearly 9000 data points. It is demonstrated that in spite of their significant density variations, the compounds under consideration can be included in the applicability range of CP-PC-SAFT. In addition, this model is an effective estimator of data in mixtures. However a major drawback of CP-PC-SAFT is the underestimation of vapour pressures away from critical points. A conceptually different predictive approach, namely the Hybrid Group-Contribution PC-SAFT (H-GC-PC-SAFT, Burgess et al., 2014) is considered as well. Although this model is often less precise than CP-PC-SAFT, in several cases, such as the low temperature vapour pressures, it exhibits noticeable advantages.
Fluid Phase Equilibria, 2004
A new group contribution method is proposed using the SAFT (Statistical Associating Fluid Theory) equation of state (EOS), in order to describe the thermodynamic properties of hydrocarbon series. The method is developed for vapor-liquid equilibrium (VLE) calculations for a large number of hydrocarbons, with the use of several group parameters. SAFT models are chosen for the physical meaning of their parameters. These can be related to the molecular structure. Two versions of the SAFT EOS are used in this work: The original SAFT equation of state, proposed by Chapman, et al. [Ind. Eng. Chem. Res., 29 (1990) 1709] and SAFT-VR (SAFT Variable Range) equation of state, proposed by Gil-Villegas, et al. [J. Chem. Phys., 106 ]. The present group contribution method consists in calculating the equation of state parameters (dispersion energy ε, segment diameter σ, chain length m and square-well range parameter λ for SAFT-VR) using group contribution rules. In this paper, we have treated pure compounds of five hydrocarbon families: n-alkanes, alkyl-benzenes, alkylcyclohexanes, α-olefins and 1-alkanols. The results obtained are compared with those of the usual approach (fitting the molecular parameters of each compound on its own properties) and seem to be nearly equivalent. The results of the present method are comparable with those of other predictive approaches.
Long chain multifunctional molecules with GC-PPC-SAFT: Limits of data and model
Fluid Phase Equilibria, 2012
The Group Contribution Polar Perturbed Chain Statistical Associating Fluid Theory (GC-PPC-SAFT) equation of state is here extended to the long chain multifunctional molecules as alkanolamines and alkanediols. The limits of data and model are discussed. For this, plots presenting the vapourization enthalpy (h T b) as a function of normal boiling temperature (T b) are used. The trend of the published data for alkanediols is found difficult to interpret using the theory. Supplementary data are generated from molecular simulations in order to further analyse the behaviour of this family, but the conclusion is that additional measurements are needed. The monofunctional (n-alkanes, primary alcohols and primary amines) and long chain multifunctional (diamines, alkanolamines and alkanediols) molecules are compared in the homologous series. The trends have led to modify the model parameters using the vapour pressure and the saturated liquid phase volume data. Two additional alcohol group (OH) parameters than these proposed by Nguyen-Huynh [D. Nguyen-Huynh et al., Fluid Phase Equilib. 264 (2008) 62-75] have been introduced. They are needed when the molecule has more than one associating group in its chain.
Developing a predictive group-contribution-based SAFT-VR equation of state
Fluid Phase Equilibria, 2009
The hetero-segmented version of the statistical associating fluid theory for potentials of variable range (hetero-SAFT-VR) is used to develop a predictive molecular-based group-contribution SAFT-VR equation of state (GC-SAFT-VR). The hetero-SAFT-VR approach models molecules composed of segments of different size and/or energies of interaction enabling an accurate description of real molecules composed of different functional groups. The differences in interactions between functional groups are maintained throughout the theory in contrast to other approaches in which the parameters for functional groups are averaged in order to model a molecule as a homonuclear chain with "group-averaged" parameters. Through the GC-SAFT-VR approach we can study the effect of molecular functionality and topology on the thermodynamic properties of real fluid systems, since parameters are determined for each functional group and chain connectivity is explicitly specified. In this initial study GC-SAFT-VR parameters are developed for key organic functional groups (CH 3 , CH 2 , CH 2 CH, C O, C 6 H 5 , OCH 3 and OCH 2) by fitting to experimental vapor pressure and saturated liquid density data for a selected group of compounds that contain these functional groups. The transferability of the parameters obtained is tested by comparing theoretical predictions with experimental data for pure fluids not included in the fitting process and binary mixtures. Using the GC-SAFT-VR approach good agreement is obtained between experimental data and the theoretical predictions for pure substances, including isomers, and their mixtures. The GC-SAFT-VR approach is able to accurately predict the effect of molecular functionality on mixture phase behavior without fitting to any experimental data for the system being studied.
Fluid Phase Equilibria, 2012
In the current study the widely implemented theoretically based model PC-SAFT and the recently proposed SAFT + Cubic have been applied for correlating and predicting various thermodynamic properties of 11 halomethanes. It has been found that thanks to the correct estimation of the experimental critical temperatures and pressures, SAFT + Cubic exhibits a superior robustness and reliability in modeling both pure compounds under consideration and their mixtures. In addition, SAFT + Cubic has a clear advantage in predicting sound velocities and isochoric heat capacities. However PC-SAFT is more accurate in modeling certain kinds of data, such as the isobaric heat capacities of saturated liquids and vapor pressures away from the critical points.
Industrial & Engineering Chemistry Research, 2008
A group-contribution statistical associating fluid theory equation of state (GC-SAFT EOS) that was proposed by Tamouza et al. [Tamouza et al. Fluid Phase Equilib. 2004, 222-223, 67-76], which was extended in the first part in this series of papers to the asymmetric systems CO 2 + n-alkane, methane + n-alkane, and ethane + n-alkane, is further tested here on binary mixtures that contain aromatic hydrocarbons, n-alkanes, CO 2 , N 2 , and H 2 S. The method for correlating the binary interaction parameters (k ij ), which is inspired by London's theory of dispersive interactions, uses only pure compound adjustable parameters ("pseudo-ionization energies" of compounds i and j, denoted as J i and J j ). A group contribution for the latter parameters also is used for n-alkane and alkyl benzene series. Numerous prediction tests on the aforementioned cited systems were performed in a systematic and comprehensive way. Predictions are both qualitatively and quantitatively satisfactory, within deviations (4%-5%) that are comparable to those obtained on previously investigated systems (n-alkane + n-alkane, n-alkane + aromatic, n-alkane + n-alkanol).
Prediction of the PC-SAFT Associating Parameters by Molecular Simulation
The Journal of Physical Chemistry B, 2012
In this work, we propose a new methodology to determine association scheme and association parameters (energy and volume) of a SAFTtype EoS for hydrogen-bonding molecules. This paper focuses on 1-alkanol molecules, but the new methodology can also be applied for any other associating system. The idea is to use molecular simulation technique to determine independently monomer and free hydrogen fractions from which the association scheme can be deduced. The 3B scheme thus appeared to be the most appropriate for 1-alkanols. Once the association scheme is defined, the association strength can be back-calculated from molecular simulation results and used as an independent property for the equation of state parameters regression, in addition of the classical phase properties such as vapor pressure and liquid molar volume. A new set of parameters for 1-alkanol for the PPC-SAFT equation of state has been proposed following this methodology. Results are found in good agreement with experimental data for both phase properties and free hydrogen-bonding sites. Hence, this new methodology makes it possible to optimize parameters allowing an accurate reproduction of pure compounds data and yielding physically significant values for associating energy and associating volume.
Energy & Fuels, 2009
A group contribution perturbed-chain statistical associating fluid theory (GC-PC-SAFT) equation of state (Tamouza et al. Fluid Phase Equilib. 2004, 222-223, 67-76) combined with a recent method for correlating k ij using only pure compound parameters (NguyenHuynh et al. Ind. Eng. Chem. Res., 2008. 47(22), 8847-8858) is extended here to model vapor-liquid phase equilibria of H 2 + alkanes and H 2 + aromatics mixtures. The correlation of k ij is inspired by London's theory of dispersive interactions, and uses "pseudo-ionization energies" J i and J j of compounds i and j as adjustable parameters. The GC-PC-SAFT parameters for alkanes and aromatics were reused from previous works when available. Otherwise, the missing parameters were estimated by regression of corresponding pure vapor-liquid equilibrium (VLE) data. Those of H 2 were determined in this work by correlating some VLE data of H 2 + n-alkane systems. Using the parameters thus obtained, the phase envelopes of other H 2 + alkane and H 2 + aromatic systems were fully predicted. The prediction tests were as comprehensive as possible. Correlations and predictions are qualitatively and quantitatively satisfactory. The deviations are within 5-6%, that is, comparable to those obtained on previously investigated systems. Mixtures containing H 2 are modeled here with deviations that compare well with those of the Grayson-Streed model (Grayson, H.G.; Streed, C.W.; Proc., 6 th World Pet. Congress, 1963, 169-181), which is often used by process engineers for hydrogen and hydrocarbon mixtures. (1) Tamouza, S.; Passarello, J.-P.; Tobaly, P.; de Hemptinne, J.-C. Fluid Phase Equilib. 2005, 228-229, 409-419. (2) NguyenHuynh, D.; Passarello, J.-P.; Tobaly, P.; de Hemptinne, J.-C. Fluid Phase Equilib. 2008, 264, 62-75. (3) NguyenHuynh, D.; Falaix, A.; Passarello, J.-P.; Tobaly, P.; de Hemptinne, J.-C. Fluid Phase Equilib. 2008, 264, 184-200. (4) NguyenHuynh, D.; Benamira, M.; Passarello, J.-P.; Tobaly, P.; de Hemptinne, J.-C.
Industrial & Engineering Chemistry Research, 2008
A group-contribution statistical associating fluid theory equation of state (GC-SAFT EOS) that was proposed by Tamouza et al. [Tamouza et al. Fluid Phase Equilib. 2004, 222-223, 67-76], which was extended in the first part in this series of papers to the asymmetric systems CO 2 + n-alkane, methane + n-alkane, and ethane + n-alkane, is further tested here on binary mixtures that contain aromatic hydrocarbons, n-alkanes, CO 2 , N 2 , and H 2 S. The method for correlating the binary interaction parameters (k ij ), which is inspired by London's theory of dispersive interactions, uses only pure compound adjustable parameters ("pseudo-ionization energies" of compounds i and j, denoted as J i and J j ). A group contribution for the latter parameters also is used for n-alkane and alkyl benzene series. Numerous prediction tests on the aforementioned cited systems were performed in a systematic and comprehensive way. Predictions are both qualitatively and quantitatively satisfactory, within deviations (4%-5%) that are comparable to those obtained on previously investigated systems (n-alkane + n-alkane, n-alkane + aromatic, n-alkane + n-alkanol).