UNIFAC approach

UNIFAC Approach Jensen et al. [16] have employed the UNIFAC group contribution approach to develop an estimation method for pure-component vapor pressures. The model developed applies to hydrocarbons, alcohols, ketones, acids, and chloroalkanes of less than 500 molecular mass and in the vapor pressure region between 10 and 2000 mmHg. Burkhard et al. [8] extended this model to chlorinated aromatic compounds such as chlorobenzenes and PCBs. 

The vapor pressure of a liquid increases with increasing temperature. Reviews on and discussion of different types of vapor pressure-temperature functions can be found in the literature [17-20]. The most common representation of vapor pressure-temperature data for a pressure interval of about 10 to 1500 mmHg [1] is the three-parametric Antoine equation  

---

The inaccuracies of the approach lead the present writer to disagree with the author s conclusions on the UNIFAC approach, which according to him is better than the one of comparing the mixture with an ideal mixture. 

The application of thermodynamic models to the correlation and prediction of pharmaceutical solubility behaviour is an underutilized technique in today s process research and development environment. This is due to the relatively poor accuracy and limited predictive ability of the previous generation of models. Recent advances in computational chemistry and an increased focus on the life science sectors has led to the development of more appropriate models with significantly improved predictive capabilities. The NRTL-SAC and Local UNIFAC approaches will be discussed here with additional examples given in section 8. 

The basis of the UNIFAC approach is the definition of submolecular groups (e.g., CH2, CH3, CH3O, CH2CI, OH) and the fitting of a given molecular property or activity coefficient to a sum of contributions based on the subgroup molecular volume and interaction terms between the groups. 

The importance of a suitable G.C. method for evaluating the pure solid properties was evidenced. The limitations of the different G.C. methods are due to the limited experimental data available in the literature for heavy multifunctional compounds. The PR EOS with classical mixing rules gives the same results than the most complex UNIFAC approach. The importance of the sublimation pressure for correlating solubility data was underlined. 

With the treatment of gases as individual groups, some binary (or multicomponent) gas-liquid mixtures are reduced to mixtures of only two groups. For example, the carbon dioxide and methanol mixture considered at the conclusion of this section is actually a molecular mixture because both molecules are treated as groups by the UNIFAC approach, Similarly, mixtures of carbon dioxide with benzene or with paraffinic hydrocarbon liquids contain only two groups. The results for such systems are remarkably successful, as will be discussed in this section. The description of mixtures with more than two groups is possible for some of the present models, and the results look promising (Apostolou et al. 1995). 

Varions modifications and extensions of the classical UNIFAC approach to polymers have been proposed. All these approaches attempt to inclnde the FV effects, which are neglected in the... 

Perhaps the most important term in Eq. (5.2-3) is the liquid-phase activity coefficient, and mathods for its prediction have been developed in many forms and by many workers. For binery systems die Van Laar [Eq. (1.4-18)]. Wilson [Eq. (1.4-23)]. NRTL (Eq. (1.4-27)], and UNIQUAC [Eq. (1.4-3 )] relationships are useful for predicting liquid-phase nonidealities, but they require some experimental data. When no data are available, and an approximate nonideality correction will suffice, the UNiFAC approach Eq-(1.4-31)], which utilizes functional group contributions, may be used. For special cases Involving regular solutions (no excess entropy of mixing), the Scatchard-Hiidebmod mathod provides liquid-phase activity coefficients based on easily obtained pane-component properties. 

If experimental activity coefficients are unavailable, they can be estimated by, for example, the UNIFAC approach (Fredenslund, Gmehling and Rasmussen, 1977). The UNIFAC model is based on the group contribution concept, calculating the activity coefficients from two parts. The combinato-... 

Table 13.2 Group interaction parameters a (K) for the UNIFAC approach in the Maurer model [6].

The GC concept is not new it has been widely implemented for example in the calculation of pure component properties (an overview is provided in ref 474), for the estimation of activity coefficients of liquids in the successful UNIFAC approach," and more recently, in equations of state." The... 

These are promising approaches, which incorporate a detailed molecular model in which groups can be differentiated. In this way, they benefit from advantages of the successful UNIFAC approach, and overcome the difficulties associated with its underlying lattice model. They are accurate over large pressure ranges and can be used consistently for liquid and vapour phases. In addition, as mentioned above, their formulation as continuum fluid theories means that the binary interaction parameters can be determined from pure component data. 

Octano/—Water Partition Coefficient. The Fragment approach (234—236) has been reviewed (227) and another method similar to the UNIFAC refit for Henry s constant has been proposed. Improved accuracy for many species and the abiUty to correct for temperature effects have been claimed for the newer method. 

One problem limiting the consideration of salt extractive distillation is the fact that the performance and solubility of a salt in a particiilar system is difficult to predict without experimental data. Some recent advances have been made in modeling the X T.E behavior of organic-aqueous-salt solutions using modified UNIFAC, NRTL, UNIQUAC, and other approaches [Kumar, Sep. Sci. Tech., 28(1), 799 (1993)]. 

Solubility modelling with activity coefficient methods is an under-utilized tool in the pharmaceutical sector. Within the last few years there have been several new developments that have increased the capabilities of these techniques. The NRTL-SAC model is a flexible new addition to the predictive armory and new software that facilitates local fitting of UNIFAC groups for Pharmaceutical molecules offers an interesting alternative. Quantum chemistry approaches like COSMO-RS [25] and COSMO-SAC [26] may allow realistic ab-initio calculations to be performed, although computational requirements are still restrictive in many corporate environments. Solubility modelling has an important role to play in the efficient development and fundamental understanding of pharmaceutical crystallization processes. The application of these methods to industrially relevant problems, and the development of new... 

The interaction between solvents is important. For example, the development of a successful crystallization process for purification and isolation of an organic compound requires the selection of a suitable solvent or solvent mixture to date, no logical method has been estabhshed for determining the best solvent combination. The process chemist or engineer often employs a trial-and-error procedure to identify an appropriate solvent system, the success of which is dependent on experience and intuition. One approach utilizes a group-contribution method (UNIFAC) to predict a... 

There are several theoretical models to estimate the solubility of a solute in a solvent. However, use of dielectric constant is one of the oldest and simplest approach and is very popular with the formulators. Fractional method to estimate the dielectric constant is the simplest approach and is not the most accurate. However, it offers a good starting point for the estimation. In addition, the solubility of a solute is dependent on the dielectric constant of a solvent mixture and not to the particular composition. Other approaches, such as solubility parameter method and UNIFAC group theory contributions are less frequently used by industry formulators. 

We recommended two of these methods for general use, estimation from octanol-water partition coefficient and a group contribution method named AQUAFAC. Three other methods are also valuable under certain circumstances, the connectivity, UNIFAC, and sol-vatochromic approaches. 

Another group contribution method that has been applied to the prediction of soil sorption is the UNIquac Functional-group Activity Coefficient (UNIFAC, where UNIQUAC = Universal Quasichemical) approach (Fredenslund et al., 1977). Ames and Grulke (1995) applied the method to a small diverse set of chemicals, with rather poor results. They did not report any correlations, but from their results it can be shown that the correlation of observed and predicted log values using the Bondi method was n = 17, R2 = 0.571, 5 = 0.524, and F = 20.0 eight chemicals were predicted with an error of < 0.5 log units, 7 chemicals were predicted with an error between 0.5 and 1.0 log units, and 2 chemicals were predicted with an error of > 1.0 log units. 

Jefferson Tester I would like to shift gears and direct a question to John Prausnitz regarding his comments. You didn t talk very much about some of your own contributions and those of your students, for instance, the NRTL equation and UNIQUAC-UNIFAC models for nonideal solutions now in widespread use throughout the chemical industry and certainly employed by many people making practical calculations. There have been extensions of that local composition approach, in particular to electrolyte systems, by C. C. Chen and others. I wonder how you personally feel about that and... 

The compositions of the vapor and liquid phases in equilibrium for partially miscible systems are calculated in the same way as for miscible systems. In the regions where a single liquid is in equilibrium with its vapor, the general nature of Fig. 13.17 is not different in any essential way from that of Fig. I2.9< Since limited miscibility implies highly nonideal behavior, any general assumption of liquid-phase ideality is excluded. Even a combination of Henry s law, valid for a species at infinite dilution, and Raoult s law, valid for a species as it approaches purity, is not very useful, because each approximates real behavior only for a very small composition range. Thus GE is large, and its composition dependence is often not adequately represented by simple equations. However, the UNIFAC method (App. D) is suitable for estimation of activity coefficients. 

Solubility data of biological compounds taken from literature are considered in this work. Different thermodynamic models based on cubic equations of state and UNIFAC are used in the correlation of experimental data. Interaction parameters are obtained by group contribution approach in order to establish correlations suitable for the prediction of the solid solubility. 

The Oishi-Prausnitz model cannot be defined strictly as a lattice model. The combinatorial and residual terms in the original UNIFAC and UNIQUAC models can be derived from lattice statistics arguments similar to those used in deriving the other models discussed in this section. On the other hand, the free volume contribution to the Oishi-Prausnitz model is derived from the Flory equation of state discussed in the next section. Thus, the Oishi-Prausnitz model is a hybrid of the lattice-fluid and free volume approaches. 

---