Model Bolles-Fair

Several methods have been proposed for predicting values of // and H as functions of system, flow conditions, and packing type. The model of Wagner et al. is directed specifically to the newer, open-style random packings. The Bolles-Fair model has the broadest validation and will be used here and is summarized by the following equations ... 

The Onda-Bravo approach has been found especially reliable for absorbers and strippers operating with lean mixtures, as is usually the case. For more concentrated cases (inlet gas > 5 vol-% solute for absorbers and inlet liquid > 5 vol-% solute for strippers), the Bolles/Fair model is more appropriate. 

In the Bolles-Fair model, the height of a gas-phase transfer unit is given by... 

This model was found to give a slightly better fit of the distillation data than did the Bolles-Fair model. It is somewhat more cumbersome to use, however, because of the need to evaluate individual-phase transfer coefficients. 

FIGURE 6.4-2 Packing parameters for the Bolles-Fair liquid-phase mass transfer model. (From Holies and Fair,4 excerpted by spacial petraissiou from Chemical Engineering, July [2, 1962 Copyright 1962 hy McGraw-Hill, New York.)... 

We will use the correlation of Bolles and Fair (1982), for which HTUs are defined in the same way as here. The Bolles-Fair correlation is based on the previous correlation of Cornell et al., (1960a, b) and a data bank of 545 observations and includes distillation, absorption, and stripping. This model and variations on it remain in common use fWang et al.. 2Q05V... 

The Bravo-Fair model for random packings involves the prediction of an effective interfacial area. The same data bank published by Bolles and Fair was used in conjunction with the individual-phase mass transfer correlation of Onda et al. to deduce values of an effective area for transfer. The final equation for area is... 

The improved Bolles and Fair model still had a standard deviation of almost 25% for the calculated values compared to the experimental data therefore. Porter and Jenkins conducted an extensive review of the data bank from which that model was developed [26]. They eliminated the runs made at very high and very low vapor rates and confined the data base only to distillation runs. Porter and Jenkins then found that this improved model needed only about a 20% safety factor to give the calculated design HETP value a 95% confidence limit for prediction of the residual data base. 

Bolles and Fair [129] present an analysis of considerable data in developing a mass-transfer model for packed tower design however, there is too much detail to present here. 

The summary of HETP values of Vital [142] for various types and sizes of packings are believed to be referenced to typical industrial distributors for the liquid. This variation can influence the value of HETP in any tabulation the effect of distributor design is discussed in an earlier section of this chapter. Porter and Jenkins [143] developed a model to improve the earlier models of Bolles and Fair from about 25% deviation to about a 95% confidence using a 20% factor of safety [139]. 

Bolles, W. L. and Fair, J. R. (1982) Chem. Eng., NY 89 (July 12) 109. Improved mass transfer model enhances packed-column design. 

In the absence of experimental data, mass transfer coefficients (and hence heights of transfer units) can be estimated by generalized models. A popular and easy to use correlation for random packings is that of Bolles and Fair (1982). The earlier correlations of Onda et al. (1968) and Bolles and Fair are also useful for random packings. 

Bravo and Fair (122) statistically analyzed the reliability of their model. They concluded that multiplying an HETP calculated from their model by a safety factor of 1.6 will give 95 percent confidence that the column is not too short This factor is slightly lower than the Bolles and Fair (55,96) and the Onds et al. (123) correlations. MacDougall (58) repeated the statistical analysis after he rearranged the data bank, and showed that a safety factor of 1.3 is more appropriate, making the Bravo and Fair correlation much better than the others. 

Bolles, W. L., and Fair,J. R., Improved Mass-Transfer Model Enhances Packed-Column Design, Chem. Eng. V. 89, No. 14 (1982) p. 109. 

The model of Bolles and Fair [Equations (12.104) through (12.106)] is appropriate. Input data for the model are as follows ... 

---