Page 244 - Packed bed columns for absorption, desorption, rectification and direct heat transfer

234

J0
3S
Pe L = 0.66Re° L Gaf . (133)
The average arithmetic error of this equation is 6.9% and the average
square error is 11,8%. For determination of the experimental constants in Eq.
(133), besides all data of Elenkov and Kolev [120] in Table 17 after taking into
account the influence of the measuring cell, the data of Dilman [121], Carlton
[122] and Sater and Levenspiel [118] in the same table, and also the data of
Richter [126] for Raschig and Pall rings 35 and 50 mm are processed. All data
are obtained either using radioactive tracer, when the effect of the measuring
cell on the results is to be neglected, or taking into account this effect.
Dunn et al, [202] developed empirical Peclet number correlation not
only for the gas phase axial mixing but also for the axial mixing in the liquid
phase. It is found that the liquid Peclet number is proportional to the liquid
superficial velocity and it is independent of the gas velocity. This independency
is taken a priori by other investigators in this field for the area under the loading
point, when the forces between the gas and the liquid phase can be neglected.
Farid and Gunn [213] measured axial and radial mixing coefficients of
the liquid under and over the loading point. They carried out their investigations
in the bulk of the packing and near the column wall. It is found that at a small
ratio of the diameters of the column and of the packing elements, this ratio
influences the axial mixing coefficient. With increasing of this ratio the effect of
the wall is to be neglected. The authors [213] assume that it is possible to
describe the dispersion by an overall dispersion coefficient that includes the
quantitative effect of local dispersion and wall flow. We think that such a
position is not perspective, at least from a mathematical point of view. It is quite
better not to complicate the mathematical model of the mass transfer process by
tacking into account the differences between the bulk of the packing and the
wall zone, but to try to eliminate these differences. The problem is discussed in
details in Chapter 8.
Macias-Salinas and Fair [204] investigated the axial mixing in the
liquid phase for 25.4 mm ceramic Raschig rings and for metal Pall rings with
the same size, together with the structured packings Sulzer BX and Flexipac 2.
The obtained date for Bo L [204] are presented in Fig. 34.
Using these data they obtained the equation:

55 ll3 2 u
Bo L = 24ARel *Gal (d pay - , (134)

where the packing height in the Bodenstein number is the test length between
the sensors.