Page 215 - Packed bed columns for absorption, desorption, rectification and direct heat transfer

205

(88
»

The equation describes with an average error of 6% the data from
different investigations of random and arranged packings of rings, saddles,
spheres, and hurdle packing from glass, ceramic, steel, and wood with sizes
from 12,7 to 50 mm. The given accuracy is determined when the pressure drop
is obtained experimentally.
Because of quite limited importance, the co-current flow is less
investigated than the countercurrent one. In this area the experiments of
Wammes, Mechielsen and Westerterp [197] could be mentioned. They are
carried out at pressure up to 6.0 MPa and gas superficial velocity up to 5.5 cm/s.
The liquid viscosity is changed using water, ethanol and 40% water-
ehtyleneglyeol solution. As a gas phase pure nitrogen is used. At gas velocity
equal to zero, the increasing of the pressure does not influence the liquid
holdup. The results for the dynamic holdup without gas flow are described by
an equation containing Re L and Gai numbers. It is found that for Re L<l 1 the
36 5S
holdup is proportional to Rei®' and for Mei»15 -to Rei°" . As it is to be
expected for the co-current flow, the increasing of the gas velocity leads to
reduction of the dynamic holdup. The experiments show also that in the case of
two phase flow operation at elevated pressures, the holdup decreases at
relatively low gas velocity.
Billet and Schultes [314] developed a physical model for prediction of
liquid holdup in two-phase countercurrent columns. The model is valid for
random and structured packings and requires an experimental constant
depending on the packing type and dimensions. For calculation of the total
liquid holdup at the flooding point H tF, the same authors [316] offered the
equation;

——*% > (89)
Pi w 0

where w^ is the value of w§ at the flooding point, m/s;
The equation is valid for

- 3<H tF<E.