Info

*Gas rate should not exceed 85 percent of the rate which gives a pressure drop of 1.6 in water per foot of packing based on Fig. 8.19a or b.

'Refers to chlorine gas drying using sulfuric acid. The reason for the low pressure drop is that chlorine is dried in a number of towers arranged in series. The overall pressure drop desired for the system is low.

'Refers to SO, absorption in sulfuric acid production. Reason for low pressure drop is to avoid entrainment of acid out of the absorber while allowing for some pressure drop rise due to accumulation of sulfonation products and packing chips in the packed bed.

^Presumably nonfoaminf services.

'"Examples include HF, HC1, ammonia, chlorine, sulfur dioxide scrubbers.

*Gas rate should not exceed 85 percent of the rate which gives a pressure drop of 1.6 in water per foot of packing based on Fig. 8.19a or b.

'Refers to chlorine gas drying using sulfuric acid. The reason for the low pressure drop is that chlorine is dried in a number of towers arranged in series. The overall pressure drop desired for the system is low.

'Refers to SO, absorption in sulfuric acid production. Reason for low pressure drop is to avoid entrainment of acid out of the absorber while allowing for some pressure drop rise due to accumulation of sulfonation products and packing chips in the packed bed.

^Presumably nonfoaminf services.

'"Examples include HF, HC1, ammonia, chlorine, sulfur dioxide scrubbers.

Which criteria to use? Some designers (5,15,57,97) abandoned the flood-point criterion in favor of the MOC. The majority (3,17,41,50, 51,55,56,58-60,96) prefer the flood-point criterion. Neither of these criteria is free of limitations (Sees. 8.2.3 and 8.2.4), but those associated with the MOC are far more restrictive. In addition, the ground rule of designing to 70 to 80 percent of flood is much better established than the equivalent rule for MOC.

The maximum pressure drop criterion is used jointly with the flood-point criterion. The column is designed to the more conservative of the two criteria. If MOC is preferred to flood point, the maximum pressure drop criterion is used jointly with the MOC criterion, and the column is designed for the more conservative of the two.

8.2.13 Average pressure drop

For computer calculations, a packed bed can be divided into several intervals. The bed pressure drop is the sum of all the interval pressure drops. Alternatively, the specific pressure drop can be calculated at the top of the bed and at the bottom of the bed. The average specific pressure drop is then calculated from (15)

Instead of using Eq. (8.30), the specific pressure drop is sometimes taken as the arithmetic average of APtop and AP^. This gives a slightly conservative estimate of the average pressure drop (15).

8.2.14 Liquid holdup

Liquid holdup is the liquid present in the void spaces of the packing. At flooding, essentially all the voids are filled with liquid or froth. Reasonable liquid holdup is necessary for good mass transfer and efficient tower operation, but beyond that, it should be kept low. High holdup increases column pressure drop, the weight of the packing, the support load at the bottom of the packing and tower, and the column drainage time. Most important, when distilling thermally unstable materials, high holdup may lead to excessive product degradation and fouling.

Static holdup is liquid remaining on the packing after it has been fully wetted and drained for a long time. The contribution of static holdup to mass transfer rates is limited (99). Static holdup can be estimated using the relationship of Shulman et al. (100), as recommended (14). Shulman's correlation was derived during the first generation of random packing, but the author is not aware of any updated alternatives.

Operating holdup is the liquid on the packing attributed to dynamic operation and is defined as the difference between total holdup and static holdup (101). Operating holdup contributes to mass transfer, as it provides residence time. Operating holdup can be estimated using Buchanan's correlation (101), as recommended (14). More recent correlations by Billet and Schultes (81), by Madkowiak (736,92a), and by Mersmann and Deixler (926) apply to second- and third-generation random packings as well as some structured packings.

Mja£kowiak (736) evaluated liquid holdup predictions from several recent correlations. His evaluation selected a simplified version of the Mersmann and Deixler correlation over alternative methods (92a), and demonstrated that it fitted experimental holdup data to within ±20 to 25 percent. This correlation has a sound theoretical basis and can be expressed in a dimensionless form. It has been extensively tested for random packings, but the author has no information on how it works for structured packings. The simplified version of the Mersmann and Deixler correlation (736, 92a) is hL

Was this article helpful?

0 0

Post a comment