Hydrocarbon Processing - March 2021 - 53

Process Optimization
the maximum amount of free-passage available, and is often in a
full cone spiral design (FIG. 9). While this design may have poorer distribution than some of the other distributors mentioned in
this article, it can be more fouling resistant.
Liquid distributor comparison. A simplified example
of characteristics of the liquid distributors described previously (based, in part, on information in literature) is shown in
TABLE 1.15 Specific vendor designs may vary but will generally
have these relative characteristics.
Packing height of an equivalent theoretical plate
(HETP). Packed tower design in SWSs will run into the same

issue encountered in trayed towers when considering the efficiency of the trays. In this case, the efficiency of the separation
in a packed tower is represented in some cases by the HETP.
Like tray efficiency, the HETP is a chemical engineering factor
that is not static from one separation to another or even from
one packed bed to another in the same tower.
Packed towers in SWS service generally use dumped (random) packing. HETP values for these types of packing are
available from many vendors. Published HETP values are not
specific to SWS operations. If general vendor-published HETP
values are used without considering the conditions that could be
present in a SWS, the system will not likely work correctly or for
long. The mass transfer limitations must be considered.
Rules of thumb for packing have been offered in discussions
at prior industry conferences, and the authors' company has
discussed these in conversations with several refinery subject
matter experts (SMEs). A rule of thumb for 2-in., second- or
third-generation packing is to use 2 ft of packing depth per actual tray. Given the previous rule of thumb for trays (three actual
trays per a theoretical stage in SWS service), and assuming a tray
spacing of 24 in., this rule of thumb makes the height of packing
the same as the height of trays that would have been present-
if trays had been chosen. If one assumes that third-generation
packing has similar efficiency to second-generation packing,
then the rule of thumb seems applicable to both generations.
However, the rule of thumb may be overly conservative. For example, one should nominally see an increase in capacity, or an
increase in efficiency, or possibly both, when going from sec-

ond-generation packing to third-generation packing. Also, better efficiency may be possible if the sour water is as clean as possible (i.e., if good feed preparation steps have been used) and if
the liquid distribution quality is as good as possible. One author
reported going from 2-in., second-generation packing to 1.5-in.,
third-generation packing, and achieving a significant improvement in efficiency in the same bed height and capacity.4
TABLE 2 shows HETP values from the literature and from
three actual operating SWSs. The data shows relatively good
agreement between the actual HETP and published HETP data
for Source 2 (0.7-0.9 ratio) and Source 3 (1.1-1.3 ratio). However, the actual/experienced HETP for Source 1 was more than
two times the published HETP. Several reasons may explain
this. Discussions with the author for Source 3 indicated that a
reasonably good distributor (in this case, meaning good balance
of liquid distribution and low fouling tendency) was used in the
stripper. Source 1 was known to have a poorer distributor type
(not one of those mentioned previously). Source 1 also had
other issues, including areas of plugged packing in sections of
the tower. Furthermore, conditioning of the sour water feed may
have been different between the sources-the literature discussing Source 3 mentions " minimal historical fouling and foaming
issues " 4, so it may have had a cleaner sour water feed stream.
TABLE 2 also shows the ratio of the SME-designed HETP to
the vendor/published HETP. This factor ranged from 1.8-2.4,
indicating that many operators choose to use a much larger
HETP, conservatively increasing the packing bed requirements
and the sizing of SWSs.
Packing recommendations. Overall, the following may be

useful when considering using packing in SWS service:
1. Distributors should seek to balance a fouling-resistant
design with good liquid distribution (adequate drip
point density); trough-style liquid distributors with

TABLE 1. Comparison of liquid distributors for SWSs
Parameter

Channel

Trough

Weir riser pan

Spray

Driving force

Gravity

Pressure

Typically
medium
to large

Typically
small

Any

Best

Lower

Low to
medium

Low

Low to
medium
(depending
on nozzle)

Must be installed
almost perfectly
level

Yes

Requires precise
nozzle aiming

Yes

Tower size

Liquid distribution
quality (of those
listed)
Propensity
to plug

FIG. 9. Sketch of a full cone spray nozzle with maximum free passage.
Hydrocarbon Processing | MARCH 2021

Hydrocarbon Processing - March 2021

Table of Contents for the Digital Edition of Hydrocarbon Processing - March 2021