Hydrocarbon Processing - August 2021 - GP-27

ENGINEERING AND DESIGN
form a fully closed loop like an amine regenerator does, so
maintaining a water balance is unnecessary. This means that
direct steam injection can be used as a stripping agent, either
alone or in conjunction with a conventional reboiler, and the
surplus water added by the condensate is simply added to the
plant's water inventory. Typical energy usage in the stripping
process is in the range of 1 lb-1.5 lb of 50-psig-equivalent saturated
steam per gallon of sour water.
Study findings and results. This assessment was completed
by systematic investigation combining simulation with hydraulic
analysis and actual field testing. Multiple simulation cases
were performed, using a mass transfer model in a proprietary
simulator to assess the maximum sour water processing capacity
of the SWS system.
As per general guidelines, SWS calculations have been performed,
using ideal stages in the Equilibrium Model.4
Ideal
stage calculations provide a fast and accurate means to quantify
the trade-off between the number of trays and the stripping
steam. In this case, the real number of trays and lengths
of packed bed were used since this is an existing column; this
gives the Mass Transfer Model.
In the Mass Transfer Model, the actual rates of mass and heat
transfer between the phases in the column are used. The mass
and heat transfer coefficients and interfacial area are based on
vapor and liquid diffusivities, viscosities, densities, heat capacities
and thermal conductivities. To estimate mass transfer coefficients,
detailed information about the column internals is
required. This information includes column diameter, surface
area per unit volume and void fraction for packing. FIG. 3 shows
the model for the SWS.
A Peng-Robinson equation of state was used, as only H2
S is
present in the sour feedwater. For systems that contain only a
weak acid or a weak base, such as the binary systems H2
or NH3-H2O, molecular models are sufficiently descriptive
S-H2
O
to model the system behavior. Examples of molecular models
are the cubic equations Peng-Robinson and Soave-Redlich-
Kwong, as well as any activity coefficient models that account
for only molecular species. This acid-only case is a special case,
and care must be taken when modeling a sour water system.
The equations are not capable of describing the behavior when
both an acid and a base are present. An electrolytic model is
needed to account for the ionic interactions of the dissociated
species in acid-base systems, and can be used for an acid-only
system, as well.
FIG. 44
and H2
shows the relation between the steam injection rate
S in treated water. It illustrates the effect of varying the
steam rate on the stripped water composition. The amount
of steam used has a major effect on the composition of the
stripped water. The H2
S concentration decreases rapidly as the
steam rate increases. NH3 concentration also initially decreases
rapidly until it is influenced by the acid-base chemistry of NH3
and HCN. The HCN concentration is initially insensitive to
steam rate and remains constant. Once the NH3
decreases sufficiently, both NH3
concentration
and HCN can be stripped.
However, if phenol ionizes in the water, then significant removal
by stripping is difficult.
The solutes to be removed are weak electrolytes that are
partially dissociated into ions in solution. Therefore, the phase
10
1
NH3
0.1
0.2
Steam rate, kg/l
H2S
FIG. 4. Steam injection rate and H2
HCN
S in treated water.
Gas Processing & LNG | JULY/AUGUST 2021 27
0.3
Phenol
0.4
equilibrium of each weak electrolyte in solution plays an important
role in the removal of pollutants from sour water through
steam stripping. The main feed components in this system are
H2
O and H2
S. Henry's law is applied to these components in
S and CO2
H2O ←→ H+
H2S ←→ H+
HS-
CO2
+ OH-
+ HS-
←→ H+ + S=
+ H2O ←→ H+
HCO3
NH3 + CO2
- ←→ H+ + CO3
+ HCO3
=
←→ NH2COO-
-
+ H+
(1)
(2)
(3)
(4)
(5)
(6)
Effect of stripper pressure. It is sometimes said that higher
stripper pressures favor sour water stripping. The experimental
data presented in Reference 3 and the gas plant actual field
tests show that higher pressures do not seem to favor ammonia
stripping, and they certainly have a negative effect on stripped
water quality in terms of H2
S.
TABLE 1 shows the effect of column head pressure on residual
ammonia and H2S levels in stripped water.3
lated performance corresponds to a steam rate of 25,000 lb/
hr of 50-psig steam (1.4 lb/gal), with the cross-exchanger
TABLE 1. Effect of stripper pressure on residual NH3
H2S levels in stripped water3
Stripper head
pressure, psig
10
15
20
25
10,000
1,000
100
Treated water
NH3
, ppmw
17.2
21.1
26
31.5
H2S, ppmw
0.00002
0.00013
0.00058
0.00197
and
The simuthe
simulation because they are light gases. The reactions that
occur when H2
any other aqueous solution. The reactions shown in Eqs. 1-6
are of interest for this study:3
dissolve in solution are the same as in
Stripped water concentration, ppmw
Hydrocarbon Processing - August 2021

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