United States
Environmental Protection
Agency
Air and Energy Engineering
Research Laboratory
Research Triangle Park, NC 27711
Research and Development
EPA/600/S2-88/047 Dec. 1988
v>EPA Project Summary
Rate Controlling Processes and
Enhancement Strategies in
Humidification for Duct SO2
Capture
D. K. Moyeda, G. H. Newton, J. F. La Fond, R. Payne, and J. C. Kramlich
The fundamental rate processes
that govern sulfur capture in power
plant ducts during humidlfication of
flue gases were Investigated. The
specific application was the
reactivation of partially sulfated
calcium-based sorbents from In-
furnace injection. The results
suggest that physical contacting
between the spray water and the
sorbent particles is necessary to
achieve significant reaction rates.
Several means of promoting such
contacting were investigated and a
general approach to contacting was
proposed. These hypothetical
predictions were tested in a subscale
rig using laser-based measure-
ments. The reactivity of slurry drops
was investigated in a dilute-phase
reactor. The results indicate that
calcium availability (I.e., dissolution
into the liquid) and droplet lifetime
were the principal constraints on
sulfur capture. Increased
concentrations of hydrate in the
slurry droplets reduced the fractional
conversion of sorbent to product.
This was unexpected since calcium
dissolution rate control would imply
that conversion is independent of
slurry concentrations. Also, the
internal structure of the hydrate
appears to contribute to the calcium
availability. This suggests that
approaches which seek to develop
high specific surface areas for the
sorbent within the slurry droplets will
enhance sulfur capture.
This Project Summary was
developed by EPA's Air and Energy
Engineering Research Laboratory,
Research Triangle Park, NC, to
announce key findings of the research
project that Is fully documented in a
separate report of the same title (see
Project Report ordering information at
back.)
Introduction
The use of calcium based materials as
in-furnace sorbents for sulfur dioxide
(SOg) has received considerable recent
research attention as a low-cost
approach to intermediate levels of SOg
control. Two ways to enhance the
attractiveness of the approach are to (1)
find an inexpensive way to improve
overall sorbent conversion to product,
and (2) develop an option for improving
electrostatic precipitator (ESP) per-
formance in collecting the furnace
injected solids. One approach to both
problems is the adiabatic humidification
of the duct immediately downstream of
the air heater to a nearly saturated
condition. Previous testing has shown
that additional sulfur capture is possible
within this region. Also, humidification
reduces both duct temperature
(increases the ESP residence time) and
ash resistivity; both of these factors
improve ESP performance.
To implement humidification, a
number of design decisions must be
made. One of these includes the amount
of humidification water and its method of
introduction. The ideal situation is one
-------
that improves performance, does not
cause negative operating effects, and
optimizes duct sulfur capture. To design
for this condition, some indication of the
processes that control the sulfur capture
is needed. The work reported here
defines the rate limiting steps that control
the capture of SOa in the duct
humidification system.
Most of the testing to date has been at
pilot-scale. Test results show that,
unless the sorbent particle is coated with
bulk moisture, the reactivity is too low to
allow significant reaction in duct
residence times. Three mechanisms by
which this physical wetting can occur
have been identified:
1. Preslurring and duct injection of the
sorbent.
2. Spraying of water into the duct,
resulting in free stream inertial
impaction of the water droplet on the
sorbent.
3. Condensation, in which liquid
collects on the surface of a cool (i.e.,
ambient temperature) particle
injected into a warm, prehumidified
gas stream.
Mechanism 2 is the only likely candidate
for duct-reactivation of the furnace
injected sorbent; condensation is not
likely because the sorbent is never
colder than the surrounding gas. Thus,
the only sorbent that is likely to be
reactive in dispersed phase is that
portion that directly collides with water
droplets. Nozzle design should include a
provision, if possible, for optimizing
sorbent scavenging by water droplets via
inertial impaction.
Once in the slurry phase, the
reaction rate can be expected to be
governed by one of the following
processes:
The rate at which SC>2 is
transported from the free stream to
the surface of the droplet.
The rate at which solid calcium
becomes available in the liquid
phase.
The aqueous reaction rate,
including ionization and direct
product formation.
The experiments were designed to
evaluate the contribution of each of these
fundamental steps to the overall rate, and
to identify candidates for enhancement.
The work was divided between
enhancement of (1) scavenging and (2)
reactivity.
Scavenging
The scavenging experiments were
conducted in a facility in which a flowing
suspension of hydrate in air passed a
subscale humidification nozzle (see
Figure 1). The absorption of the sorbent
into the water droplets was measured by
a unique laser extinction technique that
was developed as part of this program.
The results were correlated by a one-
dimensional model of the scavenging
and reactivity process developed in-
house.
Figure 2 shows typical results for an
air-blast atomizer. The results show that
the percent of the sorbent scavenged
from the duct increases almost linearly
with the nozzle water flow. The one-
dimensional model results, also shown
on the figure, are in close quantitative
agreement. Figure 3 uses the model to
show that the atomizer parameters that
most strongly favor scavenging are (1)
large droplet diameter, and (2) high
droplet velocity. Unfortunately, these
considerations run opposite engineering
requirements for fitting sprays into
confined spaces (i.e., fine droplets and
long residence times to avoid wall
wetting).
An investigation of the scavenging
process showed that most of the
scavenging takes place in the near field
spray, near the nozzle. This "high
efficiency zone" can be viewed as a
volume fixed in space immediately
downstream of the nozzle. One obvious
enhancement strategy is to focus all of
the sorbent-laden duct flow though this
zone. This, however, does not result in
any enhancement because the additional
sorbent flux is counterbalanced by the
reduced residence time available for
scavenging. Thus, no net improvement
occurs. Also, nozzles which entrain large
amounts of surrounding gas will not
improve scavenging because, again, the
additional flux will be balanced by
reduced residence time in the high
efficiency zone. Another key point is that
attempts to "one-dimensionalize" the
flow by, for example, using many small
nozzles, will not significantly improve
scavenging. Basically, this is because
doubling the volume of the high
efficiency zone halves the scavenging
rate per unit volume. Thus, the overall
integrated result is constant.
The key to enhancing scavenging is to
direct the sorbent, but not the gas flow,
into the near field of the nozzle. Figure 4
shows that the scavenging was
significantly enhanced when the sorbent
was introduced near the nozzle, rather
than mixing throughout the test duct.
This can be practically effected by: (1)
introducing sorbent near the nozzles
(this, of course, is not an option for
furnace sorbent reactivation), and (2)
separating the sorbent from the gas
and concentrating it about the no.
Because of the small size of the sort
aerodynamic separation is not a li
candidate; however, electrost
concentration is a possibility.
Reactivity
The slurry droplet reacti'
experiments were conducted in a d
phase plug-flow reactor. As illustrate
Figure 5, the slurry was atomized t
rotating disk, and a small fraction of
droplets produced were admitted thrc
a slot into the reacting flow. The drof
were collected for the desired amour
time by a heated cup probe, and
reaction extent was determined
chemical analysis. One note
importance is that the exact condit
and design of the sampling cup mus
carefully controlled to prevent signifii
probe capture.
The results indicated that at low !
concentrations the overall slurry reac
was limited by external diffusion of 5
to the droplet surface. At high J
concentrations, some form of inte
control over calcium availability \
evident. Figure 6 shows this shifl
controlling mechanism at the point wr
utilization levels off with increasing J
concentration. One key point is that
utilization of the sorbent is reduced
higher slurry concentrations, indical
that incremental sorbent reactivity
reduced as the solids concentration
the slurry increases. The practi
consequence of this is that enhancem
of scavenging will not lead to
proportional enhancement of sul
capture because of the decreas
specific sorbent reactivity. It is interesl
to note that similar overall behavior
observed in spray dryers. Here,
reduced reactivity is manifested a;
weak increase in capture when Ca/S
increased above 1.0.
The importance of internal cont
under realistic conditions led to i
investigation of sorbent parameters wh
might be expected to influence 1
possible rate controlling steps. The f
was the importance of the exter
surface area of the sorbent, which v
varied by comparing an atmosphe
hydrate with a pressure hydrate. For
materials used, the difference in exter
surface area was about a factor of 5.
shown in Figure 7, the pressure hydr
displayed essentially identical reactiv
Thus, external surface area is r
indicated to be a controlling factor. Figi
7 also shows a test of the importance
internal surface area. The normal calc
-------
Nozzle g
Water
Main Duct
Air mi
Nozzle
Air
(For Twin-Fluid
Nozzle Only)
Sorbent
Feeder
Transport
Air
Mechanical or
Twin-Fluid
Nozzle
Laser
Transmitter
Figure 1.
t
Exhaust
The scavenging rig.
Computer
hydrate was compared with an alcohol
hydrate (i.e., hydrates prepared in an
alcohol/water solution, which develop
high internal surface areas). The figure
shows that the conversion is similar in
the external control regime (less than
500 ppm SOa), but the high surface area
sorbents excel at higher 802 values. The
ionclusion is that the higher surface area
provides additional calcium availability,
which essentially allows the reaction to
remain under 862 diffusion control to
higher 802 concentrations. In other
words, the "knee" in the curve is moved
to higher SC>2 concentrations, and higher
utilization values. For practical duct
humidification, this suggests that, if a
way could be found to form high surface
area materials during the in situ
hydration, higher utilizations would result.
Conclusions
The conclusions of this study include:
Some way to physically wet the
sorbent appears to be necessary to
achieve sulfur capture rates of
practical interest for duct
applications. The principal way to
do this in duct humidification
-------
40
30
I»
I
"5
10
Model Predictions
O Test Results
2 3
Water Flow (gphj
I
01234 5
Water Flow (cc/s)
Figure 2. Scavenging results for the air-blast nozzle compared to one-dimensional model
predictions.
0 20 40 60 80 100 120 140 160 180 200
Drop Velocity fm/s)
Figure 3. Predicted scavenging performance as a function of drop velocity and diameter
(monosized spray)
-------
R. 30 -
to
o
to
Concentrated
Sorbent Field
2 3
Water Flow Igph)
0723
Water Flow (cc/s)
Figure 4. Increased scavenging by sorbent concentration.
appears to be sorbent scavenging
by inertia! impaction.
Scavenging can best be enhanced
by focusing the sorbent, but not the
flow field, onto the region about the
humidification nozzle.
Under practical conditions, calcium
utilization appears to be limited
mainly by calcium availability within
the slurry drop, and by the lifetime
of the drop before evaporation to
dry ness.
Calcium availability is enhanced by
high internal surface areas for the
hydrate.
-------
Rotary Disk
Atomizer
Skimmer
Cone
Slurry
Feed
Inlet
Gas
Flow
jcouple B^
)ted
ogen ^ ^
ffe
p
i
^B
1*
fr ' ' ' '' f» L
^->^ Reactor
-X*^* (Insulatio
*^ Removed
S~\ Hot
' Purge
Probe
1 1 f I 1 ff
If
Outlet
Figure 5. Single-drop reactor.
-------
Dolomitic
120
100
,0
to
-------
D. K. Moyeda, G. H. Newton, J. F. La Fond, R. Payne, and J. C. Kramlich are
with Energy and Environmental Research Corp., Irvine, CA 92718-2798.
Brian K. Gullett is the EPA Project Officer (see below).
The complete report, entitled "Rate Controlling Processes and Enhancement
Strategies in Humidification for Duct SC-2 Capture," (Order No. PB 88-
245 975/AS; Cost: $79.95, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Air and Energy Engineering Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S2-88/047
0000329 PS
------- |