&EFK
United States
Environmental Protection
Agency
Industrial Environmental Research,
Laboratory
Research Triangle Park NC 27711
Research and Development
EPA-600/S7-82-044 September 1982
Project Summary
Augmentation of Fine
Particle Collection in the
EPxP Scrubber
Shui-Chow Yung, Toai Le, Ronald Patterson, and Seymour Calvert
The EPxP (electrostatically aug-
mented particle by particle) dry scrub-
ber is analogous to a venturi scrubber
except that it uses relatively large solid
particles (instead of water drops) as
collection centers for the fine particles
in the gas stream. It is a novel device
for controlling fine particle emissions
at high temperatures and pressures.
Bench scale (0.5 and 1.1 AmVmin)
and pilot scale (4.8 AmVmin) experi-
ments have been run at temperatures
of 20-820°C to determine the per-
formance characteristics of the system.
Experimental results show that the
EPxP dry scrubber can operate at high
temperatures and its particle collection
efficiency can be increased by pre-
charging the particles and by polarizing
the solid collectors.
This report presents the system
design and experimental results.
This Project Summary was devel-
oped by EPA's Industrial Environ-
mental 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
Fine particle removal from gases at
high temperature and pressure en-
countered in fluidized-bed combustors
and various fuel conversion processes
places severe requirements on the gas
cleaning system. The environment of
reactive gas mixtures at temperatures
up to 1,100°C and pressures up to 15
atm can be withstood by only a few
structural materials. Particle collection
efficiency must be high to meet the
NSPS (new source performance stand-
ard) and turbine requirement if the gas
is to be expanded through a gas turbine.
Only a few control devices currently
being developed can operate at these
extreme temperatures and pressures.
A.P.T/S PxP (for particle collection by
particles) dry scrubbing system is one of
the devices. The PxP scrubber is
analogous to a venturi scrubber except
that relatively large solid particles
(instead of liquid drops) are used as fine
particle collectors. Several possible PxP
designs were evaluated in an earlier
study sponsored by EPA. It was con-
cluded that the design presented in
Figures 1 and 2 is feasible.
In this design, 12 PxPsare arranged in
four "delta" units for controlling the
emissions from a typical pressurized
fluidized-bed coal combustor. Each PxP
contactor has an angle of 60° or more to
the horizon. The solid collector granules
are injected at the contactor entrance
and are transported to a collector
separator section. The granules are
separated from the gas in this section
and are recycled through a vertical
downcomer to the next contactor. The
dense flow of the collector granules in
the downcomer and their rubbing
against each other cause them to
release their collected particles. A bleed
stream of gas from the downcomer
carries the small particles into a
-------
separate cleaning process at low tem-
perature and pressure. As a result the
collector granules are continuously
cleaned just before introduction into the
contactor.
The design (1) maximizes the flow of
collector by having a vertical down-
comer, (2) eliminates the requirement
for an external collector lifting system,
and (3) provides an integral method for
collector cleaning.
Dirty Gas
Manifolds
Since the PxP contactor is essentially
a venturi scrubber, particle collection is
achieved principally through the mech-
anism of inertia! impaction. Particle
penetration through the scrubber
increases dramatically as the particle
aerodynamic diameter decreases. Al-
though penetration can be decreased by
operating at higher collector flow rates,
greater power is needed to accelerate
the collectors to the gas velocity.
Clean Gas Manifold (Outlet)
Clean Gas Manifold
Figure 1. Delta arrangement of 12 PxPs for HTP cleaning.
Collectors
Purge Gas
to Cleanup-*——
Purge Gas
Bustle *"
Collector
^
i i
i i
i i
\ ' 1
! i
Distributor
i A
i A
V ^
/
/
/ Dirty Gas.
Clean Gas
Separator
Baffles
Collector
Downcomer
h-
Bustle
Refractory Lining
• with H. T. Resistant
Metal Liner Inside
Figure 2.
Assembly diagram of a PxP scrubber.
2
Performance of the PxP scrubber relies
on its ability to collect particles with
minimal reentrainment.
Adhesion of the particles to the
collector granule surface is a function of
the London-Van der Waals forces and
naturally occurring electrostatic attrac-
tive forces. Collision with the scrubber
wall and with other granules can
produce forces large enough to over-
come the attractive forces and dislodge
collected particles.
Electrostatic augmentation may be
used to overcome the above problems
and thereby improve the PxP scrubber
performance. The feasibility and per-
formance characteristics of the EPxP
(electrostatically augmented PxP) scrub-
ber were determined experimentally in
bench and pilot scale units under
ambient and high temperature condi-
tions. This report gives the results.
Preliminary Bench Scale
EPxP Study
To evaluate the feasibility of aug-
menting the PxP with electrostatic
forces, the EPxP system shown in
Figure 3 was designed and built. It was
made of Plexiglas which enables visual
observation of collector flow pattern. It
consisted of a blower, a particle
precharger, an EPxP contactor, a gravity
settler for collector separation, a
particle generator, a collector storage
hopper, and various flow and pressure
measuring instruments. The contactor
was inclined at an angle of 60° with the
horizon to allow vertical collector in-
jection.
Two EPxP throat configurations were
tested: the first had a round cross-sec-
tion, a diameter of 1.6 cm, and a length
of 5 cm (from collector injection point to
throat exit); and the other had a square
cross-section (1.6 cm on a side) and a
length of 28 cm.
There was no attempt to precharge
the collector granules in the bench
scale apparatus; however, the granules
did pick up some charges due to friction.
The collector granules could be polarized
with an external field. When using the
long square throat, the polarizer, which
consisted of two oppositely charged
parallel plates, was at the throat. When
using the round throat, the polarizer
was in the lift pipe after the contactor.
The apparatus was operated under
forced draft and ambient temperatures
and pressures. Test particles were
injected into the air stream upstream of
the particle charger. After the particle
charging section, the air and particles
-------
flowed into the EPxP where contact
with collector granules occurred.
Collector granules were regular
grade Florida zircon sand with a mass
median diameter of about 0.11 mm. The
granules were passed through the
system on an once-through basis.
From the contactor, the collectors
were lifted by the cleaned air up to a
gravity settler, an expanding duct with
sufficiently large cross-section to slow
the gas velocity to below the settling
velocity of the collectors. The air vented
into the atmosphere after the collector
separator.
Results
Methylene blue and fly ash particles
were used in the experiments. Particle
samples were taken at the inlet and
outlet of the EPxP simultaneously with
cascade impactors. The inlet sampling
point is after the particle precharger;
therefore, the measured EPxP collec-
tion did not include collection by the
particle charger.
The EPxP contactor was operated in
four modes: UP/NC (uncharged particle/
non-polarized collector), CP/NC (charged
particle/non-polarized collector), UP/
PC (uncharged particle/polarized col-
lector), and CP/PC (charged particle/
polarized collector).
Air Outlet
Collector
Hopper
Valve
Collector
Separator
Collector
Reservoir
<'gh Voltage
Supply
Blower
jLFIowmeter
Air Inlet
Figure 3. Charged particle, polar-
ized-collector bench*scale
experimental apparatus.
Note: "uncharged" and "non-polar-
ized" mean that the natural charges on
the particles and the collectors were not
neutralized. The collectors and particles
might actually carry charges. This is
especially true for the collectors since
they can pick up charges by friction.
Figures 4 and 5 show the measured
grade penetration curves for the two
contactors, operating at a pressure drop
of 38 cm (15 in.) W.C. The particle
collection efficiency improves by pre-
charging the particles and/or by polariz-
ing the collectors. Because it is longer,
the square throat shows a higher
collection efficiency than the round
throat.
The contactor shows a higher particle
penetration with fly ash particles,
probably due to particle bounce and
dirty collectors. The Florida Zircon sand,
used as received, contained a large
quantity of fine dust, which was
reentrained during collector injection.
In later experiments (hot bench scale
EPxP), the sand was cleaned by passing
through the system twice before taking
particle data.
High Temperature Bench
Scale Experiments
Preliminary experiments under am-
bient temperatures had shown that the
PxP scrubber could be augmented with
electrostatic force. To obtain design
information for the particle charger,
contactor, collector separator, and EPxP
performance under high temperature
conditions, the scrubber system shown
in Figure 6 was built. It was similar to
the ambient bench scale EPxP except
for a furnace to produce the hot gas and
a quencher to cool the gas before it was
vented.
1.0
0.5
0.1
0.05
!
0.01
0.005
= 47 m/s
UP/PC
CP/NC
QG
AP = 38 cm W.C.
Round Throat
di = 1.6cm
A = 5. / cm
Methylene Blue
Part.
0.1
0.5 1
Aerodynamic Panicle Diameter,
10
Figure 4. Experimental grade penetration curves of the round-throat Plexiglas
EPxP scrubber.
-------
1.0
0.5
0.1
c-
.o
5 0.05
0.01
0.005
UP/NC
Fly Ash
UP/NC
Methylene Blue
CP/NC
Methylene Blue
= 37.5 m/s
QG
AP = 38 cm W. C.
Square Throat
O.I
O.5 1
Aerodynamic Particle Diameter,
10
Figure 5. Experimental grade penetration curves for the square-throat Plexig/as
~PxP scrubber.
The system was designed for a
maximum volumetric gas flow rate of
1.1 AmVmin (38 acfm) and a maximum
collector-mass/gas-volume ratio of 8
kg/m3. The air was heated by burning
natural gas. Fly ash particles were
injected into the flame to deagglomer-
ate and mix the particles.
The particles next entered the particle
charger, made of machinable ceramic.
The charger was of the wire-and-plate
design.
The particle charger was followed by
the EPxP contactor, also made of
machinable ceramic. Two opposite
surfaces were lined with 309 stainless
steel plates so that a high potential
could be applied to one of the plates to
polarize the collector granules.
The collectors and the collected
particles were separated from the gas
stream by gravity in a baffled settler.
Then, a quencher cooled the gas and a
liquid entrainment separator removed
the water drops before the gas was
vented to the atmosphere.
High Temperature Particle
Charging Experimental
Results
The particle charger, tested at tem-
peratures up to 800°C, did not show any
operational problems. The character-
istics of the charger were then experi-
mentally determined. The negative
corona current-voltage characteristics
of the bench-scale hot EPxP particle
charger are shown in Figure 7 for
various temperatures. It is generally
acknowledged that spark-over voltage
will decrease as gas density decreases.
At elevated temperatures, the spark-
over voltage is lower and there is a
narrower operating range between the
corona starting voltage and the spark-
over voltage.
A Faraday cup, consisting of a
shielded, insulated glass fiber filter
connected to an electrometer, was used
to measure the charge/mass ratios at
various temperatures. Figure 8 shows
the results measured for fly ash
particles with a mass median diameter
of about 1.5 //m aerodynamic diameter
and a geometric standard deviation of
2.0. The charge level is lower for high
temperatures than at room tempera-
tures. However, the obtainable charge
level at high temperature is still
acceptable.
Particle Experiments
Particle penetration was determined
for the EPxP scrubber at several gas
temperatures. Figures 9 and 10 show
the results at ambient temperature and
500°C. The improvement in particle
collection efficiency due to augmenting
the PxP scrubber with electrostatic
force is clearly shown. The improve-
ment is greater for small particles. For
particles larger than 5 /urn aerodynamic
diameter, the improvement is minimal.
Present data show that high tempera-
ture particle collection is more difficult
than that at low temperature.
Discussion
Comparison of the bench scale
experimental results for UP/NC condi-
tions shows that the EPxP performance
is worse than predicted with this
mathematical model. The higher pene-
tration of the EPxP scrubber could be
due to particle bounce, attrition of
collectors, and dirty collectors.
The test particles used in the experi-
ments were redispersed power plant fly
ash, properties of which are different
from those of pressurized fluidized-bed
combustor (PFBC) fly ash. The PFBC fly
ash is sticky; therefore, particle bounce
would be less with PFBC particles. A
few experiments were done using
titanium dioxide particles which are
sticky and can be used to simulate PFBC
particles. Figure 11 shows the results.
the EPxP is more efficient in removing
titanium dioxide particles.
The Florida zircon sand used in the
experiment as collector granules is
irregular in shape and contains a large
amount of fine dust. Particle samples
collected by cascade impactors showed
two distinct colors, black in lower
stages and sandy brown in the upper.
This indicates either attrition or reen-
trainment of fine dust contained in the
sand. Cleaning the sand by passing it
through the system several times
-------
Outlet Sampling
Water.
Pump
o
•ncher
Collector
Hopper
A
Separator
\/
Vent
Natural
Gas
Panicle Injection
Figure 6. Hot EPxP bench scale system.
Collector
Reservoir
helped but did not completely alleviate
the problem.
Pilot Scale EPxP System
A 4.8 AmVmin (170 acfm) pilot scale
EPxP was designed and built for
controlling the particle emissions from
an atmospheric fluidized-bed coal
combustor (Figure 12). The pilot plant
Contained an integral collector cleaning
system for removing collected particles.
The collector continuously recirculated
in the system; i.e., collector granules
from the collector separator were
cleaned and reinjected into the EPxP
contactor. The pilot plant was operated
for several days at 870°C (1,600°F) with
no difficulties. However, attrition pre-
vented the collection efficiency deter-
mination. The particles collected by the
impactor at the outlet had a different
color than those at the inlet. Also the
particle concentration was higher at the
outlet.
Of several alternative collector gran-
ules, high density alumina spheres
were determined to be the best and are
able to stand high temperatures with
little attrition. However, only one
manufacturer produces small, 0.1 mm
diameter alumina spheres and only in
limited amounts. Since the spheres are
not mass produced, it is not economically
feasible to use them in the EPxP.
Discussion
The particle control device should be
capable of satisfying pollution control
regulations as well as minimizing
turbine erosion. There is considerable
uncertainty among experts as to the
particle tolerance of industrial gas
turbines. Westinghouse Research Lab-
oratories estimated that a particle
concentration below about 0.0046
g/Nm3 will not damage the turbine
blades. The allowable particle mass
concentration passing through a gas
turbine is estimated to be 0.0028
g/Nm3 by United Technologies Research
Center. General Electric estimated that
a GE heavy duty industrial gas turbine
could tolerate particle concentration of
about 0.1 g/Nm3, if 98% or more of the
particles are under 10 yum in diameter.
There is no current new Source
Performance Standard (NSPS) for
particle emissions from advanced energy
processes. It is likely that any standards
will be at least as stringent as those for
fossil-fuel-fired boilers. The NSPS for
fossil-fuel-fired boilers is being lowered
to 13 mg/MJ, corresponding to a mass
concentration of about 0.03 g/Nm3 for a
typical PFBC. This value is about 5-
10 times less stringent than the turbine
requirements estimated by Westing-
house and United Technologies. It is
more stringent than GE's projected
turbine tolerance.
The particles from a PFBC secondary
cyclone have a mass median diameter
of about 4.8 /vm and a geometric
standard deviation of about 3.2. The
particle concentration is about 2.3 g/Nm3
(1 gr/scf). Using the experimental grade
penetration curve for ambient tempera-
ture, the EPxP can clean the gas to meet
GE's projected turbine tolerance at
a scrubber pressure drop of 51-65 cm
W.C. when the particles are precharged
and the collectors polarized. The EPxP
cannot meet the cleanup requirements
with the same pressure drop at high
temperatures. To meet the cleanup
requirements, it is necessary to operate
the EPxP at a higher pressure drop.
Since neither experimental data at
higher pressure drops nor a reliable
mathematical model is available, the
required scrubber pressure is not
known.
Conclusions
The EPxP can operate under high
temperature conditions and has the
-------
2.5
2.0
1 1.5
c
1
| 1.0
O
0.5
Air, 1 atm
Plate-to-Plate = 3.3cm
Wire-to-Wire = 2.5 cm
Wire Diameter = 0.084 cm
650°C
20°C
potential to clean the gas to meet NSPS
and turbine requirements. The particle
collection efficiency of the EPxP can be
greatly increased by precharging the
particles and polarizing the collectors.
Recommendations
Additional experiments should be
done to obtain conditions under which
the EPxP can clean the gas to meet both
the NSPS and turbine requirements. To
optimize the scrubber design, a reliable
mathematical model should be devel-
oped.
0 5 10 15 20
Applied Voltage, kV
Figure 7. Measured voltage-current curve as a function of temperature.
25
100
50
tt
Ui
! s
13.5M
o °o
4.5 kV
0
6.5 kV
O
6 kV
1
Figure 8.
100 200 300 400 500
Gas Temperature, °C
Experimental charge/mass ratio.
600 700 800 900
-------
1.0
0.5
0.3
I
0.1
0.05
0.03
Gas Temp. = Ambient
Pressure Drop = 50 cm W.C.
Fly Ash Particles
1.0
0.5
fo.
§
I
0.1
0.05
0.5 / .5 10
Aerodynamic Particle Diameter, umA
Figure 9. Experimental grade penetratoin for the high-
temperature-hardware EPxP scrubber at
ambient temperature.
0.03
UP/NC
~ Gas Temp. = 500°C
Pressure Drop = 51 cm W.C.
Fly Ash Particles
0.5 1 2345 J
Aerodynamic Particle Diameter, umA
Figure 10. Experimental grade penetration for the
high-temperature-hardware EPxP scrubber
at 500 °C.
1.0
0.5
S.-C. Yung, T. Le. R. Patterson, and S. Calvert are with
A.P. T., Inc., San Diego. CA92117.
Dtnnic C. Drehmelis the EPA Project Officer (see below).
The complete report, entitled "Augmentation of Fine
Panicle Collection in the EPxP Scrubber/'fOrder No. PB
82-249 186; Cost: $13.50. subject to change} will be
available only from:
National Technical Information Service
5285 Port Royal Road
Springfield. VA22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park. NC 27711
•I 0.1
1
Q>
£ 0.05-
0.01
Titanium Dioxide Particles
UP/NC
Gas Temp. - Ambient
Pressure Drop = 51 cm W.C.
0.1
Figure 11.
0.5 1 5
Aerodynamic Particle Diameter, umA
10
Effect of particle type on EPxP performance.
*USGPO: 1982—559-092/0493
-------
Coal Bed
Feed Hopper
Combustor
3" Refractory Lined
1500°F
510 dm
0.2 psig
On-Off
Hydraulic
Valve
1500° 170cfm
0.2 psig
Rotary
Feeder
10-50lb/hr
Ignition
Burner
10.5 mm Btu/hr)
1500OF
170 cfm
-0.5 psig
Collector
Separator
Backup
Collector
Hopper
(Min. 2000 Ib.
Capacity)
r-—-
Fluidizing Gas
Compressor
(135 scfm - 4 psig
5-15H.P.)
11500°F
170 cfm
-0.5
psig
k \-
f Polarizing
'PxP
Contactor
, Particle
Charger
Venturi Scrubber
Exhaust Blower (250 cfm
Capacity 25" W.C. Discharge
Press 1 'A h.p.)
Figure 12. £PxP pilot plant.
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
------- |