United States
Environmental Protection
Agency
Air and Energy Engineering
Research Laboratory
Research Triangle Park NC 27711
Research and Development
EPA/600/S7-85/021 Aug. 1985
Project Summary
Pilot-Scale Investigation of
Closed-Loop Fly Ash Sluicing
S. T. Litherland, P. A. Nassos, M. L. Owen, and S. L Winton
This project was a pilot-scale demon-
stration of the technical feasibility of
closed-loop operation of fly ash sluicing
systems. Chemical species leached
from the ash increase the dissolved
solids concentration of recycled sluice
water to a point where scaling of equip-
ment may occur. Tests were conducted
at two power plants using a 50 gpm
pilot unit to demonstrate the feasibility
of closed-loop operation, both with and
without sluice water treatment. An ash
sluice computer process model was de-
veloped to predict chemistry and pro-
cess conditions in full-scale systems.
Fly ash sluicing systems handling
highly reactive alkaline ashes cannot be
operated closed-loop without treating
sluice water to control scale formation.
Acid addition for pH adjustment was ef-
fective in controlling calcium carbonate
scale formation in the sluice water re-
turn line; however, use of surfuric acid
increased the potential for gypsum
scale formation. Gypsum was ulti-
mately the limiting species which pre-
vented reliable closed-loop operation at
the plants tested. Increased ash/water
contact time in a reaction tank was not
adequate to control the potential for
gypsum scale formation at the resi-
dence times tested.
The ash sluice computer process
model proved to be accurate in predict-
ing the chemical composition and po-
tential for scale formation in the pilot
unit. This model is an effective tool for
the resolution of ash sluicing system
operating problems or to support the
design of systems in new plants.
This Project Summary was devel-
oped by EPA's Air and Energy Engineer-
ing Research Laboratory, Research Tri-
angle Park, NC, to announce key
findings of the research project that is
fully documented in two separate vol-
umes (see Project Report ordering in-
formation at back).
Introduction
The intimate mixing of coal-fired
power plant ash and water during sluic-
ing causes soluble species (e.g., cal-
cium, chloride, sulfate, sodium, and
trace elements) to be leached from the
ash. An increase or decrease in the pH
level can also result, depending on the
characteristics of the coal ash. The efflu-
ent guidelines limitations for the steam
electric power industry require zero dis-
charge from new plants with fly ash
sluicing systems. Restrictions for exist-
ing plants were not included in the final
effluent limitations promulgated in
November 1982. Operation of a zero dis-
charge (closed-loop) fly ash sluicing
system has not been common in the
utility industry. To meet the zero dis-
charge requirement for fly ash sluicing
in new plants, designers have three op-
tions: 1) operate systems closed-loop,
2) dispose of the dry fly ash, and
3) blend the fly ash with sludge from a
wet flue gas desulfurization (FGD) sys-
tem at plants with wet FGD systems.
Recycling ash pond water to create a
closed-loop ash sluicing system can
cause many operating problems. Scale
formation in the system is a major con-
cern in closed-loop operations. Cal-
cium, magnesium, sodium, sulfate, and
silica are the major species leached
from the fly ash and subsequently con-
centrated with recycle. Concentrating
these soluble species in a closed-loop
system can supersaturate the sluice
water with calcium carbonate (CaC03),
gypsum (CaS04-2H20), magnesium hy-
droxide (Mg(OH)2), or silica com-
-------
pounds, resulting in precipitation
which, in turn, causes scale formation in
parts of the system. Failure to control
the buildup of scale-forming com-
pounds in the ash sluicing system by
appropriate treatment may result in
plugging of piping, pumps, and other
system equipment encountering ele-
vated temperatures. This scaling can
adversely affect the reliability of the ash
sluice system, thus reducing the avail-
ability of the generating unit.
To address the potential impacts of
closed-loop operation on power plant
ash sluicing systems, EPA has con-
ducted three projects to define potential
chemistry-related operating problems,
and to test several control approaches.
The first project was to provide techni-
cal and economic evaluations of water
recycle/reuse and treatment options to
assist utilities operating or designing
coal-fired power plants to reduce water
consumption and treatment costs. Re-
sults of the first project indicated that
closed-loop operation of fly ash sluicing
systems would require treatment to
control scale formation. Because of the
variability and developmental nature of
the computer model used in the evalua-
tion, additional bench- and pilot-scale
studies were recommended.
In response to the recommendations
of the first project, the second and third
projects were initiated. In the second
project, funded by EPA's Effluent Guide-
lines Division in 1979, the ash sluice sys-
tems at three power plants were charac-
terized. Ash collected in the field was
used in a bench-scale simulation of a
closed-loop ash sluice system to evalu-
ate ash leaching and dissolved solids
buildup. These data were used to refine
the accuracy of the ash sluice system
computer simulation model as well as
to provide a design basis for a pilot-
scale field testing system.
The third project, which is the subject
of this report, involved the design, con-
struction, and field testing of a pilot unit
and use of the field data to evaluate the
ash sluice process model.
Project Objective
The primary objective of this project
was to investigate the technical feasibil-
ity of closed-loop operation of fly ash
sluicing systems. To achieve this objec-
tive, three major tasks were completed:
• Development, design, and construc-
tion of a pilot system—The pilot
system design was developed using
both full- and bench-scale ash sluic-
ing data. The system was designed
2
and constructed with a high degree
of flexibility to be able to 1) dupli-
cate the wide range of flow and mix-
ing conditions found in full-scale
systems, and 2) provide three treat-
ment approaches for reducing the
potential for scaling. The system
was also designed to be easily
transportable and adaptable to sev-
eral plant locations.
• Test plan development and field
testing of the pilot system—De-
tailed test plans were prepared for
operating the pilot unit at each par-
ticipating plant. Test conditions
were determined using bench-scale
ash sluice system characterization
data and ash leaching studies.
• Development/verification of the ash
sluice computer process model
using field data.
Field Testing Approach
The approach used involved operat-
ing the ash sluice pilot system with ash
and makeup water from a host power
plant. Figure 1 is a flow diagram of the
ash sluice pilot system. The system con-
tains a makeup tank for blending recy-
cled ash sluice water with makeup
water. The mixture flows to a mixing
tank where fly ash is added to produce
a slurry. The mixing tank is large
enough to provide a slurry residence
time comparable to that required to
transport ash from the plant to a settling
pond. Additional mixing time can be ob-
tained by using a reaction tank. A filter
press is used to remove ash solids from
the system. The filtered water is col-
lected in a recirculation tank and
pumped to the makeup tank for reuse.
The pilot system includes three treat-
ment options to control scaling in the
system:
• Adjusting the pH of the recycle
stream,
• Softening the sidestream of the re-
cycle stream, and
• Mixing the ash slurry longer to
allow chemical precipitation in the
reaction tank.
The system is designed so that one or a
combination of these treatment options
can be evaluated in a single test run.
Pilot testing was completed at Plants
9677 and 9991 during the program. Both
plants produce highly reactive fly ash
that results in an alkaline ash sluicing
water. The fly ash from Plant 9991 is
more reactive than that from Plant 9677.
Plant 9991 burns a western subbitumi-
nous coal, and Plant 9677 burns a bitu-
minous coal mined in western Ken-
tucky. The fly ash from Plant 9991
contains about twice as much leachable
calcium as does that from Plant 9677.
A complete series of tests with anc
without treatment were initially plannec
at each site. However, due to projec
budget limitations, only the following
tests were conducted:
• Plant 9677
- no treatment
- using the reaction tank to increase
contact time; and
• Plant 9991
- no treatment
- pH adjustment using sulfuric anc
hydrochloric acids.
Sidestream softening was not used a
either plant.
Both plants also sluice ash to settlinc
ponds and reuse pond water for asf
sluicing. A portion of the sluice water h
discharged from the ash settling pone
at Plant 9677. Nevertheless, some seal
ing has been observed at this plant
However, since the installation of a pi-
control system, no scaling has been ex
perienced in any part of the ash sluicinc
system at this plant. Plant 9991 operate:
its ash sluicing system closed-loop anc
has experienced some scaling in recir
culation pump screens. This scaling ha:
not been severe enough to cause i
major plant operating problem. Thi
plant is relatively new, and the asl
sluice water had not concentrated sig
nificantly at the time of the field test
with the pilot system. Because scalini
had been noticed at various points ii
the system, plant personnel were con
cerned about scale buildup and contro
Results
Program results can be divided int
three major areas: (Dash leachin
characteristics, (2) feasibility of closec
loop operation, and (3) verification c
the ash sluice computer process mode
Each area is summarized below.
Ash Leaching Characteristics
During the project the weight frac
tions of key chemical species leachei
from fly ash were evaluated based oi
three sources of test data: laborator
leaching tests, pilot unit tests, and full
scale system characterization tests.
Table 1 summarizes the Plant 967
ash leaching data from the laborator
batch tests, the full-scale system, am
the pilot unit. Both the average and th
range of calculated values are reportec
Also shown in Table 1 are the calciur
carbonate and gypsum relative sature
tion values calculated by the equilit
-------
Makeup
Water
Figure 1. Ash sluice pilot system flow.
-------
rium model for the leachate composi-
tion. The relative saturation values were
used to determine if a potential for pre-
cipitation existed, since precipitation
must be accounted for during the leach-
ability calculations. The relative satura-
tion values indicated that, for each data
set, the potential for precipitation of cal-
cium carbonate existed in the leachate.
A potential for gypsum precipitation ex-
isted only for the pilot unit tests.
Based on the average values, the cal-
culated quantities of species leached for
the laboratory and full-scale data agree
extremely well. The differences in the
relative saturation values in Table 1 are
due primarily to differences in the water
(recycled plus makeup) composition
prior to being mixed with the ash. The
relative saturation values were used to
determine if reprecipitation of calcium,
sulfate, or carbonate would be ex-
pected. The average calculated quantity
of calcium leached during pilot testing
was about 20 percent lower than that for
the laboratory or full-scale data. This
does not necessarily indicate a decrease
in the quantity of calcium leached, but
may reflect additional precipitation of
calcium that was not accounted for dur-
ing the calculation. The transfer of CO3
from the atmosphere was suspected
during the pilot unit test, which could
have precipitated with the calcium as
calcium carbonate resulting in a lower
calculated leach quantity. The quantity
of sulfate leaching during the pilot unit
tests could not be calculated because of
the precipitation of gypsum. For this
reason, the laboratory sulfate leach data
were used for pilot unit calculations.
The ranges of values in Table 1 indi-
cate considerable variation in calculated
leaching quantities. These variations
can be attributed to two factors: 1) there
is some analytical error (±5 percent) as-
sociated with the measured values used
in the calculation, and 2) fluctuations in
the coal composition would affect the
fly ash composition and the leachability
of the species from the fly ash. The com-
parison of Plant 9677 leaching data indi-
cates that (although there is a fairly
wide range in the values calculated
from the three sources) the average
values compare well. In addition, the re-
sults suggest that (for similar ashes) the
quantities of species leached during
pilot- and full-scale sluicing can be ade-
quately estimated from laboratory
batch leaching tests. Also, these results
suggest that the quantity of a species
leached from the ash is not a function of
the initial water composition; however,
reprecipitation of supersaturated spe-
cies may occur at high concentrations.
Table 2 gives the ash leaching data
from laboratory, pilot-, and full-scale
determinations from Plant 9991. The
ranges of the quantities of species
leached per gram of ash again show
considerable overlap between the three
data sources. In general, averages of the
estimated quantities leached compare
reasonably well for all of the species.
The most important discrepancy is the
estimated quantity of calcium leached
from each source, although all of the
calcium values are the same order of
magnitude. Relative saturation values
indicate that calcium carbonate was su-
persaturated in data obtained from all
three data sources. The estimate of the
quantity of calcium leached could have
been affected if additional C03 were ab-
sorbed and unaccounted for, and
Table 1. Summary of Ash Leaching Data-Plant 9677
Quantity Leached, mg/g ash
leached calcium was precipitated as cal-
cium carbonate during the tests. Since
the laboratory apparatus was covered,
the levels of carbonate were so low that
the quantity of calcium carbonate that
could have been precipitated would
have been minimal. More opportunity
for CO3 transfer existed in the pilot unit,
but the greatest opportunity for C03
transfer was in the ash ponds of the
full-scale system. The greatest uncer-
tainty in the ash leaching data is for the
estimates prepared from full-scale data
since the weight percent ash in the
slurry was not determined. Also, in a
full-scale system C03 could be intro-
duced from the following sources:
• as a component of fluidizing air;
and
• as a component of boiler flue gas
contained in the fly ash interstitially
or from leaking fly-ash hopper
valves.
Feasibility of Closed-Loop Op-
eration
Pilot unit test results indicated that
closed-loop operation would not be
possible at either plant tested without
some form of treatment to control scale
formation. Since funding limitations
prevented test programs using all treat-
ment options from being performed at
each plant, the optimum treatment ap-
proach and control limits could not be
determined. However, two of the three
treatment capabilities of the pilot sys-
tem were used during the two field
tests.
Table 3 summarizes the test condi-
tions and the scaling tendency (relative
saturation) of the recycle water for tests
at Plant 9677. All tests were conducted
Relative Saturation
Calcium
Data Source
Field Laboratory6
- average
- range
Calcium
4.2
3.6-5.3
Magnesium
0.02
0.01-0.04
Sodium
0.12
0. 10-0. 16
Potassium
0.17
0. 13-0.20
Chloride
0.03
0.00-0.08
Sulfate
6.5
4.0-10.1
Slurry pH
70.0-77.0
Carbonate
29
19-45
Gypsum
0.3
0. 1-0.5
Full-Scale System
-average 4.1 0.05 0.11 0.25
-range 3.2-5.0 0.02-0.07 0.10-0.13 0.22-0.27
Pilot Unit
-average 3.3 0.03 0.09 0.18
- range 2.9-3.8 0.01-0.05 0.02-0.20 0.06-0.35
0.02
0.00-0.03
6.6
4.6-8.1
6.5d
4.7-9.8
10.2-10.5
b
0.01-30.8
>100
0.2
0.1-0.3
2.2
1.9-2.5
"Performed using deionized water.
bAverage not reported due to variations in pH.
°No leaching was measured.
dCould not be measured due to precipitation of gypsum, assumed to be the same as the laboratory value.
-------
Quantity Leached, mg/g ash
Data Source
Field Laboratory3
- average
- range
Full-Scale System
- average
- range
Pilot Unit
- average
- range
Calcium
8.1
6.2-9.9
4.0
7.8-7.8
5.7
3.5-8. T
Magnesium Sodium
0.37
0.23-0.51
0.43
0.23-0.78
0.40
0.24-0.73
Potassium
0.03
0.0-0.06
0.02
0.01-0.02
0.05
0.04-0.08
Chloride
0.03
0.0-0.06
0.08
0.08b
0.12
0.1 2C
Sulfate
3.7
3.2-4.2
3.0
1.1-7.0
2.5
1.9-3.0
Relative Saturation
Calcium
Slurry pH Carbonate
7.9
12.2-12.7 7.7-8.1
94
11.1-12.4 65-120
no
11.8-12.6 50-150
Gypsum
0.2
0. 1-0.3
0.3
0.2-0.4
0.6
0.2-7.5"
"Performed using plant ash sluice system makeup water.
bBased on results from data collected on October 22, 1982.
°Based on Test 6 results only.
dGypsum was not saturated except in Test 3.
with 100 percent recycle of sluice water
with slurry concentrations in the 10-12
weight percent range. The only differ-
ence between the test runs was the in-
corporation of more ash mixing time in
the reaction/hold tank in Tests 2 and 3.
In all tests, the recycle water was super-
saturated with respect to both calcium
carbonate and gypsum. During the
tests, a light coating of scale was noted
in several parts of the pilot system. Un-
fortunately, the testing had to be com-
pleted before enough scale had accu-
mulated for sampling and identification.
Figure 2 summarizes chemistry and
flow for Test 2 at Plant 9677. The cal-
cium carbonate relative saturation is
over 400 at all points in the system indi-
cating that precipitation was probably
occurring.
The gypsum relative saturations also
indicate a potential for precipitation.
The potential for gypsum precipitation
is high because the concentrations of
both calcium and sulfate in the recycle
water are very high. A significant depo-
sition of gypsum scale may form very
rapidly in ash sluice system equipment,
especially where higher temperatures
may be encountered.
The reaction/hold tank was used to
provide an additional 5-10 minutes of
ash mixing. Bench scale studies had in-
dicated that this approach might be an
effective mechanism for increasing pre-
cipitation on the ash particles to desu-
persaturate the sluice water. The results
of the field tests at Plant 9677 indicated
that only a slight reduction in gypsum
relative saturation was obtained. The
relative saturation of calcium carbonate
(increased. This suggests that longer re-
tention times would be required to pro-
Table 3. Summary of Pilot Test Runs at Plant 9677
Recycle Water3
Test
No.
1
2
Slurry
Concentration
wt%
10.0
10.2
Treatment
_c
Reaction tank
pH
10.5
10.2
Calcium
Carbonate
180
590
Gypsum
2.5
1.9
(5 minute retention)
12.8 Reaction tank
(10 minute retention)
10.4
470
2.1
a100 percent sluice water recycled.
bRelative saturations > 1 indicate a potential for precipitation/scale formation.
cNo treatment.
mote precipitation to achieve solid/
liquid equilibrium for plants with sluice
water compositions similar to Plant
9677 (e.g., plants burning bituminous
coals producing alkaline ash sluice
water).
Table 4 summarizes the field tests
performed at Plant 9991. Six tests were
performed with varying degrees of re-
cycle and slurry concentrations in the
5.8 - 11.9 weight percent range. After
baseline tests with no treatment of the
ash sluice water, tests were conducted
using both sulfuric and hydrochloric
acids to control the pH of the recycled
water. Due to the higher reactivity of the
ash, the approach at Plant 9991 was
slightly different from that at Plant 9677.
At Plant 9991, laboratory ash leaching
data were input to the ash sluice com-
puter process model, and a blowdown
rate was calculated that would maintain
gypsum relative saturation at a value of
1.0. Tests 3 through 6 were run with
blowdown and recirculation rates elimi-
nated in this manner.
Figure 3 summarizes chemistry and
flow conditions during Test 4 with hy-
drochloric acid used to adjust pH. The
relative saturation of calcium carbonate
dropped from 122 to 12 across the recy-
cle tank where the acid is added. A rela-
tive saturation of 12 is comparable to
that measured in the effluent of a soft-
ener and, therefore, should not be high
enough to promote precipitation. Gyp-
sum was unsaturated in this test. How-
ever, a further decrease in blowdown
would increase the gypsum relative sat-
uration to a critical point. The high cal-
cium and sulfate concentrations in the
recycle water indicate that a large quan-
tity of scale may form in the ash sluice
system if gypsum becomes supersatu-
rated.
The test program at Plant 9991 fo-
cused on pH adjustment as a way to
operate closed-loop without scale for-
mation. Hydrochloric acid was effective
in reducing calcium carbonate relative
saturation to levels of low potential for
precipitation. Sluice water recycle at
-------
Concentrations in mg/l as ion except alkalinity
which, is in mg/l as CaCO3. RS stands for
8.3 1
Ca =7 Cl =2 <2'2i
Mg = 2 Alk = 25.6
Na = 3 CO3 = 3/
K = 2 SOt= 11
pH = 6.3
145 1pm
(38.3 gpm)
reiauve saturation.
ipm) Ash
— ». 7046 kg/hr
(38.4 Ib/min)
-^- Makeup \
Tank *
40.5 gpm
Mix Mfl
Ca =1210 Cl =8 Tank In9vuiv f
Ma - 2 Alk - 580 W'2 wt% K
— 1330 Cl =8
= 5 Alk = 480
- 148 CO 3 = 252
= 264 SO4 = 3370
Na =140 C03 = 285 solids \ p" ~*-* „„„
K =247 SO. =2910 1 ** CaCO =402
CaC03 = 591
CaSO*>2H20 = 1.91
Reaction
Tank
(5 min)
Ca
145 1pm f \ Mg
, , (38.3 gpm) [ Filter Press W Na
V / P"
- 73/0 Cl ~8
= 5 Alk = 550
- 152 C03 - 240
- 269 SO4 - 334C
= 9.6
Tank I RS CaS04 • 2H2O = 2.21
*
Ca = 1200 Cl =8 jH 1pm,
Mg =2 Alk =570 <~ f'fg^...
Na = 146 CO, = 300 67'6 wf/0 soMs
K =251 SO« = 2960
pH = 10.2
RS CaC03 = 596
RS CaSOt • 2HtO = 1.92
Figure 2. Summary of pilot unit data from Test 2—Plant 9677.
Table 4. Summary of Pilot Test Runs at Plant 9991
Recycle Water
Slurry Percentage of
Test Concentration Sluice Water
No. wt% Treatment Recycle pH i
1
2
3
4
5
6
6.3
8.8
5.8
5.9
11.9
8.8
_b
_b
Sulfuric acid0
Hydrochloric acidc
Hydrochloric acidc
Sulfuric acidc
0
0
68
78
56
27
11.9
12.4
9.3
8.4
8.5
8.3
Relative Saturation3
Calcium
Carbonate
50
100
46
12
12
6
Gypsum
0.2
0.3
1.8d
0.6
0.7
0.8
aRelative saturations > 1 indicate a potential for precipitation/scale formation.
bNo treatment.
cpH adjusted with indicated acid.
dDuring this test, makeup water sulfate concentration increased fivefold.
78 percent produced a gypsum relative
saturation of 0.6 in the recycled sluice
water. Sulfuric acid could not be used to
control scale; the sulfate introduced in-
creased the amount of blowdown that
was required to avoid gypsum scale for-
mation. At higher slurry concentrations
(about 10 weight percent) and pH ad-
justment with hydrochloric acid, recycle
rates were limited to about 56 percent.
Adjusting pH with hydrochloric acid can
control calcium carbonate scale in
closed-loop operation; however, it may
limit operation by gypsum scale.
Verification of Ash Sluice Com-
puter Process Model
A secondary objective of the field test-
ing was verification of the capability of
the ash sluice computer process model
to predict operation of pilot- and full-
-------
Concentrations in mg/l as ion except alkalinity
which is in mg/l as CaCO3. RS stands for
pH = 10.2 Temp = 28°C
Ma -07 Cl ~ 37 34.71pm
Na =119 S04 = 98 rri- — no
relative saturation.
Ash
498 kg/hr
133 1pm f
(35.0 gpm)
Temp — 28°C Mix .. .
9.2 Cl = 2090 Tank /^ 7^"
1310 SO. = 489 (37'6 9f)m>
171 C03 = 94 5.9 wt%
8.3 Ma = <0.5
C03 = 97.3
SO4 • 2H20 = 0.44
.02 1 pm
)00
127 1pm/ \
^ (33.5 gpm) 1 Filtf!r Pr(!SS W 1
Ca = 7720 Cl = 1830
Na = 199 SO4 = 633
K = 11 CO3 = 62 1 6.06 1pm
pH = 12.6 Mg = <0.5 f (1.6 gpm)
Ca
Na
K
pH
RS
RS
RS CaCO3 = 122 57.9 wt% solids
RS CaSOt >2H2O = 0.56
Temp = 29°C
= 1670 Cl = 2030
= 790 S04 = 622
= 11 C03 = 48
= 12.6 Mg = <0.5
RS CaC03 = 94.6
RS CaSOt • 2H20 = 0.56
RS CaSOt -2H20 = 0.62
Figure 3. Summary of pilot unit data from Test 4--Plant 9991.
scale systems. The model consists of an
internal chemical equilibrium model
supported by a set of subroutines. The
equilibrium model calculates ionic in-
teractions and the potential for precipi-
tation of solids in aqueous systems
which contain calcium, magnesium,
sodium, potassium, ammonia, silica,
chloride, carbonate, nitrate, sulfate, and
phosphate. The support subroutines
simulate the various components of an
ash sluicing system (pond, softener, re-
action/hold tank, etc) and perform mate-
rial balance calculations.
As discussed earlier under "Feas-
ibility of Closed-loop Operation," the
model was used to predict the blow-
down flowrates that would result in a
gypsum relative saturation value of 1.0
in the recycled sluice water for Plant
9991.
>The output of the model can only be
as accurate as the quality of the input
data to the model. If the input data (e.g.,
ash leaching, makeup water chemistry)
accurately reflect the characteristics of
an ash sluice system, then the model
should predict system chemistry and
flows relatively accurately. Table 5 com-
pares the chemistry, relative saturation,
and system flowrates measured during
Test 4 at Plant 9991 in the pilot system
and predicted by the model. Ash leach
rates, makeup water composition, and
the approach to solid/liquid equilibrium
measured in the pilot system were used
as inputs to the model. The model accu-
rately predicted all chemistry and flow
rate data within the limits of analytical
and flow measurement accuracy (±20
percent). The only significant deviation
is in the prediction of alkalinity, which
directly affects the relative saturation
calculations. Although the relative error
of the alkalinity is high, the actual con-
centrations involved are fairly low.
Thus, even though the driving force for
scaling (relative saturation) is high, the
quantity of material that could poten-
tially be deposited is relatively small.
To demonstrate the predictive capa-
bility of the model, data from Test 4
were used to project the minimum
blowdown that could be achieved with-
out gypsum supersaturation. Hydro-
chloric acid was used to adjust the pH of
recycle water in this test. The blowdown
composition for the minimum blow-
down case is compared with the model
prediction for Test 4 (78 percent recycle)
in Table 6. At a 94 percent sluice water
recycle rate, gypsum would become
supersaturated, indicating that closed-
loop operation may result in precipita-
tion. This prediction could not be veri-
fied in the field; however, from the
previous discussion showing the accu-
rate duplication of field test results, this
estimate should be reasonably accu-
-------
Table 5. Comparison of Ash
Test 4
System Slowdown Composition
Component, ppm as ion
Ca
Mg
Na
K
Cl
Alkalinity (as CaCO3)
SO4
pH
Relative Saturation
CaCO3
Gypsum
System Flow Rates, Ipm (gpm)
Hydrochloric acid
Makeup
Sludge (water loss)
Sluice Model Predictions
Predicted
by Model
1650
0.2
186
9.6
2670
14.8
722
8.4
4.8
0.68
0.80 (0.21)
34 (8.9)
5.7 (1.5)
with Pilot Unit Performance for
Measured in Difference
Pilot Unit percent
1700
<0.5
197
10
2740
30
647
8.4
11.9
0.62
1.0 (0.27)
35 (9.2)
6.1 (1.6)
3
N/Aa
6
4
3
50
-12
0
N/A
-9
22
3
6
were specified. Critical inputs to the
model that can significantly affect
the accuracy of model predictions
include 1) ash leaching data, 2) the
quantity of C03 transfer in the ash
ponds, and 3) estimates of the po-
tential for precipitation in ash sluice
equipment. Program results indi-
cate that laboratory ash leaching
data are adequate to use as inputs
to the model.
a/Vof applicable.
rate. Although gypsum scale formation
had not been encountered at Plant 9991,
the concentrations of species in the full-
scale system were much lower than pre-
dicted by the model. This indicates that
the plant had not yet reached steady
state operation. Due to the large vol-
umes of the ash ponds and the firing
rate at the plant, several years could be
required before gypsum scaling is ob-
served.
Conclusions
Program results indicated the follow-
ing conclusions.
• Fly ash sluicing systems handling
highly reactive alkaline ashes (char-
acteristic of low sulfur western coal
ashes) cannot be operated closed-
loop without encountering precipi-
tation in the system unless water is
treated to control scale formation.
The results obtained from the full-
scale plant characterization, the
pilot unit, and the ash sluice com-
puter model are consistent in indi-
cating precipitation will occur.
• Adding acid to adjust pH should be
effective in controlling calcium car-
bonate scale formation in the sluice
water return line. Economic consid-
erations will probably favor sulfuric
acid over hydrochloric; however,
use of sulfuric acid will increase the
potential for gypsum scale forma-
tion. Other treatments, such as side-
stream softening, will be necessary
to prevent gypsum scale formation
during closed-loop operation.
Table 6. Comparison of Minimum Blow-
down Simulation with Simula-
tion of Test 4
Model
Predicted
Minimum
Slowdown
System Slowdown
Composition
Component, ppm
as ion
Ca
Mg
Na
K
Cl
Alkalinity
(as CaCO3)
SO4
pH
Relative Saturation
CaCO3
Gypsum
3910
0.1
343
23.8
6695
7.5
1036
8.4
3.1
1.1
Pilot
Unit
Test 4
1650
0.2
186
9.6
2670
14.8
722
8.4
4.8
0.68
System Flow Rates,
Ipm (gpm)
Hydrochloric
Acid
Makeup
Sludge
(water loss)
Liquid
Slowdown
Total
Slowdown
0.80 (0.21)
12 (3.3)
5.7 (1.5)
7.6 (2.0)
13 (3.5)
0.80 (0.21)
34 (8.9)
5.7 (1.5)
29 (7.6)
34 (9. 1)
The ash sluice computer model was
an effective tool for predicting the
chemical composition and potential
for scale formation in the pilot unit
when accurate inputs to the model
-------
S. Litherland, P. Nassos, M, Owen, andS. Winton are with Radian Corp., Austin,
TX 78766.
Julian W. Jones is the EPA Project Officer (see below).
The complete report consists of two volumes:
"Pilot-Scale Investigation of Closed-Loop Fly Ash Sluicing: Volume 1. Final
Report," (Order No. PB 85-204 378/AS; Cost: $16.00, subject to change).
"Pilot-Scale Investigation of Closed-Loop Fly Ash Sluicing: Volume 2.
Appendices," (Order No. PB 85-204 386/AS; Cost: $32.50, subject to
change).
The above reports 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
. S. GOVERNMENT PRINTING OFFICE:1985/559-l 11/20626
-------
CO
•jj
00
01
o
ro
c
(A
oo
TJ
m
20
p >
Q
CO
(71
------- |