United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
Research and Development
EPA-600/S7-82-015 August 1982
Project Summary
Recovery of Calcium
Carbonate and Sulfur from
FGD Scrubber Waste
R. P. Arganbright, P. Huang, G. S. Benner, B. G. Mandelik, T. S. Roche,
and P. V. Smith
Key process steps were demonstrated
in the Kel-S process, a proprietary pro-
cess for recovering calcium carbonate
and sulfur from lime/limestone flue gas
desutfurlzatlon (FGD) scrubber waste.
These steps are reduction of the waste
to calcium sulflde (using coal as the
reducing agent), carbonation of the cal-
cium surHde to generate hydrogen sulfide
and calcium carbonate, and recovery of
precipitated calcium carbonate from
inerts (e.g., coal ash).
Conversion of 99 percent of the
calcium surfate/sulfite to calcium sulflde
was achieved both in a laboratory fur-
nace and in a pilot plant kiln. Conversion
of the sulflde to carbonate and hydrogen
sulflde in bench-scale equipment gave
very high rates of conversion (less than
1 percent calcium sulflde remaining In
the solids), and the concentration of
hydrogen sulfide in the offgas typically
exceeded 95 percent. Results from flo-
tation, centrlfugation, and filtration ex-
periments indicate that all three methods
could be used in the ash separation step,
but economics favor centrifugation. Tech-
nically, the Kel-S process was shown to
be ready for testing In a continuous pilot
plant.
However, a preliminary economic ana-
lysis of the Kel-S process was conducted
by the Tennessee Valley Authority (TVA)
for EPA. This analysis indicated that the
total cost of a limestone scrubbing FGD
process using the Kel-S recovery approach
was 30 percent higher in capital costs
and 19 percent higher in annual revenue
requirements than the Wellman-Lord/
Allied Chemical process (which has been
demonstrated on a commercial scale by
EPA). For this reason, EPA withdraw
financial support for the project.
This Prefect Summary was developed
by EPA's Industrial Environmental Re-
search Laboratory, Research Triangle
Park, NC, to announce key findings of
the research project that Is fuHy docu-
mented in a separate report of the same
title (see Project Report ordering Infor-
mation at back).
Introduction
Pullman Kellogg and EPA entered into
Contract No. 68-02-2644 which speci-
fied that Kellogg would continue research
and development of their proprietary
Kel-S process under a joint funding
arrangement. The process involves a
method of converting to usable products
the waste produced in lime/limestone
flue gas desulfurization (FGD) systems.
Major features of the process are pro-
duction of elemental sulfur using coal as
the reducing agent and process fuel and
recovery of calcium carbonate for recy-
cling to the FGD system. The following
processing steps are involved:
(1) Reduction of lime/limestone FGD
waste (calcium sulfite and sulfate)
to calcium sulfide.
(2) Carbonation of the calcium sulfide
to generate hydrogen sulfide and
calcium carbonate.
(3) Separation of inerts (such as coal
ash, unreacted coal, limestone in-
erts, and other impurities) from the
calcium carbonate.
-------
3*-<
(4) Conversion of hydrogen sulfide to
elemental sulfur in a standard Claus
plant.
The last step represents well-estab-
lished commercial technology and was
not included as part of this study.
Two distinct inerts-separation process
steps were investigated:
(1) A one-step process which con-
sists of first converting calcium
sulfide to calcium carbonate and
then removing the inerts from the
calcium carbonate by air flotation.
(2) A two-step process which con-
sists of first converting calcium
sulfide to calcium hydrosulfide
(which is water soluble) by reac-
tion with hydrogen sulfide, then
separating the inerts by filtration
or centrifugation, and finally pre-
cipitating calcium carbonate from
the calcium hydrosulfide solution
by reaction with carbon dioxide.
Reduction of the FGD waste occurs in
a rotating kiln where the waste and coal
are reacted at an elevated temperature
to convert calcium sulfite/sulfate com-
pounds to calcium sulfide. The study, as
originally planned, consisted of four
phases. Phase I included the following
tasks:
(1) Obtain FGD waste feed from a
commercial FGD system.
(2) Dry waste to the level needed for
making suitable pellets; determine
a suitable palletizing procedure;
determine the coal content needed
in the pellets for complete reduction.
(3) Determine the feed rate for drying
the pellets, the kiln operating tem-
perature, and the kiln feed rate.
(4) Perform a kiln production run to
make feed for pilot plant operation.
Phase II involved the following tasks:
(1) Perform bench-scale carbonation
tests to determine the maximum
particle size of reduced FGD waste
that could be used to obtain suit-
able carbonation rates, to deter-
mine operating conditions needed
to obtain maximum concentration
of hydrogen sulfide in the reactor
off-gas, and to provide informa-
tion for designing the pilot plant
reactors.
(2) Perform bench scale flotation,
centrifugation, and filtration tests
to show the feasibility of separat-
ing calcium compounds from coal
ash and inerts, to determine range
of operating conditions for the
pilot plant, an,?! to provide informa-
tion for sizing the pilot plant units.
All of the Phase I and II tasks, except
the continuous kiln production run (the
fourth task of Phase I), were completed.
Phases III and IV involved the design,
erection, and operation of a pilot plant
for recovery of calcium carbonate (as
well as the production of hydrogen sul-
fide) and separation of coal ash and
other inert materials. Most of the design
of the pilot plant and delivery of much of
the pilot plant equipment was accom-
plished before the contract was termi-
nated by EPA. (The rationale for contract
termination is included in the discussion
below.)
Phase I Results
Based on Kellogg in-house data previ-
ously obtained in bench scale equipment,
pellet compositions were specified for
the initial FDG waste reduction kiln test
work. Using these compositions, a pellet-
izing procedure was developed which
produced pellets strong enough to be
used in rotary kiln operations. The pellets
were produced by mixing about 4 wt
percent bentonite binder with the dried
coal/FGD waste mixture and then spray-
ing a 5 wt percent starch (in water) solu-
tion on the pellets as they were formed.
The palletizing procedure produced
pellets of variable quality because spray-
ing the starch solution on the pellets was
done manually and was difficult to con-
trol in below freezing weather. Several
mixtures of materials were tested until it
was determined that satisfactory pellets
could be produced by pelletizing mix-
tures containing 2 wt percent bentonite
and 2 wt percent wheat starch.
The green (wet) pellets were dried in a
pilot flowdryer that kept the pellets agi-
tated above a grate through which hot
air (270 °F)* flowed. Attempts to run
the dryer continuously were unsuccess-
ful; therefore, batch drying was used. The
dried pellets were collected in drums and
stored prior to kiln processing.
Using the dried pellets, kiln tests were
•made in a 24-in. outside diameter, 30-ft
'long rotary kiln (see Figure 1) at Kennedy
Van Saun, Inc. (KVS) in Danville, PA.
These tests were conducted to establish
the operating conditions necessary to
achieve a maximum reduction of the
FGD waste. Essentially complete reduc-
tion of the calcium sulfate/sulfite com-
pounds to calcium sulfide was achieved
(see Table 1).
•Readers more familiar with the metric system are
requested to use the conversion table at the end of
this summary.
Using a laboratory furnace at the
facilities in Houston, TX, Kellogg als
conducted reduction tests to expand th
data base obtained in the KVS kiln test!
These results confirmed the pilot kiln r<
suits and the minimum amount of co
required for complete reduction.
Phase II Results
Using the reduced material produce
in the KVS kiln tests, Kellogg carried 01
bench-scale carbonation tests (see Figui
2) to define operating conditions an
produce a feed for the air flotation test
to determine whether calcium carbonai
could be separated effectively froi
impurities. The bench-scale carbonatio
tests indicated that the conversion c
calcium sulfide to calcium carbonate an
hydrogen sulfide was extremely rapi
and apparently limited by the rate i
which carbon dioxide could be fed int
the system. Maximum concentration c
hydrogen sulfide in the effluent, ga
ranged from 90 to 100 percent for mos
of the runs (see Table 2), which is idei
feed gas to a Claus plant.
The effect of particle size of the n
duced FGD waste was studied usir
-12 to +140 mesh, and -140 mes
particle-size distributions, but the'resul
were not conclusive. It appears that pa
tide size within the range tested is n<
critical. The effects of pressure and ten
perature also were evaluated, but with
the range tested (20 psig, 150° 1
250°F), no significant differences wei
observed.
Product produced in the Kellogg bencl
scale carbonation tests was used in tr
air flotation test performed at Denv
Equipment facilities in Denver, CO.
was determined that three additives (f u
oil, pine oil, and a proprietary promot
manufactured by American Cyanamii
and three floats were needed to achie\
about 75 percent recovery of calciu
carbonate and approximately 50 percei
inerts rejection. These results indica
that air flotation could be used comme
cially (see one-step process, Figure 3)
Although air flotation was shown
be technically feasible, the air flotatk
additives which remained with the r
covered ash might preclude the use
the recovered ash as landfill because
waste disposal environmental regulation
Significant amounts of calcium carb
nate (25 percent) would also be rejecti
with the ash. It was then decided to e
amine the use of filtration and centrif
gation to remove ash from the systen
-------
Vent
TC = Thermocouple
S = Sample port
Firing
zone
Fuel
ID
Fan#1
Rotary
kiln
Drive Ram J
train (not used)
1
'
Kiln
discharge
^
(j
^
J)
(j
V
?)
^
G
FeedJ \
chute
)
D
Primary
fan
.Preheater
(not used) Hot cyclone
"7 bypass
Feed
Hot cyclone
(not used)
Figure 1. Kennedy Van Saun (KVS) rotary pilot kiln.
Table 1. Kel-S Pilot Plant Analyses-Mixture K1 Kiln Test (November 21-22, 1977)
Sample
Description
KD»
Port 1
KD
Portl
Port3
KD
Portl
KD
Port 1
KD
Portl
KD
Portl
Port3
KD
Portl
Time
Sample
Taken
2p.m.
2p.m.
4p.m.
4. p.m.
4p.m.
6 p.m.
6p.m.
1 1 a.m.
1 1 a.m.
2p.m.
2p.m.
4p.m.
4p.m.
4p.m.
6 p.m.
6p.m.
Date
Sample
Taken
11/21/77
11/21/77
11/21/77
11/21/77
11/21/77
11/21/77
11/21/77
1 1/22/77
1 1/22/77
1 1/22/77
1 1/22/77
1 1/22/77
1 1/22/77
1 1/22/77
1 1/22/77
1 1/22/77
CaS,
wt%
41.07
38.56
38.38
40. 05
1.30
42.83
38.10
41.07
6.40
42.59
18.45
43.48
25.77
1.21
37.91
21.88
CaSOa,
wt%
3.23
0.60
1.35
0.0
0.15
0.30
0.15
0.0
0.0
0.30
0.0
0.0
0.0
0.0
0.30
0.0
CaSOj,
wt%
0.46
0.61
0.2O
0.21
39.15
0.48
0.59
0.85
32.87
0.0
15.48
0.72
13.84
39.57
0.51
11.93
CaCOg,
wt%
0.89
3.65
0.0
3.06
21.15
0.28
4.68
0.69
20.67
0.57
20.94
1.06
18.91
19.15
2.18
17.07
CaO,
wt%
26.49
23.67
25.51
23.10
4.6O
23.25
23.72
26.74
8.63
25.72
10.32
22.45
8.41
4.43
24.86
12.13
AW*
wt%
2.70
2.88
2.0O
2.64
0.51
2.44
2.60
2.92
1.31
3.20
1.94
2.86
1.95
0.44
2.68
1.86
FeO,
wt%
2.74
2.63
1.83
2.51
0.89
2.08
2.26
2.53
1.61
2.89
1.81
2.53
1.76
0.73
2.27
1.59
MgO, Free Carbon,
wt% wt%
0.93
0.75
0.67
0.79
0.30
0.72
0.69
0.79
0.44
0.84
0.58
0.82
0.58
0.32
0.77
0.58
2.66
7.80
1.49
9.92
—
1.64
10.29
4.32
12.43
5.27
11.67
4.21
10.64
—
7.66
12.71
'Kiln discharge.
-------
Table 2.
Carbonation Test Results
Run
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
FGD Waste Slurry
Total
Percent Volume,
Solids
15
7.3
25
15
15
7.3
7.3
7.3
7.3
7.3
7.3
7.3
7.3
7.3
7.3
7.3
7.3
7.3
7.3
7.3
7.3
7.3
7.3
7.3
7.3
7.3
7.3
7.3
2
2
2.4
2
2
3
3
2
3
3
3
3
3
3
3
2
3
3
3
3
3
3
3
3
3
3
3
3
Reactor Conditions
Temp, Press.,
°F psig
260
250
250
250
250
250
250
250
250
250
250
250
250
250
250
250
250
275
250
250
200
200
250
150
200
200
250
250
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
60
74
60
60
42
42
60
35
42
24
60
60
C02
Feed Rate,
/min
1.0
1.0
1.0
0.54
0.54
0.54
0.54
0.54
0.54
1.0
0.54
1.0
0.54
1.0
1.5
1.0
1.0
1.5
1.0
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
Maximum W^S
Concentration,
%
74.5
46.4
55.9
57.5
15.6
81.5
44.4
61.3
44.5
39
96.9
93.9
97.8
88.6
83.8
92.1
88.7
83.8
90.7
97.9
97.9
98.2
—
96.8
97.4
93.9
98.1
99.3
Figure 2. Kel-S carbonation bench-scale unit.
4
The two-step process consists of (see
Figure 3):
(1) Dissolving calcium suitide by add-
ing hydrogen sulfide, thus forming
soluble calcium hydrosulfide.
(2) Removing the insoluble ash by fil-
tering or centrifugating.
(3) Precipitating the calcium as cal-
cium carbonate from solution by
adding carbon dioxide.
Kellogg conducted several bench-
scale semi-batch (i.e., continuous gas
flow, batch liquid/solids) tests which
demonstrated the calcium dissolution
and calcium carbonation steps. The
maximum percent hydrogen sulf ide gen-
erated in the carbonation step ranged
from 86-98 percent hydrogen sulfide,
which is ideal for sulfur plant feed. The
precipitated calcium carbonate was ex-
tremely pure (99.37 percent), and should
be excellent recycle to a limestone
scrubbing unit.
The two-step intermediate product (ash
and dissolved calcium/sulfur values)
was sent to the Bird Machine Company,
South Walpole, MA, for centrifuge and
filter tests. Both tests indicated better
calcium recovery and less calcium rejec-
tion than the results in the air flotation
tests.
Economic Analyses
Using the data from these tests, Kellogg
did a process evaluation to compare the
relative costs of solids removal equip-
ment for the one- and two-step processes
(see Table 3). Although the air flotation
units have the lowest capital cost, the
penalties of lower separation efficiency
and potentially polluting additives made
air flotation less attractive than filters or
centrifuges. For this application, centri-
fuges were preferable (over filters) be-
cause of lower capital costs. As a result
of the evaluation, the pilot plant design
was changed to use the two-step process
and to incorporate a centrifuge as the
ash separation device.
Although the results of the study indi-
cated that the process was technically
feasible, unforeseen problems early in
the study led to delays and additional ex-
pense, resulting in a large projected cost
overrun under the contract. To assist in
the decision of whether to continue the
study, EPA requested the Tennessee
Valley Authority (TVA) to conduct a pre-
liminary economic evaluation of the Kel-S
process. TVA based the evaluation on
design and economic premises developed
jointly by EPA and TVA to evaluate a
variety of FGD processes. Some of the
-------
Tables.
Cost Comparison of Air Flotation Versus Centrifugation/Filtration for the Kei-S Process
Raw Material Costs
Approximate Percent Percent Makeup
Capital Cost, Feed CaCO3 Feed Ash CaCOg, Additives, Total,
Method
One-Step
Air Flotation
(Using results
from Denver
Equipment
Experiment 1 7)
Air Flotation
(Using results
from Denver
Equipment
Experiment 9)
Two-Step
Centrifuge
Centrifuge
(washed)
Centrifuge
Centrifuge
(washed)
Filter
Filter
(washed)
$1000
65
110
200
400
200
400
900
900
Recovered
76
74
77
86
74
83
79
90
Rejected
52
62
92
84
97
93
100
100
$1000/yr
414
455
402
243
455
288
363
172
$ 1000/yr
307
643
0
0
33
33
0
0
$ 1000/yr
721
1098
402
243
488
321
363
172
Notes
Additives:
Pine Oil, Fuel
Oil, Aeropromoter
845
Additives:
Aeropromoter
839 and 845,
Accoalfroth
S-4005
Cake reslurried
and recentrifuged
Additive: Percol
7250.2fb/t
Additive: Percol
725 0.2 Ib/t, cake
reslurried and
recentrifuged
Cake washed
on filter
main premises were (1) the FGD system
was to be installed on a new 500-MW
boiler burning 3.5 percent sulfur, 16 per-
cent ash coal; (2) the FGD system was
designed to control SC>2 emissions to
1.2 lb/106 Btu heat input; (3) capital
costs were projected to mid-1979, rep-
resenting a mid-1977 construction start
and a mid-1980 completion; and (4) an-
nual revenue requirements (operating
costs + capital charges) were based on
a 7000 hr/yr first year operation and
were projected to mid-1980. Results of
the evaluation are summarized in Table 4.
The TVA evaluation indicated.that a
limestone scrubbing Kel-S process was
almost 50 percent higher in capital costs
than a typical limestone scrubbing pro-
cess with ponding of the waste; the Kel-S
annual revenue requirements were
about 80 percent higher than ponding
the waste. yVhen compared to the Well-
man-Lord/Allied Chemical process,
which also produces elemental sulfur,
the Kel-S process fared somewhat bet-
ter, but not enough to make it competi-
TaUe 4. Comparative Costs of Kel-S Versus Competing FGD Processes
FGD Processes
Costs, $1000
Total Capital
Investment
Total Annual
Revenue
Requirements
Limestone FGD
+ Ponding
48, 728
14, 102
Limestone FGD
+ Kel-S
73,023
26,065'
Wellman-Lord/
Allied Chemical
56,295
21,982
"By-product sulfur sales credit excluded; at $ 60/long ton, this would be about $ 2 million/
tive (see Table 4). For this reason, EPA
withdrew financial support from the
contract.
-------
Conclusions
Ail steps in Phases I and II, except the
kiln run to produce feed material for the
pilot plant, were completed. The FGD
waste sludge was reduced satisfactorily
in all three kiln test runs.
Carbonation tests conducted with the
bench scale reactor show that the rate of
conversion of calcium sulfide to calcium
carbonate and hydrogen sulfide is rapid,
and that high purity levels of hydrogen
sulfide in the effluent gas are achievable.
Flotation, filtration, and centrif ugation
experiments, environment considerations,
and analysis of equipment costs indi-
cated that the two-step Kel-S process
using centrifugation was preferable.
The objectives of Phases I and II were
successfully demonstrated. On this basis,
it would have been technically sound to
proceed with erection and testing of the
pilot plant. However, an economic eval-
uation of the Kel-S process indicated
that it probably could not compete with
other FGO processes.
Metric Conversion Factors
To Convert
from: To: Multiply by:
Btu
°F
ft
gal
in.
Ib
psig
J
°C
cm
1
cm
kg
kP
1,055.1
%(°F-32)
30.48
3.785
2.54
0.454
6.895
HtSto
claus
XHiS + C
"*
„ _ *
CO2 ..Carbonator #7
Carbonator
#2
0 CO,
scrubber
H20
Two-step
30-40 Wt %
solids
20Wt% solids
Cleanup
reactor
Ash. Ca (HS)2
Ash, CaCOa
Dispose/
H2Sto
claus
Air
flotation
additives
Disposal
Ash-rich
One-step
Figure 3. Process comparison, one- and two-step processes.
-------
R. P. Arganbright. P. Huang, G. S. Banner, B. G. Mandelik, T. S. Roche, and P. V.
Smith are with Pullman Kellogg, Industrial Park Ten, Houston, TX 77084.
Julian W. Jones is the EPA Project Officer (see below).
The complete report, entitled "Recovery of Calcium Carbonate and Sulfur
from FGD Scrubber Waste,"(Order No. PB 82-227 729; Cost: $13.50.
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:
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
•US GPO:1M2-S5*-092-4S2
-------
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
------- |