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

-------