oEPA
         United States
         Environmental Protection
         Agency
         Industrial Environmental Research EPA-600/7-79-092
         Laboratory         March 1979
         Research Triangle Park NC 27711
Proceedings:  Industry
Briefing on EPA
Lime/Limestone Wet
Scrubbing Test Programs
(August 1978)
         Interagency
         Energy/Environment
         R&D Program Report

-------
                   RESEARCH REPORTING SERIES


 Research reports of the Office of Research and Development, U.S. Environmental
 Protection Agency, have been grouped into nine series. These nine broad cate-
 gories were established to facilitate further development and application of en-
 vironmental  technology. Elimination of traditional grouping was  consciously
 planned to foster technology transfer and a maximum interface in related fields.
 The nine series are:

     1.  Environmental Health Effects Research

     2.  Environmental Protection Technology

     3.  Ecological Research

     4.  Environmental Monitoring

     5. Socioeconomic Environmental Studies

     6. Scientific and Technical Assessment Reports (STAR)

     7. Interagency Energy-Environment Research and Development

    8. "Special" Reports

    9. Miscellaneous Reports

 This report has been assigned to the INTERAGENCY  ENERGY-ENVIRONMENT
 RESEARCH AND DEVELOPMENT series. Reports in  this series result from the
 effort funded under the 17-agency Federal Energy/Environment Research and
 Development Program. These studies relate to EPA's mission to protect the public
 health and welfare from adverse effects of pollutants associated with energy sys-
 tems. The goal of the Program  is to assure the rapid development of domestic
 energy  supplies in an environmentally-compatible manner by providing the nec-
 essary environmental data and control technology. Investigations include analy-
 ses of the transport of energy-related pollutants and their health and ecological
 effects;  assessments of, and development of, control technologies for energy
 systems; and integrated assessments of a wide-range of energy-related environ-
 mental issues.
                        EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for  publication. Approval does  not signify that the contents necessarily reflect
the  views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.

This document is available to the public through the National Technical Informa-
?;?n Service. Springfield, Virginia 22161.

-------
                                    EPA-600/7-79-092

                                           March 1979
Proceedings:  Industry Briefing  on EPA
     Lime/Limestone Wet Scrubbing
                Test Programs
                (August 1978)
               John E. Williams, Conference Chairman
                  Program Element No. INE624A
              Industrial Environmental Research Laboratory
               Office of Energy, Minerals, and Industry
                 Research Triangle Park, NC 27711
                      Prepared for

              U.S. ENVIRONMENTAL PROTECTION AGENCY
                Office of Research and Development
                    Washington, DC 20460

-------
                                     PREFACE
      More than half of all "man-made" sulfur dioxide  (S0?)  is emitted  by
 electric power plants, and the use of sulfur-containing fossil fuels,  especially
 coal, to generate electricity is expected to increase dramatically  in  the  next
 10 years. Therefore,  the develoment and commercial application of SO9  control
 technologies  is one of the most important concerns of the  U.S.  Environmental
 Protection Agency (EPA).   Flue gas desulfurization (FGD)  is the most promising
 technique for control of SCL  that will be available for widespread  application
 to fossil fuel-fired  electric electric power plants for at  least the next  8  to
 ]0 years.

      The Industrial Environmental Research Laboratory - Research Triangle  Park
 (IERL-RTP)  of EPA's Office of Research and Development periodically sponsors
 symposia and  Industry Briefing Conferences for  the transfer of information
 regarding FGD research,  development and application activities with the objective
 of further accelerating  the development and commercialization of this  technology.
 One of the major  IERL-RTP  FGD efforts for the past several  years has been
 advancement of the  technology for lime/limestone  wet  scrubbing.  The focal
 point of  this program has  been the prototype testing  at EPA's Alkali Wet Scrubbing
 Test  Facility,  located at  TVA's Shawnee Steam Plant,  near Paducah,  Kentucky.
 Current  emphasis  of the  test  program at Shawnee is to optimize  lime and limestone
 systems  in the  areas  of  improved sludge disposal,  performance reliability, and
 process  economics.

      The  August ]978  Industry Briefing Conference  focused primarily on recent
 test  results  at Shawnee  in which the  predominantly calcium  sulfite  reaction
 products  were  forced  oxidized to calcium sulfate  (gypsum).   Potential advantages
 of  forced oxidation were also discussed.   Other material presented  during  the
 Conference  included:   IERL-RTP  in-house pilot plant results,  which  have contributed
 significantly  to  an understanding  of  the  lime/limestone system chemistry and in
 supporting  the  Shawnee program;  a  review  of  the IERL-RTP effort  in  the waste
 solids disposal area,  including  discussion  of waste disposal  economic options;
 and a review  of future plans,  including testing at  TVA's Widow's Creek Plant
 and the EPRI  test programs.

     More  than  125  people  representing  electric utilities,  process  suppliers,
and State and Federal  regulatory agencies,  attended the Conference.   The Chairman
of the August 1978  Industry Briefing Conference was John E. Williams,  a Chemical
Engineer  in the Emissions/Effluent  Technology Branch,  IERL-RTP.

     These Proceedings are  comprised of copies of  the  participating authors'
papers as received.  As supplies permit,  copies of  the Proceedings  are available
free of charge and may be  obtained  by contacting IERL-RTP's Technical  Information
Coordinator, Environmental Protection Agency, Research Triangle Park,  North
Carolina  27711.
                                        ii

-------
                                CONTENTS
                                                                 Page

SIGNIFICANT EPA/IERL-RTP PILOT PLANT RESULTS
     Robert H. Borgwardt 	   1

RESULTS OF LIME AND LIMESTONE TESTING WITH FORCED
OXIDATION AT THE EPA ALKALI SCRUBBING TEST                         <•
FACILITY - SECOND REPORT
     Harlan N. Head	10

SELECTED TOPICS FROM SHAWNEE TEST FACILITY
OPERATION
     David T. Rabb	54

STATUS REPORT OF SHAWNEE COCURRENT AND DOWA
SCRUBBER PROJECTS AND WIDOWS CREEK FORCED
OXIDATION
     James L. Crowe, Gerald A. Hollinden and Thomas M. Morasky  .   72

CURRENT STATUS OF DEVELOPMENT OF THE SHAWNEE LIME-
LIMESTONE COMPUTER PROGRAM
     C. David Stephenson and Robert L. Torstrick  	   94

LANDFILL AND PONDING CONCEPTS FOR FGD SLUDGE
DISPOSAL
     Jerome Rossoff, Paul P. Leo and Richard B. Fling  	 140

COMPARATIVE ECONOMICS OF FGD WASTE DISPOSAL
     J. Wayne Barrier	153
                                   iii

-------
                  SIGNIFICANT  EPA/IERL-RTP  PILOT  PLANT RESULTS

                                      by

                              Robert  H. Borgwardt

             For  Presentation at  EPA  Industry Briefing Conference

                               August 29,  1978



                                  INTRODUCTION

     The last review of  IERL-RTP  pilot plant testing was made at the FGD
Symposium  in November, 1977.  It  discussed the effect of forced oxidation on
the performance of a single-loop  limestone scrubber, showing that complete
oxidation  of the  slurry  could be  accomplished at normal pH  (to at least as
high as 6.5) without adversely affecting S02 removal efficiency or scaling
potential.  Because of its distinct advantages with respect to sludge quality
and operating simplicity, this scrubbing configuration has  continued to be the
focus of testing  at RTP.  Recent  work with the single-loop  limestone scrubber
has involved two  main areas of investigation:  1) the use of adipic acid as an
additive for improving S02 removal efficiencies while forcing oxidation, and
2) the replacement of makeup  water with simulated cooling tower blowdown as a
means of further  reducing the fresh water  requirements for  FGD and improving
the overall water management  in a power plant.  This presentation reviews
progress toward those objectives.

Adipic Acid

     This  prospective scrubber additive is a solid, straight-chain dicarboxylic
acid:  HOOC(CH2)4COOH.   It was obtained in 50 Ib bags at a  cost of 46c/lb
for the tests at  RTP.  Bulk shipments are quoted at 41c/lb.

     The tests were undertaken as a result of theoretical analyses carried out
by G. Rochelle ("The Effect of Additives on Mass Transfer in CaCO» and CaO
Slurry Scrubbing  of S02  from Waste Gases," Ind. Eng. Chem., Fundam. 16,
pp. 67-75, 1977) which indicated  that adipic acid should act as a buffer to
limit the  drop in pH that normally occurs at the gas/liquid interface during
S02 absorption.  The additional capacity of the surface film for SO  absorption
brought about by  this buffering action is expected to enhance the liquid-phase
mass transfer and improve the overall SO  removal efficiency of a limestone or
lime scrubber of  a given type operating at a given L/G. The relatively low cost
of adipic  acid, together with its particularly favorable ionization constant,
make it a  prime candidate for testing as a buffering additive.

     Further analysis by Rochelle ("Process Alternatives for Stack Gas Desulfur-
ization by Throwaway Scrubbing,"  Proceedings of 2nd Pacific Chem. Eng. Congress,
Vol. I, p. 264, August 1977)  showed that additives will be most effective,
cost-wise, when used in  scrubbers employing forced oxidation.  This approach
would minimize the loss of the additive—and thus reduce the amount required
for makeup—because of the tighter loop resulting from the better dewatering
properties of oxidized sludge.  For similar reasons, additive losses will also
be minimized when fly ash is  collected dry rather than collected in the scrubber.
A fly ash-free scrubber producing oxidized sludge should achieve the greatest
SO,., removal efficiency for a  given amount of additive.

-------
     Adipic acid has several potential advantages over other additives, such
as MgO, that function by increasing the dissolved alkalinity (primarily as
soluble sulfites).  First, since adipic acid functions by a different reaction
mechanism which does not include the sulfite/bisulfite equilibrium it is not
affected by oxidation of the sulfite in the scrubbing liquor.  It is therefore
particularly well suited for use in a single-loop scrubber employing forced
oxidation in the hold tank.  Second, the buffering mechanism by which adipic
acid enhances SO  absorption is not affected by the presence of chloride:  the
effectiveness of the alkali additives is reduced in proportion to the concen-
tration of chloride.  The  lack of interference by chloride thus favors the use
of adipic acid in systems  employing forced oxidation since these systems
concentrate chloride to much higher levels in the scrubbing  loop.  This feature
also has importance with regard to the possible application  of filter washing
to remove the soluble salts from sludge—washing will also tend to concentrate
chloride in the scrubber.

     A third important advantage of adipic acid is indicated by initial cost
comparisons made by Bechtel.  Assuming that no additive  losses occur in the
solids, the effectiveness  of adipic acid is expected to  be sufficiently greater
than that of MgO to more than compensate for its higher  cost per pound  (bulk
MgO =  llc/lb):  the cost of obtaining a given degree of  improvement  in  S02
removal should be lower for adipic acid.

     The purpose of the IERL-RTP tests was to verify the postulated  effect  of
adipic acid on SO- removal efficiency in a single loop limestone scrubber and
to determine whether any deleterious effects are associated  with its use,
particularly with regard to sludge properties and oxidation  efficiency.

Water  Reuse

     The other area of testing at  IERL-RTP involved  the  substitution of a
simulated cooling tower blowdown for the makeup water.   The  successful applica-
tion of forced oxidation within the  scrubbing loop is  expected  to  facilitate
the improvement of overall water reuse  in a power plant  by making  possible  the
use of waste water from the boiler  and/or cooling tower  as  FGD  scrubber feed.
The basis for  this assumption  is that the high  density of pure  gypsum  in the
oxidized slurry can rapidly dissipate the supersaturation occurring when the
extra  sulfate  present  in  the blowdown  (as Na2SO^)  is fed to  the scrubber.

     The tests are ultimately  aimed at  the development of a scheme whereby  a
water  treatment process,  such  as vapor  compression evaporation,  is incorporated
within the  FGD scrubber  system.  The soluble  salts,  including Na-SO, and
calcium chloride, would  thus  be  simultaneously  extracted from the  scrubbing
loop while  regenerating  enough fresh water  for  mist  eliminator  washing.  This
approach has  the  potential of  eliminating  soluble  salts  from the waste sludge
and  improving  the energy  efficiency of  water  treatment,  while reducing the
overall fresh  water  requirements.   When combined with the use of adipic acid,
such a system would  have  the  further advantage  of  maximizing,  the  SO-  removal
that can be achieved  with a given  amount  of  additive.   By also incorporating

-------
filter cake washing and additive recycle, the elimination of additive makeup
is conceivable since the relatively insoluble adipic acid could be separated
from the soluble salts during evaporation.  The tests reported here are a
first step in evaluating the feasibility of this concept.
                                    RESULTS

     Using the scrubbing configuration indicated by Figure 1, tests were made
with two absorber types, a TCA and a multigrid (simulating a spray tower),
operating at L/G = 65 gal/103cf  (8.7 liters/m3).  Adipic acid was fed with the
limestone.  Figure 2 summarizes  the effect of adipic acid on the SO- removal
efficiency in the IERL-RTP pilot plant.  At an adipic acid level of about
1600 ppm, the SO^ removal efficiency of the TCA was increased from 82 to
93 percent.  As indicated by Table 1, the limestone utilization was also
significantly better than that obtained without adipic acid when compared at
similar feed stoichiometry.                                            «•

     As expected, the S02 removal efficiency obtained with high chloride
concentrations in the scrubbing  liquor was not significantly different from
that obtained without chloride when operating at similar levels of adipic
acid.

     The improvement in SO- removal could be maintained at the 85 percent
limestone utilization level required for minimum mist eliminator fouling.
                                  _0
     An oxidation rate of 1.3 x  10   g mol/1 (min) was maintained in the
system at pH 5.9 using an unstirred, air-sparged tower of 18 ft (5.5 m)
slurry depth.  No significant reduction in oxygen transfer efficiency was
observed at adipic acid levels up to 5000 ppm.
                TABLE 1.  Effect of Adipic Acid on TCA Scrubber
                          Performance at Constant Limestone Feed
                          Stoichiometry
     SO- Removal, %
     Limestone Utilization, mol %
     Chloride Concentration, ppm
     Scrubber feed pH
     Scrubber effluent pH
     Oxidation3, mol %
     Settling rate, cm/min
     Limestone stoichiometry^3
     Adipic acid cone., ppm
No Additive

   82
   77
   20,000
   6.1
   5.1
   99
   3.4
   1.07
   0
Adipic Acid
   Added

   93
   91
   17,000
   5.9
   4.9
   99
   2.7
   1.01
   1570
     a
      Air stoichiometry = 2.7 g atoms oxygen/g mol SO- absorbed

      mol CaCO- fed per mol SO  fed to scrubber

-------
      The physical properties of the slurry that was oxidized in the presence
 of adipic acid were not noticeably different from those of fully oxidized slurries
 obtained in tests without adipic acid:   it settled at 2.5 cm/min and filtered to
 76 percent solids.

      A comparison of the pressure drop  required to obtain a given S0? removal
 in the pilot scrubber is shown in Figure 3 for single-loop systems operating
 with forced oxidation.   It indicates that the improved mass transfer resulting
 from adipic acid addition may permit a  given S02 removal to be attained with
 substantially lower pressure drop than  that required without the additive.

      The only adverse effect of adipic  acid noted at IERL-RTP was a disagreeable
 odor arising from the open tanks.   It was present during all tests with this
 additive and its cause  is unknown.   The adipic acid feed was odorless.

      The tests with sodium sulfate  addition were also made in the scrubbing
 configuration of Figure 1 replacing the makeup water with a solution containing
 66 Ib Na2S04/100 gal.*  (79 g/1).   The TCA tower was operated at L/G = 80 gal/
 106cf (10.7 liters/m3),  3000 ppm inlet  S02,  and 6 percent oxygen.   Chloride was fed
 to the system as HC1  gas to  maintain an average of 13,000 ppm Cl in the scrub-
 bing liquor.

      The principal  objective of the tests was to assess the effect of sodium
 sulfate addition on the scaling potential.   When CaCO- dissolves in the tower,
 the gypsum saturation differential  across it is increased by the presence of
 high sulfate concentrations  in the  scrubbing liquor.

      The results of eight  consecutive 90-hr  runs in this  mode overaged  1.13%
 relative sulfate saturation  in the  scrubber  feed liquor using 12 min.  total
 EHT residence  time,  including 5.6 min.  in the oxidizer.   With 15,000 ppm
 average sodium concentration in the scrubbing liquor,  sulfate scale was consis-
 tently observed on  the  bottom TCA grid  when  the S02 make-per-pass  exceeded
 8  m mol/1.   At  lower  make-per-pass  (e.g.,  with 2500 ppm inlet SC- ) scaling  did
 not  occur.

      SC^  removal  efficiencies  obtained  in the IERL-RTP pilot plant were as
 good  as,  or  better  than, those  obtained with fresh  water  makeup.   As indicated
by  the  comparison in  Table 2,  higher  SO-  removal  and  limestone  utilization
 resulted  when  the fresh water was replaced by sodium  sulfate solution.   When
 compared  to  the  results obtained without  forced oxidation (and  fresh water
makeup) the  improvement  in SO-  removal was about  10 percent.
     *The makeup solution was saturated with gypsum by adding oxidized sludge.

-------
              TABLE 2.  Effect of  Sodium Sulfate on TCA Scrubber
                        Performance at Constant Limestone Feed
                        Stoichiometrv
                                                  Fresh Water
                                                    Makeup

                                                     86
                                                     22
                                                     84
                                                     17,000
                                                     0
                                                     6.0
                                                     5.0
                                                     99
                                                     2.5
                                                     73
                                                     1.40
Na2SO
tion
                                                                   Solu-
                                                             "'" ~ Makeup
SO  removal, %
Scrubber 'P, cm HO
Limestone utilization, mol %
Cl~ cone.,  ppm
Na+ cone.,  ppm
Scrubber feed pH
Scrubber effluent pH
Oxidation,  mol %
Settling rate, cm/min
Filter cake solids, %
Limestone feed stoichiometry
   88
   18
   91
   19,000
   26,000
   6.3
   5.4
   99
   3.0
   85
   1.36
     Sludge oxidized in this mode was of excellent quality, consistently
filtering to 85 percent solids.  It contained about 2 mg of sodium per gram of
dry solid after washing with acetone.  About 0.3 mg of this sodium was non-
leachable by water washing.

     An overall oxidation rate of 1 x 10~3 g mol/1 (min) could be maintained
in the pilot plant oxidizer at pH 7 when operating with 13,000 ppm Cl~ and
15,000 ppm Na+ in the scrubbing liquor.

     The tentative conclusion from these results is that an FGD system can be
operated with full replacement of the makeup water by cooling tower blowdown
when the scrubber is designed for an SO  make-per-pass below 8 m mol/1.  The
combination of adipic acid, forced oxidation, and water treatment appears
feasible for spray towers even in high-sulfur coal situations to maximize
water reuse and eliminate the uncontrolled discharge of soluble salts.

     Adipic acid has clear merits as an additive for improving SO  removal
efficiencies of scrubbers using forced oxidation.
                                 FUTURE PLANS

     A limestone "type and grind" test program will commence at RTP next
month.  Four different limestone types, selected by Bechtel to give a wide
range of expected reactivities, will be tested at two grinds—75 percent less
than 200 mesh (coarse) and 90 percent less than 325 mesh  (fine).  Eleven tests
are planned for the four limestones over a 22 week period to compare the
limestone stoichiometries required for a given SO  removal efficiency in the
IERL-RTP scrubber.  Laboratory characterization of the physical and chemical
properties of the limestones will be made elsewhere in detail.

-------
     The prospects for improving sludge quality without forced oxidation will
be evaluated in the IERL-RTP pilot plant according to a test program devised
by Radian Corporation ("Development of a Mathematical Basis for Relating
Sludge Properties to FGD-Scrubber Operating Variables," J. L. Phillips et al.,
EPA-600/7-78-072, NT1S No. PB 281582/AS).  This test program will use a a
model based on CaSO '1/2 H-0 crystallization kinetics to select and evaluate
scrubber modifications that might lead to the growth of larger crystals of
calcium sulfite.  By increasing the crystal size, it is expected that a faster
settling and more filterable sludge may be produced.  These tests will be
undertaken after the type and grind study.

     Another additive, sodium thiosulfate, will be evaluated in the lime
scrubber as a possible oxidation inhibitor.  Thiosulfate has been tentatively
identified by Radian as an impurity in carbide lime that may be responsible
for the low oxidations observed at LG&E.  If small amounts are effective in
reducing oxidation when added to commercial lime, the propects for unsaturated
operation should be improved.

     One of the two scrubbers at IERL-RTP will be converted to a sodium-based
dual alkali system later this year.  This scrubber will be used to provide
experimental support for EPA's full-scale dual alkali demonstration facility
at LG&E's Cane Run Station.

-------
3000 ppm S02
        *
6 percent
  MAKEUP
           o°oO
           000°0
           o°o°o
           o°o°o
  WATER




^









fc^^^^^*"**^^^
lift
(5.5 m)
O
0
e °
0
o
e
o
•
O
o
e
OX1DIZER
O
0 °
0
0 °
o o
o
e
o
0
;








LIMESTONE


i


* ob
AIR
      Figure 1.  IERL-RTP scrubber configuration for adipic acid and Na2
      SO 4 tests.

-------
  100
                       TCA
   90
80
                                               MULTIGRID
                                              AP = 5 cm H20
o
ui
M
   70
   60
   50
                       50
                                     1000               1500

                                 ADIPIC ACID CONCENTRATION, ppm
                                                                            2000
2500
          Figure 2.  Effect of adipic acid on S02 absorption in TCA and multigrid scrubbers at
          L/G = 65 gal/Mcf (8.7 l/rr.3).

-------
   100
    90
i
<
>
o
K

CM
80
                      FORCED OXIDATION

                       WITH ADIPIC ACID
                                                                    FORCED OXIDATION

                                                                   WITHOUT ADIPIC ACID
    70
    60
                            10
                                    15         20         25



                                       SCRUBBER AP, cm H20
30
35
40
         Figure 3  Effect of adipic acid on the pressure drop required to attain a given S02 removal

         efficiency in the IERL-RTP limestone scrubber.

-------
  RESULTS OF LIME AND LIMESTONE TESTING

         WITH FORCED OXIDATION

AT THE EPA ALKALI SCRUBBING TEST FACILITY

            - SECOND REPORT -
               Presented by
            Dr. Harlan N.  Head
   Project Manager, Shawnee Test Facility
           Bechtel  National, Inc.
               50 Beale St.
       San Francisco, California 94119
                  at the

           EPA Industry Briefing
    Research Triangle Park, North Carolina
              August 29, 1978
          EPA Contract 68-02-1814
              John E. Williams
               Project Officer
  Industrial Environmental  Researh Laboratory
     Office of Research and Development
  Research Triangle Park, North Carolina 27711
                10

-------
                            ACKNOWLEDGEMENT


This paper was prepared  as  a team  effort by the following Bechtel personnel



                  Dr.  Harlan N. Head, Project Manager


             D.A.  Burbank,  Jr.                 D.Y. Kawahara
             G.A.  Oallabetta                  T.M. Martin
             C.L.  OaMassa                      D.T. Rabb
             D.G.  Derasary                     C.H. Rowland
             J-  Hing                           Dr. S.C. Wang
                                 11

-------
                                 Section 1


                                INTRODUCTION
In lime and limestone wet-scrubbing systems for removing  S0£  and  participate
from coal-fired boiler flue gas, disposal  of the waste solids product  has been
a major problem both technically and economically.   This  report addresses the
results of testing at the EPA Alkali Scrubbing Test Facility  to develop commer-
cially feasible forced oxidation procedures for reducing  the  volume and improving
the disposal characteristics of the waste  solids product.

The waste solids consist primarily of calcium sulfite, calcium sulfate (gypsum),
and fly ash.  The relative amounts of sulfite and sulfate depend  on the degree of
oxidation in the scrubbing system.  In most medium-to-high sulfur coal applica-
tions, natural oxidation of sulfite to sulfate in the scrubber system  amounts to
only 10 to 30 percent and calcium sulfite  is the predominant  material  in the waste
sludge.

Calcium sulfite wastes present a serious disposal  problem because of the diffi-
culty of dewatering.  The slurry can be dewatered only to about 50 to  60 percent
solids, producing an unstable, thixotropic material  unsuitable for landfill.
Where space is available, ponding of the untreated sulfite sludge has  been  prac-
ticed.  But the pond area may be impossible to reclaim, and in many locations
sufficient space is not available.

Three procedures have been considered for  converting the  sulfite  wastes to  material
suitable for landfill:


     •  Commercial fixation with additives

     •  Blending sulfite sludge with fly ash

     •  Forced oxidation of the calcium sulfite to a more tractable calcium
        sulfate (gypsum), which is easily  dewatered to greater than 80 percent
        solids


Of these 3 procedures, preliminary economic evaluations by TVA*)  have  shown
that forced oxidation is the most economical method to upgrade pond disposal to
landfill.  Cost information will be presented by TVA in a separate paper.   Further-
more, because of the superior dewatering properties, forced oxidation  results in
a smaller volume of waste solids.

                                   12

-------
 In Japan,  where  natural  gypsum  is not available, forced oxidation in scrubber systems
 has been employed  extensively to produce a high-quality q.ypsum raw material  for the
 cement  and wall board  industries.  In the United States, scrubber gypsum may  be un-
 able to compete  extensively with the widely available natural qypsum.  Thus, the
 incentive  in the United  States  has been to develop simplified forced-oxidation pro-
 cedures directed only toward improving waste solids handling and disposal  properties.
 As a disposal material,  the gypsum sludge can have high fly ash content; moreover
 the oxidation reaction need be  carried only to about 95 percent completion.

 Beginning  in 1976, studies conducted by EPA with the O.I MW pilot plant at the
 Industrial  Environmental Research Laboratory located at Research Triangle  Park
 North Carolina (lERL-RTP)Z), have shown that calcium sulfite can be readily  oxidized
 to gypsum  by simple air/slurry  contact in the hold tank of the scrubber recirculation
 loop.   Although the rate of oxidation reaches a maximum at a pH of 4.5  and then
 declines at higher pH, it was found that oxidation could be accomplished at  a prac-
 tical rate up to a pH of about  6.0.

 Based on the findings at the IERL-RTP pilot plant, a program was set up at the
 Shawnee Test Facility located at the TVA Shawnee Steam Plant near Paducah, Kentucky
 to develop procedures for forced oxidation.  Forced oxidation testing was  initiated'
 in January 1977 on the 10 MW EPA prototype scrubbers and has continued  since as the
 major part  of the Shawnee Advanced Test Program.  Results of forced oxidation test-
 ing at  the Shawnee facility from January through September 1977 were reported at
 the FGD Symposium in Hollywood, Florida in November 1977. 3)  This paper is an update
 on that report.  It includes the results of forced oxidation testing through June
Systems successfully demonstrated during this period are:

     •  Forced oxidation in the first of two scrubber loops using lime  slurry
        limestone slurry, and limestone slurry with added  magnesium oxide

     t  Forced oxidation within a single scrubber loop using limestone  slurrv

     •  Forced oxidation of a scrubber bleed stream using  limestone slurrv with
        added magnesium oxide
THE TEST FACILITY


There are two scrubber systems operating at  the F.PA sponsored  Shawnee Test Facility
each with its own independent slurry handling facilities.   Roth  systems were tested'
with forced oxidation.  The systems  have the following  scrubbers:

     t  A venturi followed by a spray tower  (venturi/spray tower)
        (35,000 acfm capacity o 3000F)

     •  A Turbulent Contact Absorber (TCA)
        (30,000 acfm capacity P 300°F)
                                    13

-------
The scrubbers receive flue gas from TVA Shawnee coal-fired  boiler No. in.   The
boiler normally burns a hiqh-to-medium sulfur bituminous  coal  producing S02 con-
centrations of 1500 to 4500 ppm.   Flue qas can be taken from either side of the
boiler No. 10 particulate removal  equipment,  allowing  testing  with high fly ash
loadings (3 to 6 grains/scf dry)  or low loadings (0.04 to 0.6  grains/scf dry).
Chlorides frorfi the flue gas concentrate in the scrubber slurry liquor over a
range of 1000 to 7000 ppm depending on the tightness of the scrubber water balance
and the chloride concentration in  the coal burned.


The Shawnee Test Facility has been operating  since March  1972.  Rechtel National,
Inc. of San Francisco is the major contractor and test director; TVA is the con-'
structor and test facility operator.   The inital  test  program  lasted through October
1974^) w-jt|r, tne major emphasis on  demonstrating reliable  operation.  The forced oxi-
dation tests are a part of an advanced test program that  is scheduled to continue
through December 1979.  Earlier results of the advanced test program are reported
elsewhere.5?6,7)


The Advanced Test Program schedule for the period covered in this report is shown
in Figure 1.  As can be seen, testing with forced oxidation has constituted the
major effort during this period.
                                   14

-------
ITEM
1. VENTUHI/SPRAY TOWEH SYSTEM
LIMESTONE TWO LOO* OXIDATION WITH FLY ASH
LIME TWO LOOP OXIDATION WITH FLY ASH
FLUE GAS CHARACTERIZATION
LIME TWO LOOT OXIDATION WITHOUT FLY ASH
LIMESTONE TWO LOOT OXIDATION WITHOUT FLY ASH
LIMESTONE TWO LOOT OXIDATION WITH MfO AND FLY ASH
LIMESTONE BLEED STREAM OXIDATION WITH M»O AND FLY ASH
Z. TCA SYSTEM
LIME WITH AND WITHOUT F LY ASH
FLUE GAS CHARACTERIZATION
LIMESTONE TYPE AND GRINO
LIMESTONE AUTOMATIC FEED CONTROL
LIMESTONE ONE LOOT OXIDATION WITH SPARGER
LIMESTONE OR LIME WITH MgO
LIMESTONE ONE LOOP OXIDATION WITH M|O AND SPARGER

1»77
JAN
















FEB








••••






MAR








1 	






APRIL







i
f
f
•

f

f
:
* f \
f
, f
f
MAY
f
t\
* f
t
JUNE










1M



•*-
JULY















	 SCH
MAI
AUG















iDULEDI
ITENANC
SEPT















3ILER
E OUTAGI
OCT
	 •
— • 	












FRI
BO
SHI
NOV
••^H












EZING PS
nt SYSTEI
IT DOWN
DEC
•• 	
__••









••


3BLEMS.-
S
1978
JAN
•::::



:::-::

•i"i:'

W|


••L;S



FEB

MM-


—







••^•H
—
MAR
















APRIL





1HBBB








BOILER I
FOR DUC
RE ROUT
• BOILER
OUTAGI
MAY





• 	








UTAGE-
T
NG
JUNE













—
•P •
FIGURE 1.   SHAWNEE ADVANCED PROGRAM TEST SCHEDULE SINCE JANUARY 1977

-------
                                  Section 2
    FORCED  OXIDATION WITH TWO SCRUBBER LOOPS ON THE VENTURI/SPRAY TOWER SYSTEM
 Forced oxidation with two scrubber loops in series has been successfully
 demonstrated in the venturi/spray tower system with three alkali  types:  limestone,
 lime, and limestone with added magnesium oxide.  In this arrangement,  the  flue
 gas  passes through two scrubbers in series, each with its own hold tank  and
 slurry recirculation loop.  The first loop is operated at a relatively low pH
 to provide favorable conditions for forced oxidation while the second  loop is
 operated at a higher pH for good S0£ removal.


 Commercialization of a two-loop scrubbing system with forced oxidation is  both
 feasible and desirable in situations where land is unavailable for ponding and
 the  higher S02 removal inherent with a two-loop system is required.  Scrubbers
 in series have already been installed commercially with the first  scrubber designed
 primarily for particulate removal  and the second primarily for S02 removal.  The
 addition of forced oxidation to such a system would be relatively  uncomplicated.


 Testing of the venturi/spray tower system in a two-scrubber-loop configuration
 with forced oxidation has been ongoing since January 1977.
SYSTEM DESCRIPTION
         ^


The venturi/spray tower system was modified for two-loop scrubber  operation with
forced oxidation as shown in Figure 2.   To separate the venturi  and  spray tower
scrubber loops, a catch funnel was installed beneath the bottom  spray  header of
the spray tower.  To eliminate slurry entrainment through the  catch  funnel, the
bottom spray header was turned upward.


The hold tank in the first scrubber (venturi) recirculation loop was used
as the oxidation tank.  The arrangement of this tank is shown  in Figure 3.  The
tank was 8 ft in diameter and could be operated at 10,  14, or  18-ft  slurry levels.
In early tests the tank contained an air sparger ring made of  straight 3-inch
316L SS pipe pieces welded into an octagon approximately 4 ft  in diameter.  It
was located 6 inches from the bottom of the tank.  Sparger rings had either
130 1/8-inch diameter holes or 40 1/4-inch diameter holes pointing downward.  The
                                     16

-------
REHEAT
                                                          COMPRESSED
                                                       VENT   AIR

                                                        I
                                                                  CLARIFIED LIQUOR
                                                                  OVERFLOW
                                                                     DESUPERSATURATION

                                                                           TANK
                                                                                              BLEED TO
                                                                                               SOLIDS
                                                                                             DEWATERING
                                                                                               SYSTEM
        FIGURE  2.    FLOW  DIAGRAM  FOR TWO-LOOP OXIDATION IN THE  VENTURI/SPRAY TOWER SYSTEM

-------
                         FIGURE 3
       ARRANGEMENT OF THE VENTURI/SPRAY TOWER
     OXIDATION TANK WITH AIR DISCHARGE THROUGH PIPE
                                          BAFFLE
                                            COMPRESSED AIR
           AGITATOR
                       OXIDATION TANK
                         PLAN VIEW
            COVER
           OUTLET
             18'
             14'    r
        ALTERNATIVE
          OUTLETS
3" DIAMETER PIPE
WITH AIR DISCHARGE
DOWNWARD
01  2345

  SCALE, FEET
                           L.
r
                                               INLET
          AGITATOR
          (56 rpm, 20 Hp)
                                         OXIDATON TANK
                                           BAFFLE
                                        COMPRESSED AIR
                      ELEVATION VIEW
                           18

-------
sparger ring was  fed  with  compressed  air to which sufficient water was added to
assure humidification.   In more  recent  tests the sparger ring was replaced
by a 3-inch diameter  pipe  with an  open  elbow discharging air downward at the center
of the tank about 3  inches from  the tank bottom.


The oxidation tank had  an  agitator with two axial flow turbines, both pumping down-
ward.  Each turbine was 52 inches  in  diameter and contained 4 blades.  The bottom
turbine was 10 inches above the  air sparger.  The agitator rotated at 56 rpm and
was rated at 17  brake Hp.


A 10-ft diameter  desupersaturation tank, operating at a 5-ft slurry level, followed
the oxidation tank to provide time for  gypsum precipitation and to provide air-free
pump suction.


Provision was made to add  alkali to either scrubber loop.  Clarified liquor from
the dewatering system could be returned to either scrubber loop or to the mist
eliminator wash circuit.
SUMMARY OF PREVIOUSLY REPORTED  TEST RESULTS
Forced oxidation test results  with  two scrubber  loops  conducted from January
through mid-September 1977  with lime and limestone  slurry  have been previously
reported.3) These early tests  were  conducted  at  25,000 acfm  flue gas rate (300°F)
which corresponds to a superficial  gas velocity  in  the spray tower of 6.7 ft/sec.
A flue gas rate lower than  the maximum possible  in  the system (35,000 acfm) was
chosen to assure that high  S02 removal (greater  than 80 percent) could he achieved.
Slurry recirculation rates  of  600 qpm in the  venturi loop  and 1400 gpm in the
spray tower loop were used.  Each run averaged about 5 to  6  days which was judged
to be sufficient time to reach kinetic equilibrium  and to  allow adequate run data
to be gathered.


Key results from these earlier tests were as  follows:

      •   Oxidation of sulfite solids to gypsum  of  90  percent or better
          dramatically improved the dewatering and  handling  characteristics of
          the waste solids.

      •   Slurry oxidation  of better than 96  percent in the  first of two indepen-
          dent scrubbing loops was demonstrated  with simple  air sparging through a
          sparger ring in an open tank with the  configuration shoun in Figure 3.

      •   Conditions under which near complete oxidation was demonstrated were
          an oxidation tank pH range of 4.5 to 5.5, an air stoichiometric ratio



                                     19

-------
          of at least 1.5 atoms 0/mole of SC^ absorbed,  and an oxidation
          tank level of at least 14 feet.

      •   Slurries with high or low fly ash loadinqs oxidized  equally well.

      •   A slurry solids concentration of 7 percent or higher in  the spray tower
          was required to prevent calcium sulfite scaling and  to maintain qood
          S02 removal.

      •   For pH control, it was necessary to add lime to both scrubber loops.
          With limestone, addition to the spray tower loop was sufficient.
UPDATE ON TWO-SCRUBBER-LOOP TEST RESULTS WITH LIME  SLURRY
Since the last report, 7 runs have been made with lime slurry  in  a  two-scrubber-
loop configuration with forced oxidation.   Results of the tests are summarized  in
Table 1.  These tests further demonstrated the feasibility of  a two-scrubber-loop
forced-oxidation system and contributed more information  for commercial desiqn.
In all these runs, the filter cake solids  concentration was at least 30 percent
and usually above 85 percent.  The following describes design  factors developed
during these tests.


Air Sparger - Previous tests were made with air dispersed into the  oxidation
tank through a sparger ring containing either 1/8-inch or 1/4-inch  holes.  Some
plugging and erosion of the holes were experienced with these  rings.  Starting
with Run 861-1A the sparger ring was replaced with a  3-inch pipe  with an elbow  on
the end to direct the air downward at the  center of the tank about  3 inches above
the tank bottom and 1 foot below the bottom agitator  blade.  Oxidation efficiency
was as good with this 3-inch air pipe as with the spargers.  For  example, in
Run 861-1A with the air pipe, 98 percent sulfite oxidation was achieved at an
air stoichiometry of 1.5 atoms 0/mole S0£  absorbed.  Based on  the success with  the
open 3-inch pipe, it was concluded that the agitator  plays a primary role in dis-
persing the air.  The agitator used in these tests had two sets of  axial flow tur-
bine blades pumping downward and operated  at a fixed  speed of  56  rpm with 17 brake
Hp.  A variable speed agitator, presently  on order, will  be installed so that
information can be obtained on agitator power versus  oxidation efficiency.


Slurry solids concentration - Virtually all  the fly ash is captured in the slurry
in the venturi loop so the spray tower slurry is essentially fly  ash free.  The
venturi loop is normally controlled at a higher slurry solids  concentration than
the spray tower loop to compensate for the fly ash.


Slurry solids concentration in the venturi  scrubber loop  was controlled at 15
weight percent in a majority of the lime runs.  Slurry solids  concentration in  the
spray tower varied from 6 percent to almost 20 percent depending  on whether the
                                    20

-------
                                                             Table  1

                            RESULTS  OF FORCED OXIDATION  TESTS  WITH  TWO  SCRUBBER  LOOPS

                                ON THE  VENTURI/SPRAY  TOWER  SYSTEM  USING  LIME  SLURRY
Major Test Conditions
Fly ash loading
Flue gas rate, acfm 
-------
clarifed liquor from the dewatering system was returned to the venturi  loop  or  the
spray tower loop.  Beginning with Run 863-1A, the returning clarified  liquor was
split between the two scrubber loops to control  the spray tower slurry  solids con-
centration at about 10 percent.  This solids level  was a compromise between  low
solids (below about 7 percent) where scaling and a  drop in S02 removal  is  experi-
enced and high- solids (above about 15 percent) where it becomes more difficult  to
keep the mist eliminator clean.
F1 ue Gas Rate - Earlier runs were made at a reduced flue gas flow rate  of  25,00.0
acfm (at 300°F) because it was assumed that forced oxidation would reduce  .S0£  removal
efficiency in the venturi  loop at lower pH to the extent that an overall S0~ re-
moval efficiency of at least 80 percent could not be achieved.   This  proved not to
be the case.  Beginning with Run 862-1A, the flue gas flow rate  was increased  to
maximum achievable of 35,000 acfm (at 300°F) which corresponds to a spray  tower
superficial velocity of 9.4 ft/sec.   In this run and subsequent  runs  at 35,000 acfm,
S02 removal averaged about 85 percent at 2000 to 3000 ppm inlet  SO^ concentration.
Spray tower slurry liquor pH - In earlier runs the spray tower inlet  slurry  liquor
pH was controlled at 8.0.  In some runs, especially those at low slurry  solids  con-
centration as mentioned previously,  this level  of inlet pH resulted in an  outlet
pH approaching 6, causing sulfite scaling.


While the pH drop across the spray tower depends  on the SO? removal and  the  inlet
S0£ concentration, it has been generally observed that sulfite scaling does  not
occur if the spray tower outlet pH stays below about 5.5  Therefore,  during
Run 863-1A, the inlet pH was adjusted downward slightly to 7.R.   In this run and
in subsequent runs, small patches of scale  were observed to appear and disappear
in a cyclic manner.  This cyclic appearance of scale did not interfere with
scrubber operation.  No effect on S02 removal  was discernable as a result  of the
slight adjustment in inlet pH.


This slight adjustment in pH is significant in that it demonstrates the  need for
good pH control in commercial installations and it demonstrates  one of several
operating adjustments that can be made to eliminate a scaling problem.
Slurry Level in Oxidation Tank - In Runs 864-1A through 867-1A,  the effect  of  oxi-
dation tank slurry level  (and consequently air and slurry residence times)  was
explored.  In these tests, air was discharged into the bottom of the oxidation tank
through an open 3-inch pipe as previously described (see Figure  3).  Major  test
                                    22

-------
conditions are listed in Table  1.   All  runs  were  nade  at  an oxidation pH of 5.5.
The effect of the tank level  is summarized below:

            Oxidation                Air  Stoichiometry,           Percent
Run No.   Tank Level, ft        atoms 0/mole 50~  absorbed    Sulfite Oxidation


864-1A         18                       1.8                          98

867-1A         14                       1.8                          98

865-1A         10                       2.1                          89

866-1A         10                     3.8/2.7                      98/R1
Oxidation efficiency was hiqh at 18 and 14-ft tank  levels  but  dropped ofV  at  a
10-ft level.  In Run 866-1A, hiqh oxidation efficiency was achieved  at  a 10-ft tank
level by increasing air Stoichiometry.   Part of Run R6R-1A was made  at  lower  air
Stoichiometry with subsequent loss in oxidation efficiency.


These runs demonstrated that 98 percent sulfite oxidation  can  he  achieved  at  14
to 18-ft tank levels at an air Stoichiometry of 1.8 atoms  0/mole  S02 absorbed.
At a 10-ft tank level an air Stoichiometry approaching 3.8 is  required.


It must be pointed out that the oxidation at a 10-ft tank  level is not  directly
comparable with those at 14 and 18 feet because the top turbine of the  agitator
is located at the 11-ft level.  In 10-ft slurry level tests, the  top turbine  is
not  in contact with the slurry and a different aqitation pattern  results.


Filter cake solids concentration during these tests was about  85  percent when
the  oxidation efficiency was 98 percent.  In test periods  when oxidation efficiency
dropped below 90 percent, the filter cake solids concentration tended toward  a  lower
range of 80 percent.


Lime Reliability Run - From mid-December 1977 through mid-January 1978, Run R63-1A,
a one-month lime-slurry reliability run, was made with the venturi/spray tower
system in a two-scrubber-loop configuration with forced oxidation in the venturi
scrubber loop.  Onstream operation for this run totaled 779 hours (32 days).  The
run  was designed to demonstrate operating reliability of the scrubber system  with
respect to  scaling and plugging and to determine if the F.PA New Source  Performance
Standards for SOp and particulate emissions could he met.


To simulate variable boiler  load, the  flue  gas flow rate was varied between
18,000 and  35,000 acfm  (4.8  and 9.4 ft/sec  spray tower superficial gas  velocity)
as the boiler load varied between 100  and 150 MW.  Flue gas with high fly ash
loading was used.  The  venturi  plug was fixed at a position to give 9 inches  H20-
                                    23

-------
pressure drop across the venturi  at  full  35,000 acfm flue qas flow rate.  The
actual venturi pressure drop ranged  from  2  to 9 inches HpO.  The slurry recircu-
lation rates to the venturi  and  spray  tower were  held constant at 600 and 1600
gpm, respectively.   The venturi  inlet  pH  was controlled at 5.5.  The oxidation
tank level was 18 ft and the oxidation air  flow rate was 210 scftn discharqed
through a 3-inch pipe.   As  previously  discussed,  the spray tower slurry inlet pH
was adjusted downward from  8.0 to 7.8  to  eliminate an observed sulfite scale buildup.
During the run, the scrubber was  shut  down  a total of 57 hours; 46 hours were due to
boiler outages, 7-1/2 hours were  for scheduled  scrubber inspections, and 3-1/? hours
were unscheduled downtime.   This  resulted in a  scrubber availability of 99.6 percent,
excluding the interruptions due to boiler outage  and the scheduled inspections.  The*
unscheduled downtime included 2 hours  for mist  eliminator cleaning and l-l/? hours
for air compressor repair.


Average SOp removal for the entire run was  88 percent at 2950 ppm average inlet SOo
concentration.  This corresponds  to an average  emission of 0.9 Ib S02/MM Rtu, well
within the EPA standard of  1.2 Ib S02/MM Btu.   However, due to unusually wide fluc-
tuations in inlet S02 concentration and slow system response time, the S02 emissions
at times exceeded the EPA standard for periods  greater than the three hours allowed
by EPA regulations.


The fluctuations in inlet SOo concentration, ranging up to 4700 ppm, resulted from
the wide variety of coals being burned during the 1977-78 coal strike.  Normally,
inlet SOo concentration ranges between 2000 and 3000 ppm.  These high S02 concentra-
tions were beyond the capacity of the  venturi/spray tower system to remove with its
limited slurry recirculation rates (liquid-to-gas ratios of 57 and 21 gal/Macf at
35,000 acfm full gas flow rate in the  venturi and spray tower, respectively).


Average particulate loading was 0.046  grain/dry scf corresponding to an average
emission of 0.09 Ib particulate/MM Btu (assuming  30 percent boiler excess air).
Although the EPA standard of 0.10 Ib particulate/MM Btu was not exceeded on the
average, a few measurements exceeded this value.
Sulfite oxidation averaged 97 percent during the  run  with  the  air  stoichiometric
ratio varying between 1.4 and 2.8 atoms 0/mole S02 absorbed.   The  filter cake was
excellent throughout the run with solids concentration  averaging 85  percent.  Lime
utilization was 90 percent in the spray tower and 98  percent overall,  reflecting
the high utilization to be expected in a two-scrubber-loop system.
At the first scheduled inspection after 160 operating hours,  the mist  eliminator
was found to be 15 percent restricted by solids.   After a  review of the history
of the mist eliminator exposure, the restriction  was attributed to excess  calcium
carbonate from the previous limestone run (limestone stoichmetric  ratio of 1.65
                                     24

-------
in the spray tower)  and a failure  to activate  the  intermittent underwash for the
first eiqht hours of the reliability run.   At  the  beginning of the reliability run,
the mist eliminator  underwash had  been changed from  continuous with diluted clari-
fied liquor (needed  for the limestone run  conditions)  to  intermittent with makeup
water (1.5 gpm/ft? for 6 minutes every 4  hours -  satisfactory for lime runs).  The
mist eliminator was  cleaned and  the  run was continued.  This mishap was frustrating
in that it broke a record of 41R3  hours of operation under widely varyinq conditions
without cleaning the mist eliminator.


At subsequent inspections at 399 operating hours  and at the end of the run, the
mist eliminator was  entirely clean.


In summary, the operatinq reliability of  the venturi/snray tower system in a
two-scrubber-loop configuration  with forced oxidation in  lime slurry service
has been demonstrated with a system  availability  of  99.6  percent.  However, under
the conditions selected, the system  was unable to continually meet F.PA New Source
Performance Standards for S02 and  particulate  emissions even thouqh the average
emissions for the run met the standards.
UPDATE ON TWO-SCRUBRER-LOOP TEST RESULTS  WITH  LIMESTONE  SLURRY


Since the last report, 6 runs have been made with limestone  slurry  in  a two-scrubber-
loop configuration with forced oxidation.   Results of  the  tests  are summarized in
Table 2.  In these tests, several  operatinq problems were  solved and operational
reliability was established.   Filter cake solids concentration stayed  consistently
above 85 percent throughout the tests.  The following  discussion highlights the
new information developed from the recent runs.
Flue Gas Rate - As with lime testing,  earlier limestone runs were made at  a reduced
flue gas flow rate under the assumption that, with  forced  oxidation, high  S0£
removal (in the range of 85 percent)  could not be achieved at  full  gas rate.  Be-
ginning with Run 815-1A, the flue gas  flow rate was increased  from  the reduced rate
of 25,000 acfm (at 300°F) to the maximum rate of 35,000 acfm.   S02  removal for this
run was 86 percent under the test conditions listed in Table ?., which is about 5
percentage points below S02 removal  achieved on an  identical run at the lower flue
gas flow rate.  All subsequent runs  were made at the higher flue gas flow  rate.


System Control - In the limestone tests with two scrubber  loops, the control
philosophy was to hold the venturi inlet pH (oxidation tank pH) at  5.5 by  adjust-
ing the limestone slurry feed rate to the spray tower effluent hold tank.  Control
in this manner proved to be difficult  and wide fluctuations were experienced in both
pH and limestone st.oichiometry.  For example, in Run 815-1A, the venturi inlet pH
varied between 4.9 and 6.3 with corresponding fluctuations in  the limestone stoichio-
metric ratios of 1.1 to 1.9 in the venturi loop and 1.2 to 2.1 in the spray tower
loop.
                                    25

-------
                                                                            Table  2

                                          RESULTS  OF  FORCED OXIDATION  TESTS  WITH  TWO SCRUBBER LOOPS

                                           ON  THE  VENTURI/SPRAY  TOWER  SYSTEM USING LIMESTONE  SLURRY
Major Test Conditions
Fly ash loading
Flue gas rate, acftn 9 300°F
Slurry rate to venturl, gpra
Slurry rate to spray tower, gpm
Venturi percent solids redrculated (controlled)
Residence times, m1n: Oxidation tank
Desupersaturation tank
Spray tower EHT
Venturl Inlet (oxidation tank) pH (controlled)
Spray tower limestone stolchiometrlc ratio
Venturl pressure drop, In. H20
Oxidation tank level, ft
Air rate to oxidation tank, scfm
Clarified liquor returned to '3'
Selected Results
Onstream hours
Percent SO, removal
Inlet SO, concentration, ppm
Spray tower percent sol Ids redrculated
Spray tower Inlet pH
Spray tower limestone stolchiometrlc ratio
Spray tower inlet liquor gypsum saturation, X
Spray tower sulflte oxidation, X
Overall sulflte oxidation, t
Overall limestone utilization, X
Venturl Inlet liquor gypsum saturation, X
Venturl Inlet liquor sulflte concentration, ppre
A1r stolchionetry, atoms 0/roole S02 absorbed
Filter cake solids, wtl'4'
Mist eliminator restriction, t'5'
815-1A
Low
35 ,000
600
1400
15
8.8
4.7
13.4
5.5
-
9
14
210(1)
S.T

306
86
2500
8.4
5.85
1.98
105
26
96
67
105
40
1.70
87
3
816-1A
Low
35,000
600
1400
15
11.3
4.7
13.4
5.5
-
9
18
210(D
S.T.

142
86
2350
7.7
5.75
1.68
105
27
98
83
100
25
1.80
86
3
817-1A
High
35,000
600
1400
15
11.3
4.7
16.8
5.5
-
9
18
210<2>
Vent

188
83
2500
8.9
S.9
1.60
100
21
97
82
105
25
1.75
86
1
818-1A
High
35 ,000
600
1600
15
11.3
4.7
14.7
5.5
-
7.5-9
18
210*2^
Vent

141
86
2550
9.6
5.9
1.64
100
19
98
81
105
25
1.70
86
2
819-1A
High
Variable
600
1600
15
11.3
4.7
14.7
5.5
-
*9
18
210^2^
Vent

840
86
2950
10.0
5.85
1.65
100
21
98
81
100
25
1.45-2.80
87
3
819-18
High
Variable
600
1600
15
11.3
4.7
14.7
-
1.6
tg
18
210<2>
Vent

126
85
3000
9V6
5.9
1.65
110
19
98
83
105
25
1.45-2.80
86
3
NJ
              dotes:
                    1  A1r discharged downward through sparger ring with 40-% Inch diameter holes about 3 Inches from tank bottom.
                      Air discharged downward through 3-Inch diameter pipe with an open elbow at center of oxidation tank about 3  Inches from tank bottom.
                      Spray tower loop (effluent hold tank) or venturl 3oop (oxidation tank).
                    ,  Ctarlfler and filter 1n series used for solids dewaterlng 1n all runs.
                    5) Continuous mist eliminator bottom wash with diluted clarified liquor at 0.4 gpm/ft2.  Sequential top wash with makeup water with one of
                      6 nozzles on at 0.53 gpm/ft2 for 4 minutes every 80 Minutes.

-------
In Run 815-1A, the oxidation tank  level was  14  feet, which was satisfactory for
forced oxidation (96 percent oxidation  at  an air  stoichiometric ratio of 1.7
atoms 0/mole S0£ absorbed).   In Run  816-1A,  the fluctuation in venturi inlet pH
was reduced to a range of 5.2 to 5.8 with  corresponding reduction in fluctuation
in limestone stoichiometry by increasing the oxidation tank level to the maximum
of 18 feet and-thus increasing hold  tank residence time.

In Run 819-1B, the control  philosophy was  changed.   In this run, the limestone
stoichiometry in the spray tower was controlled at 1.6 moles calcium per mole SO?
absorbed and the venturi  inlet pH  was allowed to  vary.  With direct control on the
spray tower stoichiometry,  the fluctuation in venturi inlet pH was 5.2 to 5.8, no
greater than in the previous runs  with  venturi  inlet pH control.

Based on these runs, control of limestone  stoichiometry in the primary scrubbinq
loop (spray tower) is recommended  over  control  of pH in the oxidation loop
(venturi).                                                                 v
Mist Eliminator - In previous runs  with  limestone  slurry  and high fly ash loadings
(Runs 805-1A  through 808-1A), problems  with  mist  eliminator plugging occured.  In
these runs, the spray tower solids  concentration was maintained at 15 percent, which
required that the clarified liquor  from  the solids dewatering  system be returned to
the venturi loop and which allowed  only  enough makeup water in the spray tower system
for an intermittent mist eliminator underside wash.  Such a wash was inadequate at
the limestone utilizations experienced in  the spray tower (60  to 70 percent) and
the mist eliminator plugged within  a matter of days.

Beginning with Run 817-1A, the mist eliminator was washed continuously at 0.4 qpm/
ft* with clarified water diluted with available makeup  water.  Excess clarified
water was returned to the venturi loop.  This wash scheme (coupled with a sequen-
tial top wash - see Table 2) proved adequate  and the mist eliminator no longer
plugged.

The continuous wash diluted the spray tower solids concentration to about 9 percent.
At 9 percent solids concentration,  SOg removal dropped  a  few percentage points to
83 percent at 2500 ppm inlet S02 concentration.

In Run 818-1A, the slurry recirculation  rate  in the spray tower loop was increased
from 1400 gpm to the maximum controlled  rate  of 1600 gpm. With this modification
S02 removal was increased to 86 percent  at 2550 ppm inlet concentration.
Limestone Reliability Run - During November 1977,  Run  819-1A, a one-month limestone
slurry reliability run, was made with a two-scrubber-loop  configuration on the
venturi/spray tower system and with forced  oxidation in  the venturi loop.  This
run operated for a total of 840 hours (35 days).   As with  the lime reliabilty
run, the run was designed to demonstrate operating reliability of the scrubber
system and to determine if the EPA New Source Performance  Standards for SOp and
particulate emission could be met.
                                    27

-------
 Flue gas and slurry flow rates  were  the  same  as with the lime reliability run.
 Flue gas with high fly ash loading was varied in  rate between 18,000 and
 35,000 acfm (at SOOOf) to follow the boiler load.  The venturi plug was fixed to
 give 9 inches HgO pressure drop at 35,000  acfm flue gas rate.  Slurry recircula-
 tion rates were held constant at 600 gpm and  1600 gpm in the venturi and spray
 tower loops respectively.  The  venturi inlet  pH was controlled at 5.5 by control-
 ling the limestone feed rate to the  spray  tower hold tank.*  The oxidation tank
 was maintained with an 18-ft slurry  level  and an  air flow rate of 210 scfm dis-
 charged through a 3-inch pipe.


 During the run, the scrubber was shut down for 18 hours due to a boiler outage,
 5 hours total  for weekly inspections, plus 3-1/2  hours of unscheduled downtime'for
 a total  of 26-1/2 hours.   Based on unscheduled downtime, the scrubber system had
 an availability of 99.6 percent.   The unscheduled downtime included 3 hours to
 clean a partially plugged slurry return  pipe  from the venturi to the oxidation
 tank and 1/2 hour to clean a plugged mist  eliminator nozzle.


 The plugged mist eliminator nozzle was discovered after 391 hours of operation.
 The mist eliminator in the vicinity  of the slurry nozzle was severely restricted
 by slurry solids (7  percent overall  mist eliminator restriction).  The nozzle
 was cleaned but the mist  eliminator  was  not disturbed.  By the end of the run
 (840 hours), the mist  eliminator restriction  had dropped to 3 percent, demon-
 strating that  a restricted area can  be self cleaning.


 For the entire run,  the SO? removal  averaged  86 percent at 2950 ppm average
 inlet SO? concentration.   This  removal efficiency corresponds to an average emis-
 sion of 1.0 Ib S02/MM  Btu  which meets the  EPA New Source Performance Standard of
 1.2 Ib S02/MM  Btu.   However, as with the lime reliability run, fluctuations to
 unusually high inlet SOg  concentrations were  experienced and the standard was
 at times exceeded  for  periods greater than the three hours allowed by EPA regu-
 lations.


 The outlet  particulate  loading  ranged from 0.021 to 0.063 grain/dry scf with an
 average  of  0.042 grain/dry scf.  Assuming  30  percent excess air to the boiler
 the average outlet  particulate  loading corresponds to 0.08 Ib/MM Btu which meets
 the EPA New Source Performance  Standard of 0.1 Ib/MM Btu.  However, a few of the
 outlet particulate  loading measurements exceeded the standard.


 Sulfite  oxidation averaged  98 percent during  the run with the air stoichiometric
 ratio  varying  between  1.4  and 2.8  atoms 0/mole SO? absorbed.   The filter cake
 solids concentration averaged 87 percent.  Overall limestone utilization was 81
 percent  while  the spray tower limestone utilization was 61  percent, again demon-
 strating the advantage  of  a two-scrubber-loop system in achieving high alkali
 utilization.


* As previously  discussed, this mode of control was later changed to Sto1ch1ometr1e
   ratio  control  in the  spray tower (Run 819-1B).


                                 28

-------
To summarize, the operating reliability of  the  venturi/spray tower system in a
two-scrubber-loop configuration with forced oxidation  in  limestone slurry service
has been demonstrated with a system availability  of  99.6  percent.  However,
under the conditions selected,  the system was unable to continually meet EPA New
Source Performance Standards for SO? and particulate emissions even though the
average emissions for the run met the standards.
TWO-SCRUBBER-LOOP TEST RESULTS USING LIMESTONE SLURRY WITH  ADDED  MAGNESIUM OXIDE


Beginning in March 1978, a series of six runs were made in  which  magnesium oxide
was added to the spray tower hold tank along with the limestone slurry.   The  pri-
mary purpose of the magnesium oxide addition was to enhance S0£ removal  efficiency
in the spray lower loop by increasing the sulfite ion concentration in the liquor
for S02 scrubbing.  In a two-scrubber-loop configuration as shown in Figure 2,
the magnesium ion concentration in the venturi loop is higher than that in the
spray tower loop because of the water loss in humidifying the flue gas in the
venturi loop.  But because the sulfite ion is converted into nonscrubbing sulfate
ion by forced oxidation, the higher magnesium ion concentration in the venturi
loop does not enhance $62 removal in the venturi loop.  The secondary purpose of
the magnesium oxide addition was to determine whether the presence of magnesium
ion had an effect on oxidation efficiency.


Typical operating conditions and results of these tests are summarized in Table 3.
The expected enhancement of S02 removal was achieved and oxidation efficiency was,
if  anything, improved.  Thus, magnesium oxide addition is compatible with a two-
scrubber-loop forced oxidation system.
 S02  Removal  -  Run  820-1A was made  under  identical conditions to Run 818-1A
 (Table  Z)  except for  the addition  of magnesium  oxide.  Effective magnesium
 ion  concentration* averaged 5150 ppm in  the  spray tower.  The  anticipated removal
 enhancement  was achieved as the average  S02  removal was 96  percent at 2250 ppm
 average inlet  S02  concentration compared with 86  percent  removal at ?550 inlet
 ppm  for Run  818-1A.


 The  spray tower inlet slurry  liquor was  100  percent  saturated  in gypsum and no
 scale was observed.   This  condition was  typical of all the  tests in the limestone/
 magnesium oxide,  forced-oxidation  test block.
 * Effective magnesium ion concentration is defined as the total  magnesium ion
   minus that magnesium ion concentration equivalent to total  chlorides.   Magnesium
   chloride has no effect on S02 removal.


                                     29

-------
OJ
o
                                                                            Table  3

                                          RESULTS  OF  FORCED  OXIDATION  TESTS WITH TWO  SCRUBBER  LOOPS

                        ON THE  VENTURI/SPRAY  TOWER  SYSTEM  USING LIMESTONE SLURRY WITH ADDED  MAGNESIUM OXIDE
Ma.lor Test Conditions
Fly ash loading
Flue gas rate, acfm 9 300°F
Slurry rate to venturl , gpm
Slurry rate to spray tower, gpm
Venturl percent solids redrculated (controlled)
Residence times, m1n: Oxidation tank
Desupersaturatlon tank
Spray tower EHT
Venturl Inlet (oxidation tank) pH (controlled)
Spray tower limestone stolchiometrlc ratio (based on solids)
Effective Mg** concentration (S.T. loop), ppm
Venturl pressure drop, 1n. HjO
Oxidation tank level, ft
Air rate to oxidation tank, scfm' '
Clarified liquor returned to "'
Selected Results
Onstream hours
Percent SO, removal
Inlet S02 concentration, ppm
Spray tower percent sol Ids redrculated
Spray tower Inlet pH
Spray tower limestone stoichiometrlc ratio (based on total slurry)
Spray tower Inlet liquor gypsun saturation, X
Spray tower sulflte oxidation, X
Effective Hg concentration (S.T. loop), ppm
Overall sulf1te oxidation, X
Overall limestone utilization, X
Venturl Inlet liquor gypsum saturation, t
Venturl Inlet liquor sulflte concentration, ppm
A1r stolchlometry, atom 0/mole SO., absorbed
Filter cake solids, wtX (3)
M1st eliminator restriction, x'4'
820- 1A
High
35,000
600
1600
15
11.3
4.7
14.7
5.5
-
5000
9
18
210
Vent

462
96
2250
6.0
6.05
1.16
100
30
5150
98
92
130
50
1.70
85
-
820- IB
High
35,000
600
1600
15
11.3
4.7
14.7
-
1.6<6)
5000
9
18
150
Vent

137
94
2500
8.3
5.9
1.28
105
17
4985
92
90
130
950
1.10
82
0
820- 1C
High
35,000
600
1600
15
11.3
4.7
14.7
-
1.6(6)
5000
9
18
0
Vent

134
91
2750
10.5
5.9
1.52
90
20
4700
36
82
145
5585
0
63
0
821-1A
High
35,000
600
0
15
11.3
4.7
-
5.5
-
5000
9
18
210
Vent


(7)













0
822- 1A
High
35,000
600
1600
15
11.3
4.7
14.7
-
1.6(6)
5000
9
18
210
Vent

232
91
2750
8.0
5.75
1.55
100
21
4985
97
79
125
735
1.45
85
-
822- IB
High
35,000
600
1600<5>
15
11.3
4.7
14.7
-
1.6(6)
5000
9
18
210
Vent

85
90
2400
5.6
5.55
1.21
110
23
4845
98
93
130
410
1.70
B5
0
                    Notes:
                            Air discharged downward through 3-Inch diameter pipe with an open elbow at center of oxidation tank about 3 Inches from tank bottom.
                            Venturl loop (oxidation tank).
                            Clarlfler and filter used for solids dewaterlng In all runs.
                            Continuous mist eliminator bottom wash with diluted clarified liquor at 0.4 gp»/ft2. Sequential top wash with makeup water with one of
                            6 nozzles on at 0.53 gpn/ft2 for 4 minutes every 80 minutes.
                            Spray tower turned off for 30 minutes every 8 hours to obtain SO; removal with venturl alone.  Venturl SO? removal averaged 29X.
                            In runs with control by spray tower stolchjnetrlc ratio, the venturl Inlet pH mriged 5.0.
                            Run failed due to low NoO dissolution rate In spray tower effluent hold tank.

-------
In Run 821-1A, an attempt was made to  determine  the  S02  removal in the venturi by
turning off the slurry recirculation to the  spray tower.  Unfortunately, the
magnesium oxide, added to the spray tower hold tank, would not dissolve without
recirculation and the run was aborted.


A second effort was more successful.   Run 822-1R was an  extension of 822-1A in
which the spray tower slurry recirculation was turned  off once a shift for only
30 minutes.  This short time period did not  upset the  system balance.  S02 removal
in the venturi loop was found to be 29 percent which is  typical of removal effi-
ciency with limestone slurry in the absence  of magnesium ion.  Thus, it has been
demonstrated that magnesium ion does not enhance S02 removal in a scrubber loop
with forced oxidation.


Run 822-1A was made in an effort to improve  removal  efficiency by minor
changes in piping configuration to locate makeup and bleed streams at their
optimum locations i-n the venturi slurry recirculation  loop.  Referring to
Figure 2, the bleed from the spray tower loop was sent to the desupersaturation
tank instead of the oxidation tank as  shown.  Also,  the  bleed to the solids
dewatering system was taken from the oxidation tank  instead of the desupersatura-
tion tank as shown.  Improvement in SOp removal  efficiency, if any, was too small
to observe.
Oxidation Efficiency - In this whole test block,  the 3-inch  pipe was used for dls-
charging air into the oxidation tank and an oxidation tank level of 18 feet was
maintained.  Runs 820-1A, B,  and C were a series  to explore  the air stoichiometry
required to achieve near complete oxidation.   Results were as  follows:
                   Air Stoichiometric Ratio,                  Percent
                   atoms 0/mole S02 absorbed             Sulfite Oxidation
820-1A                       1.7                               98

820-1B                       1.1                               q?

820-1C                       0                                 36
The oxidation efficiency was marginally acceptable at an air Stoichiometric ratio
of 1.1.  Although the oxidation efficiency averaged 92 percent,  it  fluctuated
widely, indicating that barely enough air was available.  During the  last 40 hours
of Run 820-1B, air Stoichiometric ratio increased to 1.3 and the oxidation efficienc
was steady at 98 percent.  Thus, sulfite oxidation efficiency appears to be unaf-
fected, if not improved, by the addition of magnesium oxide.
                                    31

-------
Filter cake solids concentration at 98 percent oxidation  averaged  85  percent,
demonstrating that magnesium oxide addition does .not adversely affect dewatering
characteristics of oxidized sludge.  This series of runs  also demonstrated  the
effect of forced oxidation on solids dewatering characteristics.   Filter  cake
solids concentration decreased from 85 percent to 63 percent  as the oxidation
of sulfite decreased from 98 percent to 36 percent.


An additional  observation in this series of runs was that overall  limestone
utilization decreased from 92 to 82 percent as the air rate to the oxidation
tank was reduced from 210 to 0 scfm.  Presumably, the higher  air rate gave better
agitation of the slurry and promoted the limestone dissolution.
                                    32

-------
                               Section  3
         FORCED OXIDATION WITH ONE  SCRUBBER  LOOP ON THE TCA SYSTEM
Forced oxidation with good S0£  removal  in  a  sinqle scrubber loop has been
demonstrated in the TCA system  using  limestone slurry.  In this arrangement,
sulfite oxidation is achieved by contacting  the slurry with air in the scrubber
hold tank.  A compromise must be made in the scrubber liquor pH between a higher
pH desired for good SOg removal  and a lower  pH desired for good oxidation.
Although the optimum oxidation  rate occurs at about 4.5 pH, it has been found
that the oxidation rate is adequately fast up to a pH of about 6.  Thus, the
oxidation pH range is compatible with the  limestone scrubbing pH range of 5
to 6.


Forced oxidation in a single scrubber loop is detrimental to lime slurry scrubbing
because sulfite ion, a major scrubbing species in a lime based scrubbing system,
is removed in the oxidation process.   Thus,  forced oxidation substantially re-
duces SOg removal efficiency in a single loop lime system.
The single loop configuration  is  of  prime  interest commercially because the
majority of commercial  installations,  both operating and planned, are of this
type.  Modification of  these installations for forced oxidation would require as a
minimum a compressor (or blower)  plus  an air  sparger in the scrubber hold tank.


Two devices for air/slurry contact have been  tested on the TCA system.  From
late June through early October 1977 an air eductor was tested.  Experience with
the air eductor has been previously  reported. 3)   Because of erosion problems and
high energy consumption, the eductor has been replaced with an air sparger similar
to the one used in the  venturi  oxidation tank.  Tests with the air sparger were
conducted from early December  1977 through late January 1978.  All tests were
conducted with flue gas containing high fly ash loadings.


SYSTEM DESCRIPTION


Two operating configurations were used in  the single loop tests.  With one hold
tank as shown in Figure 4, effluent  slurry from the scrubber  is discharged to
the oxidation tank where limestone is  added and the slurry is recycled back to


                                    33

-------
MAKE UP WATER
FLU EG AS
LIMESTONE
CLARIFIED LIQUOR ^
\



BLEED TO SOLIDS
_ DEWATERING SYSTEM
REHEAT





r-LUb laAJS

/ \
* * *
Siii.!
>>M55!>
LLTJ





1 	 N





f


1

000 TCA
O OO O O
oo o
o o o o o
o O O
o o o o_c
D

<=3C
cdc


'
3
WATER
3 I COMPRESSED
^ f AIR


                   ^^    OXIDATION TANK

   FIGURE4.   FLOW DIAGRAM FOR SINGLE LOOP
FORCED OXIDATION IN THE TCA SYSTEM WITH ONE TANK
                   34

-------
the scrubber.   With two tanks  in  series as shown in Fiqure 5, effluent slurry
is discharged  to the oxidation tank  and the slurry then passes to a second tank
where limestone is added.   Slurry is  recycled from the second tank back to the
scrubber.  Although the one-tank  configuration  is simpler, the two-tank configura-
tion allows the oxidation  to take place at the  lower pH of the scrubber effluent
before limestone is added.   The two-tank configuration also provides longer
residence time for better  limestone  utilization.


The oxidation  tank arrangement is shown in Figure 6.  The tank is 7 ft in diameter
and was operated at a 17 to 18-ft level.  All tests were conducted with an air
sparger ring made of straight  3-inch  316L SS pipe pieces welded into an octagon
of approximately 4-ft diameter.   It  was located 8 inches from the bottom of the
tank and had 40 1/4-inch diameter holes pointed downward.  The sparger ring was
fed with compressed air to which  sufficient water was added to assure humidifi-
cation.


A major shortcoming of this oxidation system was the agitator which was rated at
only 3 Hp and  rotated at 37 rpm (compared with  17 brake Hp and 56 rpm for the ven-
turi oxidation tank).  This agitator was similar in configuration to the agitator
in the venturi oxidation tank  with two axial flow turbines (49 inches in diameter)
pumping downward.  Because of  the weaker agitation, runs with similar oxidation
tank environment (pH, air  stoichiometry, tank level, percent slurry solids, and
limestone utilization) had lower  oxidation efficiency in the TCA oxidation tank
than in the venturi oxidation  tank.


A 20 Hp variable speed agitator is on order and will be used to develop the rela-
tionship between oxidation tank agitation and air requirements.


A second shortcoming was the existing Shawnee air compressor which did not have
sufficient capacity to serve the  venturi and the TCA oxidation tanks simultaneously
at full flue gas load.  To circumvent this problem, several of the TCA runs were
made at reduced flue gas flow  rates.   An additional air compressor has been ordered
to correct this limitation.


A clarifier was used for dewatering  in all runs except Run 821-2A where a clarifier
followed by a  centrifuge was used.


SUMMARY OF PREVIOUSLY REPORTED TEST  RESULTS WITH AIR EDUCTOR
Forced oxidation test results with one scrubber loop conducted  from  late June
through early October. 1977 with limestone slurry using the air  eductor have been
previously reported.-3)   These tests were conducted at 30,000  acfm  (300 F) flue
gas rate which corresponds to a superficial  gas velocity in the TCA  of 12.5 ft/sec.
The slurry recirculation rate was 1200 gpm.   Each run averaged  about 5 to 6 days.
All runs were made with flue gas having high fly ash loadings.

                                    35

-------
                         REHEAT
 MAKEUP WATER
FLUE GAS
LIMESTONE
CLARIFIED LIQUOR
   BLEED TO SOLIDS
   DEWATERING SYSTEM
                                        FLUE GAS
                                A
1
                                   popoo
                                   OOOOO
                                   OOJDOOO
                                              TCA
                                               COMPRESSED
                                                   AIR
                                                       A ?Vft
                                EFFLUENT HOLD
                                   TANK
               OXIDATION
                 TANK
               FIGURE 5.    FLOW DIAGRAM FOR SINGLE LOOP
           FORCED OXIDATION IN THE TCA SYSTEM WITH TWO TANKS
                               36

-------
                FIGURE 6
         ARRANGEMENT OF THE TCA
    OXIDATION TANK WITH AIR SPARGER
AGITATOR
   BAFFLE
                                SPARGER
                           COMPRESSED AIR
             OXIDATION TANK
                PLAN VIEW
1
J
OUTLET *— '

BAFFLE v.
^s.



SPARGER WITH
401/4-inch HOLES
(DOWNWARD DISCHARGE) v

01 2345



^



\





1 i


c.
F™5'
1 t
V

»

• — •• i ii _


!3-

j
y
[ 1
7
/





"


1

•

AGITATOR
^^(37 rpm, 3 Hp)

— - OYinATIOM TAIJIT


x COMPRESSED AIR
_ INLET
r
              ELEVATION VIEW
              37

-------
 Key results from these  earlier tests were as follows:


       t   The dewatering  and  handling characteristics of slurry solids oxidized
           to 90  percent or better in a single loop system were as good as those
           in a double loop system.


       •   Sulfite oxidation to 98 percent with qood S0£ removal was demonstrated
           in a single scrubber loop with two hold tanks using an air eductor for
           air/slurry contact.


       •   Conditions under which near complete oxidation was demonstrated were
           slurry feed to  the  eductor from a small downcomer hold tank at  5.15 pH,
           eductor discharge to the oxidation tank held at 8-ft slurry level  and
           5.5  pH, and an  air  stoichiometric ratio of about 2.5 atoms 0/mole  SOg
           absorbed;


       •   S02  removal was enhanced slightly by single loop forced oxidation  with
           limestone scrubbing.


       •   The  rubber lined eductor diffuser eroded severely in less than  1500
           hours of operation.
ONE-SCRUBBER-LOOP TEST RESULTS WITH AIR SPARGER


Eight forced oxidation runs with limestone slurry were made on  the TCA system in
a one-scrubber-loop configuration with an air sparger.  Results of these tests
are reported in Table 4.  Despite agitator and air compressor limitations, forced
oxidation with an air sparger in a single scrubber loop was demonstrated.
Air Stoichiometrv - Runs 815-2A through 818-2B were made  with two hold tanks in
series as shown in Figure 5.  The primary effort during these runs was to identify
the air stoichiometric ratio required for near complete oxidation.  In the first "
two tests, run at the maximum achievable flue gas flow rate of 30,000 acfm, it
was found that the air compressor did not have a high enough capacity to supply
both the venturi/spray tower system and the TCA system.   With an air rate of 210
scfm to the venturi oxidation tank only 180 scfm was available for the TCA system.
Further tests were conducted at reduced flue gas flow rates (20,000 to 25,000 acfiri\
to allow higher air stoichiometry at the available air rate.  Results of these tests
conducted over an oxidation tank pH range of 5.4 to 5.7 were as follows:

                                     38

-------
                                                                Table 4

                                                RESULTS OF  FORCED OXIDATION TESTS

                          WITH ONE  SCRUBBER  LOOP ON THE  TCA SYSTEM USING LIMESTONE  SLURRY
Major Test Conditions
Fly ash loading
Flue gas rate, acfm e 300°F
Slurry flow rate to TCA, gpm
Percent solids recirculated
Residence times, min: Oxidation tank
EHT
Oxidation tank level, ft
Airflow rate to sparger, scftr1'
Limestone stoichicmetric ratio (controlled)
TCA inlet pH (controlled)
Effective Mg*+ concentration, ppm
Limestone addition point
Total static height of spheres, inches
Selected Results
Onstream hours
Percent S02 removal
Inlet S02 concentration, ppm
Percent sulfite oxidation
Air stoichiometry, atoms 0/mole SO^ absorbed
TCA Inlet pH
Oxidation tank pH
Limestone utilization, X
Gypsum saturation in TCA inlet liquor, %
Mist eliminator restriction, X(
815-2A
High
30,000
1000
15
5.2
14.4
18
130
1.3
-
-
EHT
20

75
89
3000
40
1.0
6.25
-
80
110
0.5
816-2A
High
30,000
1000
15
5.2
14.4
18
180
1.3
-
-
EHT
22.5

48
91
2850
54
1.40
6.25
5.7
76
100
-
817-2A
High
20,000
1000
15
5.2
14.4
18
130
1.3
-
-
EHT
22.5

161
79
3000
94
1.70
5.8
5.45
81
100
-
818-2A
High
25,000
1000
15
5.2
14.4
18
130
1.3
-
-
EHT
22.5

131
85
3000
67
1.25
6.2
5.65
77
95
-
818-2B
High
25,000
1000
15
5.2
14.4
18
0
1.3
-
-
EHT
22.5

140
82
3300
24
0
6.2
-
81
no
0
819-2A
High
20 ,-000
1000
15
4.9
-
17
130
1.3
-
-
Oxid. Tk
22.5

164
75
2800
94
1.90
5.55
5.55
77
110
-
820- 2A
High
20,000
1000
15
4.9
-
17
130
-
5.9
-
Oxid. Tk
22.5

259
79
2500
92
2.0
5.65
5.65
62
115
0
821-2A
High
30,000
1200
15
4.1
-
17
170
1.2
-
5000
Oxid. Tk
15

182
84
2500
95
1.65
5.35
5.35
79
110
1.5
Notes:
      1) Air  discharged  downward through sparger ring with 40->s inch diameter holes about 8  inches from tank bottom.
      2) Clarifier used  for solids dewatering except for Run 821-2A where clarifier and centrifuge was used.
      3) Continuous mist eliminator bottom wash with diluted clarified liquor at 0.4 gprn/ft? for Runs 815-2A & 816-2A and at 0.3gpm/ft< for Runs 817-2A
         through 820-2A. Intermittent bottom wash with makeup water for Run 821-2A at 1.5 gprn/ft^ for 4 minutes each hour. Sequential top wash for all
         runs using makeup water with one of 6 nozzles on at 0.55 gpm/ft' for 3 minutes every 10 minutes.

-------
                   Air Stoichiometric  Ratio                Percent
Run                atoms 0/mole S02  absorbed          Sulfite Oxidation
817-2A                    1.7                               94

816-2A                    1.4                               54

818-2A                    Io25                              67

815-2A                    1.0                               40

818-2B                    0                                24
Thus, with two hold tanks in series,  an air Stoichiometric  ratio of about 1.7 was
required to achieve greater than 90  percent oxidation,.  Under similar conditions
in the venturi oxidation tank,  higher oxidation  efficiency  was achieved.  This
better performance in the venturi  oxidation tank was  attributed to the superior
agitation in the venturi tank.
Runs 819-2A and 821-2A were made with the oxidation tank  as the only hold tank as
shown in Figure 4.  In these runs, the pH in the oxidation tank was higher because
of the limestone addition.   Because of the higher pH,  a higher air.stoichiometr.y
was required.  This effect  can be seen by comparing Runs  B17-2A (2 hold tanks)
and 819-2A (1 hold tank) made at essentially the same  operating conditions.  Ninety-
four (94) percent sulfite oxidation was achieved in both  runs.  An air Stoichiometric
ratio of 1.7 atoms 0/mole S02 absorbed was used  in the run with two hold tanks (5.4
oxidation tank pH) while an air Stoichiometric. of 1.9  was required in the run
with one hold tank (5.65 oxidation tank pH).
S02 Removal Efficiency - S02 removal  efficiency in these runs  appeared to be in-
dependent OT oxidation efficiency.   S02 removal  efficiency was primarily a func-
tion of flue gas flow rate and inlet S02 concentration,  closely following pre-
viously developed correlations for the TCA system in  limestone service without
forced oxidation.  At 3000 ppm inlet S0£ concentration and a limestone Stoichio-
metric ratio controlled at 1.3 moles Ca/moles  SO? absorbed, S0£ removal efficiency
ranged from about 90 percent at 30,000 scfm to about  fin  percent at 20,000 scfm.
Limestone Utilization - Limestone utilization was  higher in the  runs using two
tanks in series tnan in the single tank runs.  Again comparing runs 817-2A and
819-2A, limestone utilization with one tank was 77 percent while with two tanks
it was 81 percent,  SOg removal  efficiency was also improved  from 75 percent with
one tank to 79 percent with two tanks.  The improvement  can be attributed to higher
                                    40

-------
residence time (19.6  minutes  with  two tanks  versus 4.9 minutes with one tank) and
the approach to plug  flow reaction inherent  with  tanks in  series.


Because of the relatively poor  S02 removal efficiency in Run  819-2A, the next run
(820-2A)  was made at  a slightly higher pH.   The oxidation  tank pH was  increased
from 5.55 to 5.65.  S02 removal  increased only slightly from  75  percent at 2800
inlet ppm to 79 percent at 2500 inlet ppm.   However, the limestone utilization
decreased from 77 percent to  62 percent.
Magnesium Oxide Addition  -  The addition of magnesium oxide  should  not enhance
S02 removal  in a scrubber loop with  forced oxidation.   This was  demonstrated in
Run 821-2A.   Magnesium ion  in the scrubber liquor improves  $02  removal  by  in-
creasing the sulfite ion, an  effective  SOg scrubbing component.  Rut forced oxi-
dation converts the sulfite to sulfate  which is non-reactive.
In Run 821-2A,  with 5000 ppm effective magnesium ion concentration  and with forced
oxidation,  the  S02 removal  efficiency averaged  84 percent,  no higher  than  expected
without magnesium oxide-addition.   In a previous run with magnesium oxide  addition
equivalent  to Run 821-2A but without forced oxidation,  S02  removal  averaged 92
percent. Thus, the enhancement on SO? removal  with  magnesium oxide addition  is
not achieved in a scrubber loop with forced oxidation.
                                    41

-------
                                 Section  4-


            FORCED OXIDATION OF THE  VENTURI/SPRAY TOWER'BLEED STREAM
Forced oxidation within the scrubber loop  requires a compromise between the
conditions needed for good oxidation and those  heeded for qood SOn removal.
Such would not be the case if it  were possible  to oxidize the slurry bleed
stream by simple air/slurry contact.  Unfortunately, tests at the IERL-RTP
pilot plant^) and at the Shawnee  Test Facility-*) have shown that air sparging
of the bleed stream increases the rate of  dissolution of the residual alkali
and causes the pH to rise, slowinq down the  oxidation rate to an impractica-1
level.  Furthermore, tests conducted with  sulfuric acid addition to control the
bleed stream pH have produced oxidized sludqe with inferior dewaterinq and
handling characteristics.3)


Despite the generally unfavorable results, batch oxidation tests at the Shawnee
Laboratory indicated that near complete sulfite oxidation could be achieved by
simple air sparging of lime or limestone slurry when magnesium ion was present
in concentrations of 1,000 ppm or higher.  Magnesium ion apparently has two
effects:  it tends to buffer the  pH rise from dissolving residual alkali in the
waste slurry solids; and it tends to promote dissolved sulfite availability,
allowing oxidation to take place  at a higher pH.


Starting in mid-May 1978, bleed stream forced oxidation with limestone slurry
and added magnesium oxide was successfully demonstrated in a month Ipnq series
of tests on the venturi/spray tower system.  Oxidized slurry from-these tests
had good dewatering properties with filter cake solids concentration averaging
about 85 percent.  Thus, it is commercially  feasible to improve the quality and
reduce the volume of waste solids in installations incorporating maqnesium ion
in the slurry liquor by simple air/slurry  contact of the bleed stream.
SYSTEM DESCRIPTION


The venturi/spray tower system was  arranged  as  shown  in Figure 7 for the bleed
stream oxidation tests.  Both the venturi  and the  spray tower slurries discharged
into a single hold tank to which  limestone and  magnesium oxide were fed.  A bleed
stream was taken from the spray tower downcomer to take advantage of the low pH
                                    42

-------
FLUE GAS
IV\gO
LIMESTONE
                       REHEAT
                  VENTURI
                               FLUE GAS
                                A
                                **»*»»
                                \L- xU ^l/
MAKE-U P WATER
                                       SPRAY TOWER
                               D
                                              OVERFLOW
                               EFFLUENT
                              HOLD TANK
                          COMPRESSED
                    VENT  AIR

                      t
                              CLARIFIED LIQUOR
                                                                OXIDATION
                                                                  TANK
                                                    BLEED TO
                                                     SOLIDS
                                                   DEWATERING
                                                     SYSTEM
                FIGURE 7.    FLOW DIAGRAM  FOR  BLEED STREAM OXIDATION  IN  THE  VENTURI/SPRAY TOWER SYSTEM

-------
 at that point.  The bleed stream was discharged to the oxidation tank which was
 arranged  as  shown in Figure 3 and is described in Section 2.   During these tests,
 the 3-inch pipe was used to discharge air into the oxidation  tank.   All  tests were
 conducted at an 18-ft oxidation tank level.  Bleed from the oxidation tank was
 dewatered by a clarifier and a filter in series.
BLEED STREAM OXIDATION TEST RESULTS


Four bleed stream, oxidation runs were made on the venturi/spray tower  system
using limestone with added magnesium oxide.  All  tests were conducted  with approx-
imately 5000 ppm effective magnesium ion concentration in  the slurry liquor.
Test results are reported in Table 5.  Percent S0£ removal was high as expected
in runs with magnesium oxide enhancement.  Oxidized slurry solids  in all runs had
good dewatering properties, averaging about 85 percent filter cake solids con-
centration.


In Runs 823-1A and 824-1A, conducted at 18,000 acfm and 35,000 acfm, respectively,
97 to 98 percent sulfite oxidation was achieved at an air  stoichiometry of about
1.6 atoms 0/mole S0£ abosorbed.  Oxidation was consistently high even  though the
oxidation tank pH averaged 6.3 in Run 823-1A and  at times  rose as  high as 6.7.


In these runs, 30 gpm of oxidized slurry was recycled from the oxidation tank back
to the scrubber hold tank.  The purpose of this recycle was to reduce  the pH
difference between the oxidation tank and the scrubber hold tank.  However, the
opposite occurred.  The hold tank pH was depressed, requiring excess limestone
feed to maintain a pH of 5.3.  The net result was a limestone utilization of less
than 40 percent for these runs.


Runs 825-1A and 826-1A (at 18,000 acfm and 26,500 acfm, respectively)  were con-
ducted without this recycle.  In both of these runs the pH difference  between
the scrubber hold tank and the oxidation tank was only about 0.1 and 0.2 with the
oxidation tank pH averaging 5.65 or less.  Near completL oxidation (high 90's)
was easily achieved in both runs with air stoichiometric ratios  of 1.6 and 2.0,
respectively.  Time was not available in the test block to determine minimum
air stoichiometry. Control of limestone feed was  poor in these runs resulting in
relatively low limestone utilization (64 and 61 percent, respectively).


This short series of runs has shown that in systems containing magnesium ion,
the slurry bleed stream can be readily oxidized.   Furthermore, oxidation of the
bleed stream does not interfere with enhancement  of S0~  removal  by the magnesium
ion as was experienced when oxidation was accompli sheer within the  scrubber loop.
                                  44

-------
                                                              Table  5

                    RESULTS OF FORCED  OXIDATION  TESTS  ON  THE VENTURI/SPRAY  TOWER  BLEED  STREAM

                                     USING LIMESTONE SLURRY WITH ADDED MAGNESIUM OXIDE
Major Test Conditions
Fly ash loading
Flue gas rate, acfm ? 300°F
SI urry rate to venturi , gpm
Slurry rate to spray tower, gpm
Percent solids recirculated (controlled)
EHT residence time, min.
Spray tower inlet pH (controlled)
Scrubber limestone stoichiometric ratio (control led) (based on solids)
Effective Mg+ concentration, ppm
Venturi pressure drop, in. H-0
Oxidation tank level, ft
Air rate to oxidation tank, scfnr '
Recycle flow from oxidation tank to EHT, gpm
Selected Results
Onstream hours
Percent SO, removal
Inlet S02 concentration, ppm
Scrubber percent solids recirci/lated
Scrubber inlet liquor pH
Oxidation tank pH
Limestone utilization, % (based on total slurry)
Sulfite oxidation in oxidation tank, %
Sulfite oxidation in scrubber inlet slurry, %
Gypsum saturation in scrubber inlet liquor, %
Gypsum saturation in oxidation tank, %
Effective Mg++ concentration in scrubber inlet liquor, ppm
Oxidation tank liquor sulfite concentration, ppm
Air stoichiometry, atoms 0/mole S02 absorbed
Filter cake solids, wt* '2^
Mist eliminator restriction, % '3)
823-1A
High
18,000
600
1600
15
11.2
5.3
-
5000
9
18
110
30

205
94
2600
13.3
5.25
6.30
36
98
86
120
115
4990
65
1.55
83
0
824-1A
High
35,000
600
1600
15
11.2
-
1.9
5000
9
18
210
30

159
88
2600
14.1
5.25
5.90
38
97
49
85
90
5215
105
1.60
85
0
825- 1A
High
18,000
600
1600
15
11.2
-
1.4
5000
9
18
110
0

229
95
2500
14.7
5.45
5.65
64
97
39
105
115
5380
220
1.60
85
0.5
826-1A
High
26,500(4)
600
1600
15
11.2
-
1.4
5000
9
18
210
0

246
89
2750
15.2
5.35
5.45
61
96
29
105
115
4970
230
2.00
84
0.1
Notes:
     1) Air discharged downward through 3-inch diameter pipe with an open elbow at center of oxidation tank about 3 inches from tank bottom,
     2) Clarifier and filter in series used for solids dewatering in all runs,           ,,          2
     3) Continuous mist eliminator bottom wash with diluted clarified liquor at 0,4 gpm/ft  (0.3 gpm/ft  for Run 823-1A),   Sequential top wash
        with makeup water with one of 6 nozzles on at 0.53 gpm/ft^ for 4 minutes every 80 minutes.
     4) Desired flow rate was 35.000 acfm but problems with the venturi lifting mechanism limited the rate to 26,500 acfm.

-------
Additional testing will  be conducted to fully characterize  the limestone/MqO
bleed stream oxidation system.  Better operational  control  is required to improve
the limestone utilization.  Higher limestone utilization, however, is not expected
to have an adverse effect on the oxidation  efficiency  because of the reduced amount
of residual  alkali and the correspondingly  less  possibility of pH rise in the
oxidation tank.

The possibility  of bleed  stream oxidation on a lime/MqO system must also be investi
gated.
                                   46

-------
                               Section 5


         DEWATERING CHARACTERISTICS OF THE OXIDIZED SLURRY SOLIDS
The settling and dewatering characteristics of slurry solids are  routinely
monitored in the Shawnee laboratory by cylinder settling tests and  vacuum
funnel filtration tests.  Results of these monitoring tests  are presented in
this section.  The test results are summarized in Table 6.


Results of sludge disposal  studies at the Shawnee Test Facility are presented
on a separate paper by the Aerospace Corporation.


Cylinder settling tests are performed in 1000 ml  cylinder containing a  rake which
rotates at 0.16 rpm.  The initial settling rate and ultimate settled solids con-
centration are recorded as indices of dewatering characteristics.   The  initial
settling rate is a qualitative index of the solids settling  properties  only.  De-
sign rates for sizing clarifiers must take into consideration the hindered settling
rate as the solids concentrate.  The ultimate settled solids from the cylinder
tests represent the highest achievable solids concentration  in a  settling pond.


Funnel filter tests are performed in a Buchner funnel with a Whatman ?.  filter  paper
under a vacuum of 25 in. Hg.  The funnel tests correlate well with  the  Shawnee
rotary drum vacuum filter when not blinded but the funnel test cakes tend to have
lower solids concentrations.


As can be seen in Table 6 the benefits of forced oxidation are clear, the dewatering
characteristics of oxidized sludge are markedly better than  those of unoxidized
sludge.  The initial settling rate is higher by a factor of  4, and  both the settled
and filtered solids concentrations are higher by a factor of 1.4.


Without forced oxidation, the average initial settling rate  was about 0.2 cm/min.
With forced oxidation the average initial settling rate was  higher, ranging from
0.42 cm/min. to 1.20 cm/min.


The presence of magnesium ion tended to decrease the initial settling rate  -  slightl
with oxidized slurry and more with unoxidized slurry.  For oxidized slurrv   in  the
2-loop mode of oxidation, the average initial settling rate was 1.0 cm/minute
                                    47

-------
                                                                      Table 6
                                                 SUMMARY OF  THE  DEWATERING CHARACTERISTICS

                                                          OF THE  SHAWNEE WASTE  SLURRY
OO
Oxidation
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
No
No
No
Ho
No
No
Fly Ash
Loading
High
High
High
High
High
Low
Low
High
High
High
High
High
Low
Low
Alkali
LS
LS
LS
LS
L
LS
L
LS
LS
LS
L
L
LS
L
Onldatlon Hade
1-loop
2-loop
Bleed Stream
2-loop
2-loop
2-loop
2-loop
-
-
-
-
-
-
-
Initial Settling Rate. ca0i1n|uUlMte Settling Solids, wtt
Avg. | Range | Hvg. | Range
1.06 0.63-1.27 74 67-84
0.70(1) 0.55-0.8711'
1.20(Z1 0.96-1.41(2> 72 62-86
1.12(3) 0.81-).47(3)
0.42 0.29-0.61 71 61-84
0.75 0.23-1.19 66 46-73
0.98 0.77-1.23 73 61-85
0.88 0.61-1.15 74 61-87
1.20 0.45-2.44 70 60-81
D.20 0.07-0.54 54 41-67
0.20 0.07-0.37 45 30-60
0.05 0.01-0.11 41 32-46
0.20 0.19-0.49 50 48-66
0.79 0.22-1.15 42 31-52
0.17 0.05-0.47 43 33-54
0.35 0.09-0.87 40 30-55
Funnel Test Cake Solids. wtS Slui
Avg. | Range Sol
76 73-80 16
72 65-88 15
73 71-76 15
70 46-76 15
71 64-78 15
73 64-82 15
76 64-83 15
57 48-66 15
57 45-64 15
55 47-69 15
53 51-55 15
52 43-63 8
50 41-59 15
45 40-50 8
'ry Effective Hg
ds Concentration, ppn
0
0
5000
8000 l
0
0
0
0
5000
9000
0
2000
0
0
                Note: Values for forced oxlditlon runs are only from data where solids oxidation
                     was greater than or equal to 90 percent,
                     (1) Ox1d1«r pH
                     (Z)
              4.5
   OxIdUer pH « 5.0
(3) Oxldlzer pH - 5.5

-------
without magnesium and 0.75  cm/min. with 8000 ppm effective magnesium ion concentra-
tion.  For unoxidized limestone  slurry with high fly ash loading, the average initial
settling rate was reduced from 0.20  cm/min. without magnesium to 0.05 cm/min. with
9000 magnesium ion, a decrease by a  factor of 4.  This magnesium effect is probably
the result of an increase in liquor  viscosity and density due to the increased
amount of total dissolved solids.

Figure 8 is a plot of percent oxidation versus  settling rate for a lime system with
low fly ash loading. The presence of fly  ash in oxidized slurry appears to decrease
the settling rate slightly.  For example, oxidized  lime slurry with low fly ash
loading has an average initial settling rate of 1.20 cm/min., whereas oxidized lime
slurry with high fly ash loading has an average initial settling rate of 0.98 cm/min.

Similar results were obtained with the ultimate settled solids and the funnel test
cake solids.  Without forced oxidation, the ultimate settled solids were generally
in the range of 40 to 50 weight  percent solids; with forced oxidation, the range was
65 to 80 weight percent.  Funnel test results  indicated 45 to 60 weiqht percent solids
without forced oxidation and 65  to 85 weight  percent with forced oxidation.  On the
rotary drum vacuum filter used in the scrubber dewatering system, cake solids con-
centration was always above 80 percent with  oxidized slurry while averaging  50 to 60
percent with unoxidized slurry.

The  unoxidized solids tended to be thioxotropic,  like  quicksand, while the oxidized
solids were more  like moist soil.

The  presence of magnesium  ion did not affect  the  ultimate  settled solids  or  the
funnel test cake  solids  results.  This result was also seen  in  the  test  facility's
operating  data.

The  product solids  at Shawnee are a mixture of unoxidized  calcium  sulfite hemihydrate
 (CaS03'l/2H20),  the oxidation product calcium  sulfate dihydrate (CaSO^Zt^O),  fly
 ash, unreacted alkali,  and  other inert materials.  In slurry that  is 10 percent  oxi-
 dized,  the main  component  is  calcium sulfite in platelets and rosettes of only a few
microns  in diameter.  The  solids in slurry 95  percent oxidized  are  mainly calcium
 sulfate crystals having a  bulky rectangular shape and ranging in size from 20 to
 100  microns.

 It is currently thought that  in the case of unoxidized slurry,  the calcium sulfite
 fines are the limiting  factor in the  inital settling rate.  Rut in the case of oxi-
 dized slurry,  the fly ash  may be the  limiting  factor.  Fly ash is extremely fine
 when compared with the  calcium  sulfate particles and hence settles at a slower rate.
 At the Shawnee Test Facility, the  limit  for oxidized slurry with high fly ash
 loading appeared to be  1.5 cm/min.   For  oxidized slurry with low fly ash loading,
 the limit was 2.4 cm/min.   Since,  at  Shawnee,  solids from flue gas with low fly ash
 loading contain up to 1 weight  percent fly ash, the limits of the calcium sulfate
 settling rate may be even higher.

 The residence time in the oxidation tank may  also  affect the size of the gypsum
 crystal; the longer the residence time the larger  the crystal.  This effect has yet
 to be thoroughly explored at Shawnee.
                                    49

-------
                       FIGURE 8

  EFFECT OF OXIDATION ON INITIAL SETTLING RATE
  2.5
   2 -
e
E
  1.5 -
Z
-I

UJ
CO
1  -
   .5 -
         10   20   30   40   50   60   70
                    %SULFITE OXIDIZED
                                    80
90   100
                       50

-------
                                 Section 6
                              FUTURE TESTING

Testing with forced oxidation will  be continued with emphasis on:

     •  More fully developing  forced oxidation within  a  single  scrubber loop
     •  Exploring the conditions under  which  bleed  stream oxidation  is applicable
     t  Determining compatibility of forced oxidation  with chemical  additives
        such as adipic acid
Based on the encouraging results at the IERL-RTP pilot plant, an extensive program
to develop adipic acid as an additive for enhancing S02 removal  efficiency has re-
cently been initiated.  Adipic acid acts as a buffer to limit the drop in pH,
thereby improving the liguid-phase mass transfer.  The advantages of adipic acid
are  listed below:
     •  Lower cost compared to MgO based on the quantity needed.  For example, for
        a similar degree of Sr>2 removal enhancement in limestone scrubbing:
              With MgO:  6,000 ppm Mg++ at $0.17/lb MgO requires
                         $14/1,000 gal of discharged  liquor
              With adipic acid:   1,000 ppm adipic  acid at $Q.42/lb
                         acid  requires only $3.50/1,000 gal  of
                         discharged liquor
      • Optimum adipic  acid concentration for  effective improvement in S02 removal
         is  only 5-10  m-moles/liter (700 - 1,500  ppm)
      •  Adipic  acid  improves  SO?  removal  and may also improve 1imestone  utilization,
         whereas MgO  may reduce the limestone dissolution  rate
      t  In the  limestone scrubbing system,  S02 removal  efficiency  is  no  longer
         limited by the limestone  dissolution  rate when adipic  acid is present in
         sufficient quantity

                                     51

-------
     •  Forced oxidation does not  affect  the  effectiveness of adipic acid.
        Forced oxidation reduces the effectiveness  of the maqnesium ion by
        converting the scrubbing S03= into  non-scr.ubbinq S04= species

     •  Unlike MgO addition,  where two chloride  ions tie up a maqnesium ion
        to form neutral  MgC^, adipic acid  is not affected by chloride

     •  Adipic acid is nontoxic (used as  a  food  additive)

     •  Both IERL-RTP pilot plant  and preliminary Shawnee results show that
        the solids quality (filterability,  settling rate) is not affected by
        adipic acid


Tests with adipic acid as an  additive are scheduled with and without forced
oxidation and with both lime  and limestone  slurries.


Tests are also planned to investigate the effect of limestone type and grind on
SO? removal and limestone utilization.  Initial  screening tests have already been
scheduled for the IERL-RTP pilot plant.
A 3-month test block is planned for the Shawnee spray  tower with  various  internal
configurations (different number of headers, number of nozzles, type of nozzles
nozzle pressure drop, etc.).   The primary objective of the testing will be to
provide a better basis for designing full-scale spray  towers.
Long-term (over one month) lime and limestone tests  are  also  planned, which will
combine the most promising operating conditions,  including forced  oxidation and
organic acid addition, to demonstrate system reliability and  conformance to the
existing Federal emission standard.


Other concurrent future activities include:


     •  Transfer of Shawnee-developed technology  to  full-scale  plants,  including
        if necessary, simulation of commercial  plant operation  at  Shawnee        '

     •  Continued development and updating of the Economic Study Computer Program
        in conjunction with TVA

     •  A study of the overall  power plant water  management as  it  relates to FRO
        plant operation
                                  52

-------
                              Section 7

                              REFERENCES
1.  Barrier,  J.W.  et  al,  Comparative Economics of FGD Sludge Disposal.
    Presented at the  71st Annual Meeting of the Air Pollution Control
    Association, Houston, Texas, June 25-30, 1978.

2.  Borgwardt, R.H.,  Sludge Oxidation in Limestone FGD Scrubbers
    EPA-600/7-77-OG1, June 1977.     ~~~

3.  Head, H.N. et al, Results  of Lime and Limestone Testing with Forced
    Oxidation at the  EPA  Alkali Scrubbing Test Facility. Proceedings:
    Symposium on Flue Gas Desulfurization - Hollywood, Fl., November 1977
    EPA-600/7-78-58a, March 1978.                                        '

4.  Bechtel  Corporation,  EPA Alkali Scrubbing Test Facility: Summary of
    Testing  through October 1974.  EPA 650/2-75-047, June 1975.

5.  Bechtel  Corporation,  EPA Alkali Scrubbing Test Facility: Advanced
    Program.  First Progress Report. EPA-6no/2-75-050, September  1975.

6.  Bechtel  Corporation,  EPA Alkali Scrubbing Test Facility: Advanced
    Program.  Second Progress Report. EPA-600/7-76-008, September  1976.

7.  Bechtel  Corporation,  EPA Alkali Scrubbing Test Facility: Advanced
    Program.  Third Progress Report. EPA-600/7-77-105, September  1977.
                                    53

-------
         SELECTED TOPICS FROM

    SHAWNEE TEST FACILITY OPERATION
              Presented bv
              David T. Rabb
        EPA Program Site Manager
         Bechtel National, Inc.
            50 Beale Street
     San Francisco, California 94119
                 at the

         EPA Industry Briefing
 Research Triangle Park, North Carolina
            August 29, 1978
          EPA Contract 68-02-1814
             John E. Williams
             Project Officer
Industrial  Environmental  Research Laboratory
      Office of Research  and Development
Research Triangle Park, North Carolina  27711
                54

-------
.LNTROBUCTI-ON


The Shawnee Test Facility is  an EPA funded  wet  lime/limestone  scrubbing test
facility that has operated since 1972.   The facility  currently consists of two
10 MW equivalent scrubbing systems each  treating  approximately 7  percent of the
flue gas produced by a medium-to-high sulfur coal-fired  150  MW boiler.  One
system is a venturi  followed  by a spray  tower (V/ST)  the other is a Turbulent
Contact Absorber (TCA).

The information presented here reflects  selected  topics  from the  operating
experience obtained between June 1977 and May 1978 at this facility.  These
include:

      a)  Scrubber operation and maintenance

      b)  Oewaterinq systems

      c)  Forced oxidation systems

      d)  Automatic control of limestone addition

      e)  Operational development
 SCRUBBER OPERATION -AND MAINTENANCE


 SCRUBBER OPERATION


 Over the past year the V/ST and the TCA have maintained a high operational
 availability that  is  summarized in Table 1.  The V/ST system operated 7040
 hours during the year or 80 percent of the time.  The TCA system operated 7272
 hours or 83 percent of the time.  The systems downtimes were attributed to
 four categories:

       a)   Boiler outages

       b)   Weather  affecting scrubber operations

       c)   Scheduled scrubber  system inspections and modifications

       d)   Scrubber equipment  failures


 The unit  10  boiler operated 91  percent  of  the time and consequently caused 9
 percent downtime  for  each  of  the  two  systems for the year.  Of the over ROO hours
 of boiler  downtime approximately  350  hours were the result  of a scheduled major
 overhaul.  The hours  of  downtime differ  for the two systems  because the V/ST and TCA
 had to start-up separately due  to operating  personnel  limitations.
                                        55

-------
                                 TABLE 1
                       SCRUBBER OPERATION
OPERATING TIME. Hours

DOWN TIME, Hours

    BOILER & DOE
    WEATHER
    INSPECTION & MODIFICATION

    EQUIPMENT PROBLEMS
TOTAL PERIOD*. Hours
                             VENTURI/SPRAY TOWER
7040 (80%)
                         TCA
                                                              7272 (83%)
832(9%)
220 (3%)

421 (5%)
247 (3%) ^
810 (9%)
171 (2%)
) 1720 (20%)
323 (4%)
184(2%)^


) 1488(17%)


8760 (100%)
8760 (100%)
*JUNE 1977 THROUGH MAY 1978

-------
The second major cause of downtime  was  attributed to  routine scrubber inspec-
tions and system modifications inherently  associated  with the goals of this test
facility.  Typically a test  would last  5 to  6  days with occasional runs lasting
as long as 9 days to insure  sufficient  steady  state data.  The time spent between
test runs varied from a few  hours for a scrubber inspection to 3 or 4 days for a
system modification.  The factors caused downtimes for the year of 5 percent for
the V/ST and 4 percent for the TCA.


The remaining two causes of  outages  were   scrubber equipment failures and
weather related problems. Equipment failures  and subsequent maintenance
will be discussed in depth in  the next  section.  The  weather related problems
were the result of the abnormally severe winter experienced in Kentucky last
January.  Alkali addition streams were  the most troublesome in that the one-
inch utility hoses used as alkali addition lines froze repeatedly.  Maintenance
was seriously hampered by the  cold.  The downtime for the year caused by the weather
was 3 percent for the V/ST and 2  percent for the TCA.


Curtailed power generation because  of scrubber inavailability is of serious concern
to utility companies.  For this paper,  scrubber availability is introduced and is
defined as the percent of time that  a commercial scrubber operates while the
accompanying boiler is operating.   Scrubber  equipment failures and weather created
problems were used in calculating the availability term at Shawnee.  Modifications
and inspections due to test  requirements were  not included because they were in-
consistent with the concept  of a  commercial  installation.
For the V/ST and TCA, the average availability for the  last year was 94 and 95
percent, respectively.   On a monthly basis  the availability ranged from R2 to
100 percent for the V/ST and 81  to 100 percent for the  TCA.
MAINTENANCE

At the Shawnee Facility, the high availability  has  been strongly  influenced by
an effective maintenance program and an adequate spare parts  inventory.  These
factors are particularly important when considering pump maintenance; the
facility does not have stand-by spare pump capacity as is common  practice  in
industrial systems.  The total  hours and frequency  associated with equipment
maintenance that resulted in system shut-downs  are  outlined in  Tables 2  and 3
for the V/ST and TCA, respectively.


In the V/ST system the primary cause of downtime was scrubber related problems
that consisted, in part, of a corroding air sparge  pipe, pipe plugging and
leaks, nozzle plugging, and solids build-up in  the  outlet ductwork!  The
corroding sparge pipe occurred only once and was caused by mistakenly using
304 stainless steel as part of the material of  construction.
                                       57

-------
                                        TABLE 2

                  V/ST EQUIPMENT MAINTENANCE CAUSING DOWN TIME

                                JUNE 1977 TO MAY 1978
Ul
00
ITEM



SCRUBBER INTERNALS AND PIPING



I. D. FAN



INSTRUMENTATION


PUMPS



ALKALI FEED


OTHER



TOTAL
                                             FREQUENCY OF EVENT
TOTAL HOURS
12
4
1
2
1
1
114
78
39
12
3
1
                                                   21
                                                    247

-------
                                          TABLE 3

                    TCA EQUIPMENT MAINTENANCE CAUSING DOWN TIME

                                  JUNE  1977 TO MAY 1978
Ul
VO
ITEM


SCRUBBER INTERNALS AND PIPING



AGITATOR


PUMPS


ALKALI FEED



INSTRUMENTATION


OTHER
                                              FREQUENCY OF EVENT    TOTAL HOURS
                                                     6
108
 39
4
2
1
2
17
8
7
5
                 TOTAL
                                    16
                                                                    184

-------
 The longest downtime requirement for the TCA was  the  installation of the
 Penberthy eductor.  The agitator shaft replacement  in the main hold tank was
 second, followed by routine maintenance of  the main slurry pump which consisted
 of repacking Allen-Sherman-Hoff Centri-Seals.  Other  operating problems included
 plugging of the alkali  feed lines,  bearing  failures of the I.n. fan and general
 solids build-up in the outlet duct.


 The personnel  available for maintenance requirements  and system modifications
 are listed in Table 4.   To avoid possible conflicts of priorities between the
 powerhouse and scrubbing facility and to allow the crafts people to develop
 expertise with specific equipment,  the maintenance personnel are assigned exclu-
 sively to the test facility.
 DEWATERING SYSTEMS

 The primary dewatering  of the  purged slurry in both scrubbers is achieved bv
 clarifiers. further dewatering of the clarifier sludges is accomplished in the
 V/ST system by  a filter and in the TCA system by a centrifuge.  The information
 discussed below  represents actual data that might be used as a guide in understand
 ing the  operating problems and costs associated a with centrifuge or filter.
 CENTRIFUGE


 A continuous centrifuge is one process used to dewater scrubber waste sludge
 and  to  recover the dissolved scrubbing additives.  The normal  operating condi-
 tions usually consist of a feed stream flow of 15 gpm at 30 to 40 wt. percent
 solids,  a centrate of 0.1 to 3.0 wt. percent solids, and a cake of 55 to 65  wt,
 percent  solids, for unoxidized slurry.  Approximately 30 percent of the total
 solids  is fly ash; the remaining solids are predominantly calcium sulfate
 and  sulfite.


 The  machine is a Bird 18" x 28" solid bowl continuous centrifuge which operates
 at 2050  rpm.  The material of construction is 316L stainless steel  with stellite
 hardfacing on the feed ports, conveyor tips and solids discharge ports.   The howl
 head plows and case plows are replaceable.  The pool depth is  set at  1-1/2 inches
 No cake washing is performed in this machine.


 The  centrifuge was inspected in June 1978, after 6460 hours of operation since
 the  previous factory servicing. The inspection was prompted by the  gradual and
 continued increase of centrate suspended solids to a level  of  approximately  3
wt.  percent.  The machine was judged to be in generally fair condition but
 certain components were in need of factory repair.  Serious wear was  observed
 at the conveyor tips on the discharge end and at the junction  of the  cylinder
 and  the 10*  section of conveyor.   Wear was also present at  the casing head plows
and  solids discharge head near the discharge ports.   The bowl  and effluent head
were in good condition.
                                     60

-------
                   TABLE 4



            SHAWNEE FACILITY



         MAINTENANCE PERSONNEL






 I  MAINTENANCE FOREMAN                       1



   BOILERMAKER-WELDER                       2



   CARPENTER                                1



   ELECTRICIAN                               1



   PIPEFITTER-WELDER                         2



   TEAMSTER                                 1



   LABORERS                                 2



   HEAVY EQUIPMENT OPERATOR                   1



   PAINTER                                   1



   MACHINIST                                 1




   INSULATOR                                 1





II  INSTRUMENT FOREMAN                        1



   SENIOR INSTRUMENT MECHANIC                  2




   JOURNEYMAN INSTRUMENT MECHANIC             2





III  TOTAL                                    19
                      61

-------
                            TABLE 5

      CENTRIFUGE MAINTENANCE & POWER REQUIREMENT

                     JUNE 1977 TO MAY 1978
   MAINTENANCE

                                       ESTIMATED        ESTIMATED
                                      TOTAL ONSITE    TOTAL MATERIAL
                      FREQUENCY OF         LABOR            COST
   EVENT               OCCURRENCE        (MAN-HOURS)     	($)	



   FEED PIPE REPAIR             1               16              20
II  POWER REQUIREMENT - 30 HORSEPOWER

-------
The current plans call  for  the  centrifuge to be shipped to the factory for an
overhaul.  The following items  will  he accomplished:


      a)  Inspect and service the  gear and bearing unit

      b)  Rebuild all worn  conveyor  surfaces and add hardfacing on the tips

      c)  Rebuild and add hardfacing to the discharge ports

      d)  Replace all seals and bushings  in the effluent and discharge head

      e)  Replace case plows and discharge plows as necessary
In an attempt to improve performance and  machine  life, tungsten carbide
hardfacing wi.ll be applied to the conveyor tips instead of the previously
used stellite.  The estimated cost for the complete  factory service, including
the hardfacing, is $17,000.


The machine has been a minimum maintenance item;  as  shown in Table 5 the only
maintenance of the past year has been the replacement  of the feed pipe in the
centrifuge.  The power requirement for the machine is  30 horsepower.
FILTER


An Ametek 31 x 6' vacuum drum filter without  cake wash  is  operated at the facility
for waste sludge dewatering and dissolved scrubbing additive recovery.  The feed to
the filter is usually 15 gpm of 30 to 40 wt.  percent solids.


The filtrate generally contains less than 0.02  wt. percent solids.  The filter
cake varies from 55 to 85 wt. percent solids  depending  mainly on whether the
sludge is unoxidized or oxidized.


The filter, with the exception of the filter  cloth, has been a moderate
maintenance item.  Table 6 is a breakdown of  maintenance categories, frequency,
approximate total manhours required, and approximate replacement material
costs.  Also, included in Table 6 is the power requirement for the filter and
vacuum pump.


Contrary to experiences at other scrubbing facilities,  filter cloth  replacement
as noted in Table 7 has been a serious problem at Shawnee.  The causes  of cloth
blinding and fraying are not satisfactorily understood  as yet.  However,  in the
last few months  operating experience indicates a relationship between  cloth
                                       63

-------
                                TABLE 6
            FILTER MAINTENANCE & POWER REQUIREMENT
                       JUNE 1977 TO MAY 1978
 I  MAINTENANCE
   EVENT
   SPEED CONTROL REPAIR
   CAKE DISCHARGE AIR REPAIR
   CLOTH REPLACEMENT
FREQUENCY OF
OCCURRENCE
    15
   ESTIMATED
  TOTAL ONSITE
LABOR (MAN-HOURS)
  ESTIMATED
TOTAL MATERIAL
    COST ($)
2
2
16
16
1000
0
                                                90
                                    1500
II  POWER REQUIREMENT - 20 HORSEPOWER

-------
CLOTH TYPE*
                          TABLE 7
                 FILTER CLOTH SERVICE
DATE INSTALLED
HOURS IN SERVICE
COMMENT
1
2
3
4
5
6
7
8
9
10
11
12
13
14

15
TFI
LAMPORTS
LAMPORTS
LAMPORTS
LAMPORTS
LAMPORTS
TFI
TFI
LAMPORTS
AMETEK
AMETEK
LAMPORTS
LAMPORTS
AMETEK

AMETEK
6-6-77 292
7-7
7-13
7-19
8-3
8-15
9-19
9-23
10- 18
127
142
302
249
540
187
501
187
10-26* 2096
2 - 6 78 188
2-14
2 22
3-21

4-19
190
290
535
•
1197
BLINDED
HOLE IN CLOTH
BLINDED
BLINDED
BLINDED
BLINDED
BLINDED
BLINDED
BLINDED
HOLE IN CLOTH
BLINDED
HOLE IN CLOTH
BLINDED
BLINDED

HOLE IN CLOTH
*AMETEK - AMETEK OLEFIN (STE - F9D8 - HJO)
 LAMPORTS- LAMPORTS POLYPROPYLENE (7512 - SHS)
 TFI -TECHNICAL FABRICATORS INCORPORATE, POLYPROPYLENE (9162)
                            65

-------
life and the technique by which  the  cloth  is fitted to the drum.  A carefully
controlled amount of looseness in the fit  between the dividers appears to be
desireable for cake discharge  and non-blinding.  The looseness evidently allows
the cloth to "snap" the cake off when the  air puff of the cake discharge cycle
is applied to the given filter cloth section.  Two additional observations have
been that oxidized sludge exhibits less tendency towards cloth blinding and
Ametek olefin appears to provide the most  satisfactory service of these cloths
tested.  The reason for the better service life of Ametek is suspected to be
attributable to the looseness  of weave that the Ametek has in comparison with
Lamports and TFI.

As noted, some limited progress  has  been achieved towards understanding and re-
ducing cloth blinding.  Future operations  at Shawnee will include continued efforts
to improve filter cake performance.
OXIDATION -SYSTEM-DISCUSSION


The desirability of an oxidized  calcium base sludge has for a number of years
been recognized when considering ease of  sludge filtration, handling and disposal
Forced oxidation has been determined  to be one method in achieving those ends.  i
forced oxidation, air is introduced and dispersed into the slurry as required to
oxidize the calcium sulfite to calcium sulfate.


At Shawnee, forced oxidation has been investigated in a series of tests using
a Penberthy eductor and an air sparger.   The intended goal of these tests was
to investigate the operating parameters necessary to achieve "near complete"
oxidation,  i.e., greater than 90 percent  total sulfur as sulfate.

One important operating experience resulting from these tests was a tight water
balance with no observable deterioration  of mechanical components or scruhbinq
chemistry.
PENBERTHY EDUCTOR


The Penberthy eductor is  a  device  similar in concept to a laboratory aspirator.
A high velocity slurry passes  through  a constricted nozzle, into an eductor
chamber and then through  a  moderately  restricted jet throat.  In the chamber
the slurry induces a vacuum that draws ambient air into the chamber via an entry
pipe located at right angles to the  slurry flow path.  The air is then entrapped
and mixed in the slurry as  the fluid leaves the eductor chamber and passes throuah
the jet throat.                                                                *n

At Shawnee, a Penberthy Model  ELL-10 eductor was tested.  The materials of con-
truction were stellite for  the nozzle  and neoprene-lined carbon steel for the
eductor chamber and exit  jet throat.  The system was operated at 1600 gpm.
                                      66

-------
Test results indicated that  near  complete  oxidation could be achieved, hut at
the same time serious erosion  problems  developed.  The neoprene lininq in the
jet throat was observed to be  "chipped  off" after only 620 hours of operation.
After approximately 1800 hours of operation,  bare carbon steel was exposed, and
after an epoxy patch failed  after 2055  hours  of operation, the tests were
terminated.

Currently, no plans exist to resume  testing of eductors primarily because no
advantages in oxidation capabilities were  seen in comparison with sparqe air
systems. Secondary reasons included  the materials erosion problem and the unfa-
vorable operating and capital  costs  in  comparison to  sparge air systems.
AIR SPARGER


The sparge atr system used in sludge oxidation  is  a  simple concept that involves
bubbling air into the bottom of a slurry tank  in conjunction with simultaneous
slurry agitation.  At Shawnee the oxidation tanks  are  purposely tall and thin
(7 to 8 feet diameter x 20 feet tall).   This shape enables a long air/slurry
contact time.

In the TCA an octagonal sparge ring with 40 holes  is currently installed at the
bottom of the effluent hold tank.  The holes are 1/4 inch diameter and located
on the underneath side of the ring; no sparge  ring plugging had occurred to
date.  The ring is constructed of 316L stainless steel'and is materially in
good condition except for minor erosion at the  1/4 inch air holes. Currently
a conventional tank agitator (37 rpm, 3 hp) is  in  service.  Future tests will
use a variable, high speed agitator.  Near complete  oxidation has been achieved
with the current configuration.

Testing was done on the V/ST system with an octagonal  sparge ring similar in
design to that of the TCA.  After 2400 hours of operation the ring was removed
from service and replaced with a 3-inch diameter open-ended sparge pipe.  The air
and slurry are mixed with an axial flow agitator (56 rpm, 20 hp).  Near complete
oxidation has been obtainable with both the sparge ring and sparge pipe in con-
junction with the agitator.

Both V/ST and TCA sparge systems operate from the  same Worthington oil-free air
compressor under the following conditions: 50  psig,  270 F, and flow  rates normally
at 210 scfm per scrubber.  The compressor loading/unloading cycle is normally
4 seconds/10 seconds for 210 scfm.  The oil-free compressor was  chosen to minimize
the possibility of slurry contamination by oil  which contains oxidation inhibitors.

Maintenance on the sparge systems has been mainly  confined to the compressor.   The
suction valve has been serviced and the compressor cooling water jacket has  been
flushed periodically.

Further testing of forced oxidation with the sparqe  ring and  sparge pipe  configura-
tions is planned.
                                      67

-------
 AUTOMATIC LIMESTONE  ADDITION CONTROL


 An automatic limestone feed control system was installed and initially tested on
 the TCA in April-June 1977.  The control loqic was based on a material balance
 concept of maintaining a desired stoichiometric limestone feed in relation to the
 amount of S02 absorbed in the scrubber.  The results of the tests indicated that
 limestone addition was satisfactorily controlled durinq normal fluctuations of
 S02 inlet mass flow  rates.  As will be discussed, one observed weakness of the
 control  system was that the stoichiometric ratio was independent of the SO? inlet
 mass flow rates and  therefore the control system could not effectively compensate
 for unusually large  fluctuations in S02.


 The basic control scheme is represented by the following equation and  by Figure 1


           L=Gx(S-K)xR

 where:     L =  limestone slurry addition rate, gpm

           G =  flue gas flow rate, acfm

           S =  inlet S02 concentration, ppm

           K =  a manually adjustable constant  related to destred outlet
               S02 concentration, ppm

           R =  a manually adjustable constant  proportional  to the desired
               stoichiometri c rati o

            =  (unit conversion factor) x (stoichiometric ratio)
                        Q
            =  2.34 x 10    x  (stoichiometric ratio)


 The  factor G(S-K) represents the amount of S02 absorbed per unit time.   Thus
 at a set  value of R which is proportional  to  the  desired stoichiometric  ratio,
 the  limestone  addition rate  is automatically  adjusted tc maintain the  desired*
 stoichiometry.

 In practice, the inlet S02 concentration,  S,  and  the flue qas  flow rate, G
may  vary within a wide range depending on  the sulfur content  in  the coal and
the boiler load.  Therefore, the control  scheme also includes  overrides which
are activated when the following situations arise:


      a)  If the measured outlet S02 concentration exceeds  a set  maximum, the
          limestone addition rate will  be  stepped up to a  preset  maximum!

      b)  If the measured outlet S02 concentration drops  below a  set minimum
          the limestone addition rate  will be maintained  at a  preset minimum!
                                       68

-------
                                   FIGURE 1
                        ALKALI FEED CONTROL SCHEME
               SCRUBBER
               INLET GAS
                  SCRUBBER
                  OUTLET GAS
                        INLET
                      GAS FLOW
                         FE
DESIRED
OUTLET
  S02
  HIC
               LIMESTONE
                  TO
               SCRUBBER
1
                                    PUMP
                                                                       /OUTLET\
                                                                       (SO2 LIMIT?

-------
The former provision insures compliance with the SOg emission  standard,
while the latter reduces the posibility of pipe plugging.


Operating under typical conditions of 30,000 acfm, 1200 gpm liquor rate, and 3
beds each with 5 inches static height nitrile foam spheres, the  desired'limestone
stoichiometric eatio for the TCA was set at 1.35 moles  Ca  added/mole S02 absorbed
(R = 2.34 x 10"°  x 1.35 = 3.16 x 10"8).  Actual stoichiometric  ratio varied
between 1.18 and 1.55, with an average of 1.37.  Actual  limestone addition rate
was generally within 10 percent of the rate calculated  from the  control equation
presented above.  A desired outlet S02 concentration, K, was set at 430 ppm.


Initially, overrides were set at 500 ppm and 200 ppm outlet S02  concentration.
Because of wide variation in the inlet S02 concentration (2100 to 3400 ppm),
the outlet S02 concentration frequently exceeded the 500 ppm upper override*!imit
This resulted in the actuation of the override control  limit and interferred with'
the testing of the major control logic.  Subsequently,  the upper override limit
was raised to 800 ppm and the proportional  control  functioned more smoothly.


In theory, the higher the inlet S02 concentration, the  higher the required
percent S02 removal  in order to meet the S02 emission standard.   The higher
required percent S02 removal would, in turn, call  for higher limestone stoich-
iometric ratio.  In the present setup, the constant R can  only be reset manually
Therefore, as a second generation control  logic, R as a function of the inlet
S02 concentration would be desirable.
OPERATIONAL -DEVELOPMENT


In the 6 years of operation the intent  of  the test facility has been to accelerate
the development and application of lime/limestone scrubbing technology.  In addi-
tion to better understanding the chemistry of scrubbing, improving the sludge
disposal properties and enhancing S02 removal,  countless developments and improve-
ments have been made in the operation of the facility.  A few are listed here:

   a) Mist eliminator plugging at Shawnee  is extremely infrequent.  Wash patterns
      and wash sequences have been refined to such a degree that an excess of 7700
      hours of operation was obtained recently  between cleanings on the TCA mist
      eliminator.  The reason for the cleaning  was that 12% plugging was present
      and a new series of tests using adipic acid were to begin.  The V/ST mist
      eliminator was cleaned only once  during the last 8000 hours of operation due
      to problems with the wash system  that caused 15% plugging.  Considering the
      wide range of operating conditions for the systems, these are outstanding
      records.

   b) The variable speed Allen-Sherman-Hoff rubber lined pumps have demonstrated
      their effectiveness as slurry pumps  under a wide range of operating condi-
      tions.
                                     70

-------
  c)  The  process  pH  is measured using Uniloc Model 321 submersible electrode
     assemblies.  Originally, Uniloc Model 320 flow-through meters were installed.
     But  because  of  line plugging problems and frequent sensing electrode breakage
     this type  of sensor was abandoned.  Current service requirements for the sub-
     mersible assemblies consist of periodic cleaning and buffering of the electrodes,
     generally  every 2 or 3 days to insure accuracy.  Also to minimize the service
     requirement, the instrument electrodes are placed in water when the scrubbers
     are  not operating.  Years of operation have shown that to insure the accuracy
     of the process  pH meters, a laboratory measured pH should be taken once every
     four hours for  comparison purposes.  This procedure enables a normal operation
     to within  _+  0.2 pH units of any desired set point.

  d)  A Dupont Model  400 UV split-beam photometer is used to measure S02 concentrations
     In the last  few years the instrument has been accurate and reasonably trouble-
     free. Maintenance requirements are limited to cleaning the sample cell and
     sample lines approximately once every 1 or 2 months and cleaning the particulate
     filter usually  once every 3 to 4 weeks.  Ultraviolet lamp failure has been
     the  only component problem and has been caused by uncontrollable and
     momentary  power fluctuations due to the switching of station power.  The
     effective  particulate filter for the instrument at Shawnee is a cylindrical
     chamber constructed of a fine mesh screen.  The screen cylinder is surrounded
     by a solid protective cylinder.  The gas sample lines have operated leak free;
     the  lines  are 316L stainless steel tubing with heat tracing.

  e)  Oxygen in  the inlet flue gas is measured with a Teledyne Model 9500 which uses
     a micro-fuel cell.  Operating performance has not been acceptable.   A fre-
     quent problem has been the rapid deactivation of the special micro-fuel cells.
     Service life has varied from one day to approximately 1-1/2 months.  In some
     cases, the cells have arrived from the factory in a deactivated condition.   The
     causes of  the cell deactivation might be due to exposure to a COg-free environ-
     ment or factory defective cells,  the problem with cell life continues and is
     being studied.

  f)  The  Foxboro  2800 series and 1800 series magnetic flow meters have shown no
     serious problems.  Periodic scale cleaning is required to improve accuracy
     and  sensitivity but the meters are considered reliable, acceptably accurate
     and  easily serviced.

  g)  Both Dynatrol Model CL-10HY U-tube density meters and Ohmart radioactive density
     meters are used at Shawnee.  Both meters provide acceptably accurate and de-
     pendable  service.  From an operations point of view, the U-tube meters did have
     some initial problems with line plugging.  The cause was attributed to operator
     error in  setting too low a flow through in instrument.  The problem has since
     been resolved.


In the future,  operational areas of interest will include but not be  limited to
the continued  effort to understand and control scale  growth,  improve  dewatering
equipment operation, reduce the frequency of routine maintenance, and optimize
pump seal water requirements.
                                     71

-------
                     STATUS REPORT OF

       SHAWNEE COCURRENT AND DOWA SCRUBBER PROJECTS

                            AND

               WIDOWS CREEK FORCED OXIDATION
                        J.  L.  Crowe
                      G. A.  Hollinden
                      Energy Research
                Tennessee Valley Authority
                  Chattanooga,  Tennessee

                            and

                      Thomas Morasky
             Desulfurization Processes Program
           Fossil Fuel  Power Plants  Department
             Electric  Power  Research  Institute
                  Palo  Alto, California
              Prepared for Presentation at
              Industry Briefing Conference
Results of EPA Lime/Limestone Wet Scrubbing Test Programs
  Sponsored by the U.S. Environmental Protection Agency
      Royal Villa Motel in Raleigh, North Carolina
                     August 29, 1978
                         72

-------
Introduction




     In 1970, the Tennessee Valley Authority and the Environmental Protec-



tion Agency began a joint program at TVA's Shawnee Steam Plant to evaluate




processes which would remove sulfur dioxide and particulates from the




gaseous emissions of a coal-fired power plant.  The major areas of concern



investigated are:  the cost of removal, the reliability of the process,




the availability of materials needed and waste disposal of the byproducts



from the removal processes.




     Three 10-MW scrubbers were constructed for testing by TVA.  Each has



the capability of pulling flue gas from the No. 10 boiler either before



or after the electrostatic precipitator.  Simultaneous testing has con-




tinued on various lime and limestone removal processes in an attempt to



lower capital and operating costs to make each process more reliable, to




optimize the processes, and to stabilize or utilize the waste products



from the processes.




     Two advanced systems are presently being prepared for testing at a



10-MW size.  The first is a recurrent scrubber which has the potential




advantage of a smaller scrubber vessel, thus lower capital cost over a



conventional scrubber.  Testing of the cocurrent scrubber has been com-



pleted on an 1-MW pilot plant at Colbert Steam Plant and the 10-MW




prototype is in initial testing phase.  The other process is the DOWA




process.  This Japanese process is being marketed in the U.S. by Universal




Oil Products (UOP).  This process can produce a stable, storable product



(gypsum) and, where demanded, a sellable product.




     Forced Oxidation has been demonstrated at Shawnee Steam Plant on a




10-MW facility.   TVA plans to demonstrate forced oxidation on a full-




scale facility at Widows Creek Steam Plant.  Testing will include technical
                                73

-------
feasibility and environmental acceptability of forced oxidation as a




method of sludge disposal.






Cocurrent 1-MW Pilot Plant Results



     EPRI funded TVA to evaluate the cocurrent scrubber concept at the




Colbert 1-MW pilot plant and to provide design data for a 10-MW prototype




to be constructed and operated at the Shawnee Scrubber Test Facility.




     The emphasis of the evaluation was to study (1) the gas-liquid distribu-




tion characteristics of the absorber, (2) S(>2 and particulate removal




efficiencies as a function of gas velocity and liquid rates, and  (3) the




effect of spray nozzle type and location and scrubber internals,  such as




grids and packing, on S02 removal.



     The gas-liquid distribution study consisted of operating the absorber




with  (1) air only to determine the gas distribution and (2) air and water




to determine the liquid distribution.  With the scrubber containing no




internals and operating at gas velocities of 12.6 and 19 ft/sec,  the air-




flow was unsymmetrical at the scrubber inlet but became symmetrical at




the lower portions of the tower.  Five superimposed sections of bar grids




(straightening vanes) were installed at the scrubber inlet, which improved




the flow profile down the absorber.  Waterflow traverses during the air-




water tests revealed a poor liquid distribution at all levels.  The total




flow  rate and nozzle pressure had little effect on liquid distribution.




The data indicates that a large portion of the waterflow was on the walls




at the upper levels but disengaged from the wall as  it proceeded  down the




tower.



      Gas velocity profiles were not  as good and the  liquid  distriuution was




still poor, after the addition of six inches of Poly Grid packing to each




of the five bar grids.  However, the packing did tend to cause  less  flow




on the scrubber walls.




                                74

-------
     Tests were performed using sodium carbonate as the absorbent to




determine the maximum efficiency of the scrubber.  With no internals




but the straightening vanes in the absorber the SO. removal ranged from




90 to 97 percent with an average of 94 percent which indicated good mass




transfer.  The liquid rate had the greatest effect on S0_ removal; increasing




the liquid rate increased S02 removal.  The spray nozzle type (pressure




drop) and its location had a lesser effect; an increase in nozzle pressure




drop increased S02 removal.  A higher SO- removal was observed when the




total liquid flow was routed through the uppermost nozzles at the scrubber




inlet than when the liquid was distributed at various levels down the




absorber.




     With lime as the absorbent, five 6-inch sections of the Poly Grid




packing were added to the bar grids to increase SO. removal from the




62 percent obtained with no internals to a prerequisite 85 percent at




established conditions (gas velocity of 19 ft/sec, liquid rate of 212 gpm,




and liquid to gas ratio of 56).  During the test program, the SO- removal




ranged from 79 to 95 percent and averaged 88 percent.  The effect of




the liquid rate was more dominant than that of the gas velocity.  Increasing




the liquid rate was more dominant than that of the gas velocity.  Increasing




the liquid rate increased SCL removal, while decrease in SO. removal




occurred when the gas velocity was increased.  Higher SO- removals occurred




when the slurry was routed through the top than when it was distributed




throughout the tower.  The nozzle pressure drop had no significant effect




on SO^ removal.  Additional lime tests were made to determine the effect




of the number of scrubbing stages  (gas-liquid contact time) on SO. removal.




With the same internals, the scrubbing slurry was  successively introduced




above each grid down the absorber.  Increasing the gas-liquid contact  time




was found to improve SO. removal.
                                 75

-------
      Comparing  the  liquid  distribution data with the SO. removal effi-




 ciencies for  these  tests indicates that good SCL removals can be achieved




 with a less than ideal gas-liquid distribution in the absorber.  Other




 scrubber types  may  also exhibit a similarly poor gas-liquid distribution




 but there is  little published data for comparison.




      When limestone was used as the absorbent, an additional 3 inches of




 Poly Grid packing (five 9-inch sections) were added to obtain the pre-




 ferred S0_ removal efficiency of 85 percent at set conditions.  During the




 test program,  the S02 removal ranged from 77 to 92 percent with an average




 of 84 percent.  The gas velocity had the greatest effect on SO- removal.




 Increasing the gas velocity from base conditions lowered SO, removal




 efficiencies.   An increase in liquid rate increased SO. removal.   The




 nozzle pressure drop had no significant effect on S02 removal.




      The particulate removal averaged 99.4 percent for both the lime and



 limestone tests.




      TVA completed  the lime/limestone-scrubbing tests with  the 1-MW cocurrent




 pilot scrubber in mid-1977.  These  tests  successfully demonstrated that the




 cocurrent scrubber  is  an effective  S02 scrubber.   The results  of  these




 pilot-plant  tests were used to guide  the  design of  a  10-MW  improved proto-




 type  scrubber  which was  installed in  the  idle Hydro-Filter  scrubber train




 at  the  Shawnee facility.  Figure 1  is  the  process  flow  diagram for  the




 new 10-MW prototype  scrubber in the cocurrent mode.   The possible advan-




 tages of  a cocurrent scrubber system over  a conventional countercurrent




 system are:




     o The equipment configuration is more compatible with most power




plant duct and fan arrangements.  The gas would enter the scrubber




at a higher elevation and exit near ground level.  The entrainment
                                76

-------
                                     FLUE GAS
                                       INLET
                                                                                                              RIVER
                                                                                                              WATER
                                                                                   LIMESTONE SLURRY
                                                                                      FEED PUMP
ENTRAINMENT
SEPARATOR
CIRCULATION
PUMP

RIVER  WATER -»
ENTRAINMENT
SEPARATOR
CIRCULATION
TANK

s
T
o
O
:RUBBER ,
r
-£
C
TANK
TO POND DISPOSAL
                                                            SCRUBBER
                                                            CIRCULATION
                                                            PUMPS
                                                                               THICKENER
                                                                               UNDERFLOW
                                                                               PUMP
                         DISPOSAL
                          PUMP
                  FILTER CAKE
                   RESLURRY
                     TANK
                                                VACUUM  BELT FILTER
                                      FILTRATE
                                       PUMP
RECYCLE LIQUOR
 SURGE TANK
                                                                    RECYCLE LIQUOR
                                                                     RETURN PUMP
                                                   FIGURE  I
              Process  Flow Diagram -  Improved  Prototype  Scrubber  (CocurrenE Mode)

-------
 separator and reheat systems (likely to require  the  most  attention) would



 be located near ground level.   Likewise,  the induced-draft  (ID)  fans  would



 be on the ground and the connecting ductwork to  the  stack would  be shorter



 and probably less complex.




      o The physical arrangement of the proposed  cocurrent system causes



 the gas to change direction in the base of the unit  before  it enters  the




 entrainment separator.   The change in direction,  together with the vertical



 position of the entrainment separator,  promotes  good liquid separation



 and drainage.   Also,  a  separate entrainment wash loop can be used if




 needed.



      o Scrubbing liquid would  tend to coalesce into  larger  droplets before



 it disengaged  from the  gas  stream  near the base  of the scrubber.  This



 would further  facilitate efficient operation of  the  mist  eliminator.



      o Flooding of the  unit and associated high-pressure  loss and



 excessive entrainment of scrubbing slurry,  even  if grids  are added to



 improve  gas-liquid contact,  is  less likely.   Also, during normal  cocurrent



 operation,  the  gas-side pressure loss would be lower since  some  liquid-



 side  energy would be  recovered.



      o Higher gas  velocites  (smaller  scrubbers) are  expected because of



 the  reduced tendency  to flood and  because  more efficient  mist elimination



 is  likely.   (The  Colbert pilot  tests  were  successfully performed  at



 approximately 30  ft/sec superficial gas velocity.)   Therefore, smaller



 or fewer  scrubber modules would be  required  in a full-scale system.




      To evaluate  further  the merits of the  cocurrent  scrubbing concept and



 to obtain additional  data for scale-up to  a  full-scale facility,  EPRI



 contracted with TVA to  design, procure, erect and test the improved




prototype wet-scrubbing  system  (gas flow equivalent to 10-MW of generating



capacity) at the TVA Shawnee Steam Plant, Paducah, Kentucky.






                                78

-------
                            Test  Objectives




     The objectives  of the  scrubber  test  program are  as  follows:




     1.    To perform an operating variable study that will  identify the




          optimum operating conditions  of the  improved prototype  scrubber



          with respect to SO-  and particulate  removal efficiency.



     2.    To develop the scale-up similarities and differences  between the



          Colbert Test Program (1-MW) and the  Shawnee Test  Program (10-MW).



     3.    To determine design  parameters  for scale-up to a  full-scale  FGD



          system.




     4.    To perform a long-term  reliability demonstration  of a cocurrent




          prototype  limestone  scrubber  system  at optimum operating conditions.



     5.    To evaluate the collection efficiency, reliability, and materials



          of construction of the  entrainment separator.




     6.    To evaluate the performance of  an  inline-indirect steam reheater




          with respect to heat transfer characteristics, materials of



          construction, and reliability.




     7.    To determine the  physical  properties of the waste solids which




          are produced during  the reliability  demonstration.




     8.    To evaluate the performance of  other mechanical components within




          the scrubber system, such  as  pumps,  piping materials, valves,



          and instrumentation.






               General Test Program  Outline  and Schedule




     The type of tests and  order of  testing  will be generally as  outlined



below:




     1.    Preoperational Testing (Final Equipment Checkout).




     2.    Sodium carbonate-scrubbing factorial tests to establish the




          maximum removal efficiency of the  scrubber in various operating



          modes.
                                79

-------
     3.    Limestone-scrubbing operating variable study with a cocurrent




          scrubber.



          a.   Factorial testing.



          b.   Short-term reliability tests.




     4.   Lime-scrubbing operating variable study with a cocurrent scrubber.




          a.   Factorial testing.




          b.   Short-term reliability tests.



     5.   Reliability demonstration at optimum operating conditions—




          cocurrent limestone scrubbing.




     A proposed schedule for the test program is shown in Figure 2.




The schedule is included to indicate the order of testing and relative




time to be assigned to each test block.  It will be subject to change as




the test program proceeds.  The length of individual tests will vary from




an 8-hr shift (factorial tests) to a month (reliability demonstration).




However, most tests will probably be 1-2 weeks in length to allow the




composition of the process liquor and solids to approach equilibrium.



TVA technical personnel will determine when the test objective has been




achieved and when the tests are to be terminated.






DOWA Process




     In this process, SO- is absorbed in a clear solution of basic aluminum




sulfate.  The spent absorbent is oxidized with air, then neutralized with




limestone to remove the sulfur in the form of gypsum.  The regenerated basic




aluminum sulfate solution is recycled to the absorber.






PROCESS DESCRIPTION



     The flue gas from the boiler flows directly into the absorber




(Figure 3).   SO  is absorbed by a counterflowing stream of basic aluminum




sulfate (Al (S0,)3  • A12 03).  The absorption mechanisms are given in







                                80

-------
                Proposed  Test   Program    Schedule
                    Improved   Prototype   Scrubber
TA** Dl/i*«k
1 CSI BIOCK
Preoperationat Tests

Sodium Carbonate Factorial Tests
•
Cocurrent Mode Tests
Limestone - Factorial
Short-Term Reliability Tests
Lime- Factorial
Short -Term Reliability Tests

Long -Term Reliability Demonstration






Months
1
••

-














2





^••i











3


















4


















5







MM









6








••








7


















8


















9

















10

















II

















12

















13

















14

















Topical Reports
I-Report  on  sodium carbonate tests
2-Report  on  limestone tests
3-Report  on  lime  tests
                                   Figure  2
                                   81

-------
                                                                                                                 Limestone Silo
                            Reheater
                                            Desulfurized
                                            Flue Gas
     Mist /\
Eliminator
    Gas  from
    Boiler—
OO
ro
                     Absorber
              1
                                        *—Water
                                               AI2(S04)3
                                Aluminum Sulfate
                                  Make-up Tank
                                    •Air
              Absorber Liquor
              Mold Tank
                                                                                                                           Feeder
                                                                                                                   Limestone
                                                                                                                 Slurry  Storage
                                                            Neutralizing
                                                               Tanks     Thickene7
                                                         Vacuum
                                                    Vacuum  Receiver
                                                                                                                  Reclaiming
                                                                                                                  Absorbent
                                                                                                                  Tank
                                                                                                           To  Pond
                                                                                           Waste Slurry
                                                                                              Tank
                                     Figure  3.   Dowo   Typical   Process  Flow   Diagram

-------
Reaction 1 (below).  The cleaned gas passes through a mist eliminator



before entering the stack.




     The spent absorbent is delivered to an oxidizer into which fine



air bubbles are injected.  The sulfites in the solution are oxidized



to sulfates by Reaction 2.  The bulk of the resulting solution is



recycled to the absorber.




     The reminader of the oxidizer effluent is channeled to neutralizing



tanks.  Limestone (CaCO,) is added to recover the basic aluminum sulfate



and precipitate the sulfur in the form of gypsum (Reaction 3).  The



liquor from the neutralizing tank overflows into a thickener, where the



absorbent liquor is separated from the gypsum.  Further separation of



the gypsum slurry from the thickener takes place in a filter.  The



resulting gypsum is sent to disposal.  The liquor from the filter is



mixed with that from the thickener and recycled to the absorber.






REACTIONS



     Basic aluminum sulfate is used in the absorber to remove SO-.  The




oxidizer is used to convert the resulting sulfites to sulfates.  The




products are then regenerated in the neutralizer.



     Absorber:




          A12(S04)3 • A1203 + 3SO  	> Al (S04)g  • A12(S03)3       (1)



     Oxidizer:



          2A12(S04)« A12(SO«)- + 30 £ 	> 4A1-(SO,)^                (2)



     Neutralizer:




                       3CaC03 + 6H20 	>                           (3)




                       • A120- + 3CaS04  • 2H20 + 3CO
                               83

-------
Advantages



      (1)  The process  is  not  very  complex.




      (2)  Since  limestone is  used  as  a  neutralizing  agent, operating  costs



          are low.




      (3)  Due to  the low  liquid-to-gas  ratio,  equipment size requirements



          are low.



      (4)  Recovered gypsum my be of high quality and, thus, sellable,




          wherever a demand for it exists.   In this  project, however,




          the gypsum will be  a disposal byproduct.






DISADVANTAGES/PROBLEMS



      (1)  Gypsum  is not a very desirable commodity for recovery, since



          other sources can easily supply the  existing market.-






STATUS



     TVA has negotiated with  EPA for  the use of the  TCA for testing of




this process.



     Initial engineering  design and procurement are  underway.  TVA is




negotiating contracts with both UOP and EPRI for the project.  An optimis-




tic schedule (Figure 4) calls for  modification to the TCA to begin



during October with operation of the  scrubber to begin in early January.






Test Objectives



     The major objectives of this  test and demonstration program are




as follows:



     1.    To demonstrate  that the DOWA process can effectively treat




          flue gas from a boiler which is fired with high-sulfur coal



          and meet current emission standards.
                              84

-------
oo
            Months from
          date  of  UOP-
          TVA  contract
          Phase I
        Engineering  and
        Procurement
           Phase  II
        Engineering  and
        Procurement
              Phase III 8
              Phase IV
            Startup  and
            Operation
8
10
II
12
13
14
15
                                          Figure  4.   Dowa   Project   Schedule

-------
     2.   To evaluate the physical properties of the gypsum byproduct



          and its suitability as a landfill material.




     3.   To determine the S02 removal efficiency and particulate removal




          efficiency of the DOWA process over a broad range of operating



          conditions.




     4.   To determine or confirm parameters for scale-up to a full-scale



          FGD system.






General Test Program Description and Schedule




     The types of test blocks and operating variables which will be con-



sidered for investigation during the test program are outlined below:



     1.   Equipment shakedown with air and water.




     2.   System shakedown and process demonstration at operating condi-



          tions  which  are based on previous commercial  experience in



          oil-fired boiler applications.




     3.   Factorial tests  - S02  and particulate  removal efficiencies;



          determine the effect  of  the  following  variables upon SO.  and



          particulate  removal efficiencies:



          a.   Al  concentration



          b.   Basicity




          c.   TCA pressure drop




          d.   TCA liquid  recirculation rate



          e.   TCA superficial gas velocity



          f.   Absorber hold tank retention time




    4.   Factorial tests  - Oxidation; determine the effect of the following



         variables upon the oxidation efficiency:




         a.    Oxygen stoichiometry in the absorber loop
                               86

-------
          c.   Al concentration




          d.   Basicity




          e.   Absorber hold tank retention time




     5.   Factorial tests - Neutralization and byproduct production;




          determine the effect of the following variables upon neutraliza-




          tion reaction rate, settling rates of precipitates, filtera-




          bility and final settled bulk density of byproduct gypsum,  and



          aluminum losses.




     6.   Short-term reliability tests.




     Following the system shakedown and initial process demonstration,




process variable studies will begin with factorial tests which are designed




to screen the effect of each of the above listed variables.  If technically




feasible and to conserve time, several of the factorial tests may be




performed simultaneously.  Following the factorial tests, a series of




short-term reliability tests will be performed.  The selection of operating




conditions for these tests will be based upon the results of the factorial




tests.  The purpose of the short-term reliability tests is to obtain more




definitive information on the effect of operating conditions upon the




reliability of system components and to obtain operating data which will




be the basis for selection of operating conditions for a long-term




reliability demonstration.  The conditions of the scrubber equipment will




be evaluated at the end of each test.  Although the operating conditions




for all tests will be specified before each test block begins, the test




conditions will be subject to change as the results of the initial tests




are evaluated.




     If results from this test program are favorable, additional funding




will be needed for more extensive variable studies and a long-term relia-




bility demonstration of the DOWA FGD system.  As indicated above, the





                               87

-------
 selection of the operating conditions for the long-term reliability test




 should be based on the short-term reliability test results.   System relia-




 bility and economy of operation will be the major criteria which  should  be




 used in the selection of operating conditions.   Also,  only operating




 conditions which have met the  current emission standards  should be  selected




 During the reliability test, all the scrubber operating conditions  should




 be  held constant except for the flue gas  rate which will  be varied  in




 proportion to  the boiler load.   During this test the physical and chemical




 properties of  the gypsum byproduct  should be  routinely determined,  including




 settling  rate,  final  settled bulk density,  compressive strength, filtera-




 bility, particle  size  distribution,  chemical  analysis,  and general  crystal




 form.   To  the extent possible, pertinent physical properties should be




 correlated with the operating conditions of the  process unit.




     The preoperational tests (equipment shakedown) and the test program




will be conducted during a 4-month period.  A test program schedule which




indicates the order of tests and the relative amount of time assigned to




each test will be proposed to EPRI and UOP  one month prior to startup.   The




test program schedule will be subject to revision as the test results are




evaluated.  All schedule changes will be approved by EPRI and UOP.
                               88

-------
                 DEMONSTRATION OF FORCED OXIDATION AT
                          WIDOWS CREEK UNIT 8
     The purpose of oxidizing lime/limestone FGD product sludges is to
convert calcium sulfite (CaSO_), the normal product of these scrubber
processes to calcium sulfate (CaSO,).  Calcium sulfate, which is commonly
known as gypsum, is a more desirable waste product because it improves the
settling and dewatering properties of the sludge.  This in turn reduces
the volume of material for disposal and makes a material which may be
suitable for landfill without need of additives or the use of mixing or
blending equipment.
     This project is designed to develop, demonstrate, and evaluate, at the
full-scale level, the technical feasibility and environmental acceptability
of utilizing oxidation as a method of sludge disposal.  This will be
accomplished by testing a forced oxidation system on the "D" train
(140-MW equivalent) of the Widows Creek unit 8 wet limestone scrubber.
A slipstream of 25 percent of the full load flow of the effluent slurry
will be treated in a thickener and vacuum filter.
     Tests will include two-stage forced oxidation, single-stage forced
oxidation, and oxidation in both stages  (venturi and absorber).  The effect
of such variables as air stoichiometry, pH, and  limestone stoichiometry
will also be evaluated.
     Combustion Engineering, Inc., under a contract with the Tennessee
Valley Authority will be responsible for the design, procurement,  erection,
testing, and reporting of this oxidation demonstration program.

A.   Widows Creek Simulation at Shawnee
     Essentially complete oxidation  of calcium  sulfites from the  scrubbers
operated at the Shawnee Test Facility has been  routinely achieved during

                                89

-------
 forced  oxidation testing on both the  TCA  and  the venturi/spray tower systems.



 However,  in pilot-plant  scale  tests conducted at the TVA Colbert plant,




 the  sludge  could not be  oxidized in any practical manner.  These opposed




 results indicate some undetermined site-specific conditions that either




 promotes  oxidation  at Shawnee  or conversely deters oxidation at Colbert.



 Since the natural oxidation at Widows Creek and Colbert is essentially



 the  same  and is  significantly  lower than  that obtained at Shawnee, con-



 siderable concern has arisen over the Colbert-Shawnee correlation since



 extrapolation may be possible  to the  forced oxidation mode.  Should this




 be the  case,  little to no oxidation could be  achieved on the Widows Creek



 system  if a  forced  oxidation should be desired for use as a method of




 treating  the  sludge.   In order to predict achievable oxidation at Widows




 Creek,  a  special  coal burn  at  Shawnee using Widows Creek coal, scrubbing




 with Widows  Creek limestone, and using the venturi-spray tower oxidation



 mode as operated  in the  EPA  test program  is planned.  The main problem



 would be  the  difference  in boilers for the two systems and the possible




 effect  this  could have on oxidation;  Shawnee—B&W, Widows Creek unit 8—



 CE.  These tests would simulate  the Widows Creek scrubber design as much



 as possible  so that problem  areas could be identified.






 B.   Widows Creek Forced Oxidation




     TVA will also  evaluate  oxidation on one of the scrubber trains at the




Widows Creek unit 8 facility such that data can be collected on the




feasability of oxidizing the sludge under Widows Creek operating condi-




tion.  Figure 5 is a  flow diagram of  the Widows Creek Forced Oxidation



System.   Should oxidation prove to be technically feasible an economic



evaluation will be made for comparison to other methods of disposal,




i.e., sludge-fly ash blending.   This project is designed to develop,
                               90

-------
                                                                Reheat
      Absorber Circ. Tank
Compressor
                                  Entrapment
                                  Separator
t
7
I
I
1





>« >* Tr
/

> Stack
                                                                             From Powerhouse
   —Venturi  Circ. Tank
                       Bleed  Stream


                    Thickener
Thickener Overflow
*•  To Effluent
   Slurry Surge
   Tank
                                                      Vacuum
                                                      Filter
                                                                                        To Truck
        Figure  5.  Flow  Diagram - Widows  Creek  Forced Oxidation

-------
demonstrate, and evaluate the technical feasibility and environmental



acceptability of oxidation as a method for disposal of sludge from the




No. 7 and No. 8 wet/limestone scrubber units at Widows Creek.  The primary




purpose for oxidizing the scrubber solids is to improve waste solids



dewatering and landfill disposal characteristics.  This objective will be




accomplished through specific tests which will be made to meet the following




criteria.



     1.   The project should result in an evaluation that will provide a




          reference point for making confident decisions on the feasibility



          of converting the scrubbers at the Widows Creek station to the




          forced oxidation mode.



     2.   Provide the results of this evaluation by March 1979 so that a



          sludge disposal system can be installed and be ready for use




          before the Widows Creek pond is filled.



     Figure 6 is a schedule for the Widows Creek Forced Oxidation System.



This study will also include a comparative economic study of forced




oxidation vs. other sludge disposal processes as they relate to the Widows




Creek system.



     Other major factors that will be included for evaluation are:



     1.   Transportability of the oxidized solids after dewatering.




     2.   Primary application for the dewatered solids.




     3.   Material handling equipment required.



     4.   Methods of disposal or reuse of waste water resulting from




          dewatering.



     5.   Effect of dewatering for landfill treatment.
                                92

-------
Engineering  and  Procurement
                Construction
                    Testing
                              AUG
                              1978
SEPT
1978
 OCT
1978
NOV
1978
 DEC
1978
JAN
1979
FEB
1979
MAR
1979
             Figure  6.   Widows  Creek  Forced Oxidation Schedule

-------
CURRENT STATUS OF DEVELOPMENT OF THE SHAWNEE LIME-LIMESTONE

                     COMPUTER PROGRAM
           C. D. Stephenson and R. L. Torstrick
           Emission Control Development Projects
      Office of Agricultural and Chemical Development
                Tennessee Valley Authority
                  Muscle Shoals, Alabama
               Prepared for Presentation at
               Industry Briefing Conference
 Results  of  EPA Lime/Limestone Wet Scrubbing  Test  Programs
   Sponsored by the  U.S.  Environmental  Protection  Agency
       Royal Villa Motel in Raleigh,  North  Carolina
                      August 29,  1978
                           94

-------
      CURRENT STATUS OF DEVELOPMENT OF THE SHAWNEE LIME-LIMESTONE

                             COMPUTER PROGRAM


GENERAL SCOPE AND PURPOSE

In conjunction with the U.S. Environmental Protection Agency (EPA)-
sponsored Shawnee test program, Bechtel National, Inc., and the Tennessee
Valley Authority (TVA) have jointly developed a computer program capable
of projecting comparative investment and revenue requirements for lime
and limestone scrubbing systems.  The computer program has been developed
to permit the estimation of relative economics of these systems for varia-
tions in process design alternatives (i.e., limestone vs lime scrubbing,
alternative scrubber types, or alternative sludge disposal methods) or
variations in the values of independent design parameters (i.e., scrubber
gas velocity and L/G ratio, alkali stoichiometry, slurry residence time,
reheat temperature, and specific sludge disposal design).  Although the
program is not intended to compute the economics of an individual system
to a high degree of accuracy, it is based on sufficient detail to allow
the quick projection of preliminary conceptual design and costs for
various lime-limestone case variations on a common design and cost basis.


PROGRAM DEVELOPMENT

The responsibility in the development of the computer program was shared
by Bechtel and TVA.  Bechtel*s major responsibility was to analyze the
results of the Shawnee scrubbing tests and develop models for calculating
the overall material balance flow rates and "tream compositions.  Bechtel
provided TVA with a complete computer program for specifying this informa-
tion.  TVA was responsible for determining fhe size limitations of the
required equipment for establishing the minimum number of parallel equip-
ment trains, accumulating cost data for the major equipment items, and
developing models for projecting equipment and field material costs as a
function of equipment capacity.  Utilizing these relationships TVA
developed models to project the overall investment cost breakdown and a
procedure for using the output of the material balance and investment
models as input to a previously developed TVA program for projecting
annual and lifetime revenue requirements.
                                   95

-------
PROGRAM DESCRIPTION AND DOCUMENTATION

TVA has presented two papers which reported the status of the program
development and displayed  its capabilities.  The first, titled "Shawnee
Limestone-Lime Scrubbing Process Computerized Design-Cost Estimate
Program:   Summary Description Report," was given at the Industry Briefing
Conference sponsored by EPA at Raleigh, North Carolina, October 14, 1976.
The second, titled "Economic Evaluation Techniques, Results, and Computer
Modeling  for Flue Gas Desulfurization," was presented at the EPA Flue
Gas Desulfurization Symposium in Hollywood, Florida, November 8-11, 1977.

A  significant number of revisions have been incorporated into the program
since  these earlier reports.  The present paper describes the current
capabilities of the computer program.  Since the design basis for the
lime and  limestone scrubbing systems has not changed appreciably from
the earlier publications,  it will not be included  in this paper.   TVA
is currently in the process of preparing a users manual for the overall
program which will include all the information required for running the
Turbulent Contact Absorber (TCA) program option.   It is anticipated that
the users manual will be available within the next few months.
CURRENT PROGRAM SCOPE

Uses and Limitations

The present computer program has the capability of projecting a complete
conceptual design package for lime or limestone scrubbing utilizing a
TCA with any one of four sludge disposal options (discussed later).  The
program is designed to consider new coal-fired power units ranging in
size from 100-1300 MW.  Equipment size and layout configurations are
modeled based on coals ranging in sulfur contents from 2-5%.  To
limit the extremely wide variations in equipment sizes and layout con-
figurations which can result with changes in other key independent
variables, the following range of values for these variables was
established.

          Scrubber gas velocity                   8-12.5  ft/sec
          Liquor  recirculation rate               25-75 gal/kft9
          Slurry  residence  time  in hold  tank     2-25 min

However,  operating parameters and plant sizes outside of these  ranges
will not  necessarily be invalid.

It  is  expected that the results may within limits  also  be valid for
extrapolation of sulfur content of  coal beyond the range actually tested
at  Shawnee.  The Shawnee models are based on scrubbing  results  over an
S02 concentration  range  of approximately  1500  to 4000 ppm.

The effect  of variations in  any of  the  inputs, such as  scrubber gas
velocity, degree of S02  removal,  reheat temperature, alkali stoichi-
ometry, L/G  ratio,  etc. ,  on process  design and economics may be  determined.

                                  96

-------
For both lime and limestone scrubbing,  SQ2 removal,  stoichiometry (pH for
the lime option), and L/G ratio may all be specified and results projected.
Alternatively, S02 removal and stoichiometry (or pH) may be specified and
L/G calculated, or S02 removal and L/G ratio may be specified and stoichi-
ometry (or pH) calculated.  An additional option is being incorporated
into the program allowing for the calculation of S02 removal based on
input values of L/G and stoichiometry (or pH).

The outputs of the overall computer program include (1) a detailed
material balance including properties of the major streams, (2) a detailed
water balance Itemizing water availability and water required, (3) speci-
fications of the scrubber system design, (4) a revised method for dis-
playing overall pond design and costs, (5) specifications and costs of the
process equipment by major processing area,  (6) a detailed breakdown of
the projected capital investment requirements,   (7) an itemized breakdown
of the projected revenue requirements by component for the first year of
operation of the system, and (8) a lifetime revenue requirement analyses
showing projected costs for each year of operation of the plant, as well
as lifetime cumulative and discounted costs and equivalent unit revenue
requirements.

New Program Options and Modifications Since the November 1977 FGD Symposium

To illustrate the current program inputs and outputs, an example run of the
updated computer program is shown in the appendix for limestone scrubbing
with an onsite pond disposal option.  Discussions of the modifications incor-
porated into the program since the November 1977 FGD symposium are given below,

Particulate Removal—

The quantity of ash in the coal which is emitted overhead as fly ash is an
input.  Additional inputs are required to specify ash removal upstream of
the scrubber and within the scrubber.  Th^se are input either as a removal
efficiency in percent or as an equivalent  'iitlet emission in pounds per
million Btu.  If removal is not input, a 33% efficient mechanical collector
is provided for protecting the fan from abrasion by large fly ash particles.
A cost model is available for optionally including the costs for the mechani-
cal collector.  As discussed later, cost models for a high-efficiency
electrostatic precipitator (ESP) or baghouse are presently being incorpo-
rated into the program, but are not yet available for the current version.
An example output of the fly ash removal option is shown on pages 10  and  13 of
the appendix under the heading "Fly Ash Removal."

SO2 Removal—

The percentage of the sulfur in coal which  is  emitted overhead  as S02  is  an
input to the program.  The degree of S02 removal may be  input by  specifying
either

   1.  % S02 removed,

   2.  Ibs S02 emitted per million Btu of heat input to  the  boiler   or
   3.  ppm S02 in outlet  flue gas.

                                  97

-------
 For each computer run, equivalent S02 emissions are displayed on all
 three basis, regardless of the method for inputting the degree of S02
 removal.  The alternative methods for specifying S02 removal are
 illustrated on page 14 under the heading "Flue Gas to Stack."

 Redundancy—

 In addition to designing the flue gas desulfurization (FGD) system with
 spare operating pumps and optionally with spare scrubbing trains as
 described in the earlier publications, spare feed preparation units may
 now be specified.  This is applicable for both limestone scrubbing in
 which spare ball mill trains can be provided, and lime scrubbing in
 which spare slakers can be specified.  The number of redundant alkali
 preparation units and redundant scrubber trains are specified under
 "Raw Material Handling Area" (pp.n  and 20 )  and "Scrubber  System Variables"
 (pp. 11, 16, and 22)  respectively.

 Water Balance—

 The program has previously assumed no net accumulation or loss of water
 due to rainfall, evaporation, and seepage.   The current version of the
 program allows for specific rainfall, evaporation,  and seepage rates to
 be input for accurately projecting makeup water requirements.   The water
 balance inputs and outputs for an example run are shown on  page 15.

 Waste Disposal Options—

 The program allows for four alternate waste disposal alternatives to be
 assessed including:

    1.   Onsite ponding

        a.   Unllned pond

        b.   Clay-lined  pond (cost  of  clay and  depth  of  lining input)

        c.   Synthetic-lined pond  (cost of liner  input)

    2.   Thickener - ponding

    3.   Thickener - fixation fee

    4.   Thickener - filter  - fixation  fee

The  onsite ponding options  may also be run with  fixation  fees  applied  to
them.  For alternatives  3  and 4,  the  fixation fee must  include  costs for
transportation and disposal  of the fixed sludge  offsite.  For alternatives
1 and 2, however,  only  the  costs  for  fixation need be provided  since the
fixed sludge can  be disposed of at the existing  pond site.

For the ponding alternatives the  program allows  for the onsite  pond to
be sized larger or smaller  than the normal projected lifetime capacity.

-------
This option, hag been incorporated (1)  to account for variations in the
sulfur content of fuel, (2) to evaluate design philosophy in construc-
tion ponds for less than the total amount of sludge to be disposed of
(requires assessment of additional cost for expanding pond later), or
(3) to allow the feed preparation and scrubbing areas to be sized based
on maximum sulfur contents expected, while sizing the pond based on
average sulfur contents.  An example output of the revised onsite ponding
mpdel is shown on page 19.

Equipment Cost Breakdown—

The program has been modified to provide a breakdown showing projected
equipment specifications and cost for each equipment item.  Both material
and labor components of equipment costs are displayed for each of the
three major areas.  The equipment list and costs for limestone scrubbing
and onsite ponding are illustrated on pages 20-23.  Although a complete
printout for the lime and alternate sludge disposal options is not
included in this paper, an equipment list for those options is illus-
trated on pages 27-30.

Operating Profile—

The current version of the program allows for the specification of three
alternative operating profiles as indicated below for projecting lifetime
revenue requirements:

   1.  Profile similar to that utilized in the  report Detailed Cost
       Estimates for Advanced Effluent Desulfurization Processes"
       by G. G. McGlamery, et al.   (EPA-600/2-75-006, January 1975)

   2.  Historical power plant operating profile based on  FPC Form 67 data

   3.  Variable profile with annual load factors as input

See pages 31-32 for illustration.
FUTURE PROGRAM DEVELOPMENT

Further additions to the program are expected to be made as additional
test data  from Shawnee become available.  Bechtel and TVA are currently
incorporating the results of the venturi-spray tower tests at Shawnee
into a design and cost model for that option.  In addition, cost models
for upstream fly ash removal by hot or cold side ESP, baghouse  collectors
and venturi scrubbers a,re being incorporated into the program.  Other
options which are being considered for incorporation as sufficient data
become available include  (1) series scrubbers/high alkali utilization
systems and (2) forced oxidation systems.

-------
PROJECTED PROGRAM USE

Upon completion of the overall effort, the program will be useful for
projecting a complete conceptual design package for lime-limestone scrub-
bing including material balance, capital investment estimate, and
projected revenue requirements.  It is expected that the program will
be used by utility companies and architectural and engineering con-
tractors involved in the selection and design of S02 removal facilities
for specific applications.  It is not intended to be used for projecting
a final design of a given system, but to assist in the evaluation of
system alternatives prior to development of a detailed design.  Also,
the program will be useful for evaluating the potential impact of various
process variables on economics as a guide for planning research and
development activities.

Although the program was not meant to be used for comparing projected
lime-limestone economics with economics for alternate processes, these
comparisons should be valid as long as the basis for the alternate process
economics are comparable to those included in the computer program for
lime-limestone systems.

Method for Attaining Results

TVA is in the process of loading the current version of the program on
a Control Data Corporation (CDC) timesharing computer system and publish-
ing a users manual for utilizing the program.  After this effort is
complete, outside users will be allowed to access the computer program
for making computer runs.   Until these activities are completed, TVA,
under a Technology Transfer Contract with EPA, can upon request make
computer runs for interested users, or can release copies of the program
to interested users along with available documentation for running it.

Current Users and Program Applications

A significant number of responses have already been handled under the
above arrangement.

Given below is a list of people who have been provided tape copies of
the Shawnee lime-limestone computer program.

          Robert H.  Boeckmann, Gibbs and Hill, Inc.
          Paul S. Farber, Argonne National Laboratory
          D.  J.  Hagerty, University of Louisville
          J.  Scott Hartman, PEDCo
          R.  G.  Knight, Michael Baker, Jr., Inc.
          M.  Lieberstein, The City of New York,
           Environmental Protection Administration
          F.  Y.  Murad,  Combustion Equipment Associates, Inc.
          Edward S.  Rubin, Carnegie-Mellon University
          J.  G.  Stevens, Exxon
          John Valente, Air Correction Division, UOP
          John Wysocki, Burns and Roe, Inc.


                                100

-------
A list of people who have been provided results of specific computer
runs is shown below:

          Randy M. Cole, Tennessee Valley Authority
          Wilson Cramer, U.S. Steel Corporation
          Richard Furman, Florida Power and Light
          Robert Gosik, Environmental Protection Agency (Denver)
          Robert Lane, Illinois Commerce Commission
          S. J. Lutz, TRW, Inc.
          M. F. Patterson, Linde Division, Union Carbide Corporation
          A. V. Slack, SAS Corporation
          John Wile, National Economic Research Associates, Inc.

To date several uses of the program other than those for which it was
intended have been tried.  The program has been run simulating both
industrial and utility boilers, smelter situations, partial scrubbing
situations, plant optimization studies, and for comparisons of front
end coal cleaning economics with total scrubbing.   Probably the most
important use to date has been support work for the National Economic
Research Association's assessment of the impact of possible NSPS
revisions on the electric utilities industry on a national scale.
[EPA issued Federal Standards of Performance for New Stationary Sources
(often called "new source performance standards" or NSPS).]  Examples
of some of the sensitivities which might be assessed are shown on
pages 33-46.
                                  101

-------
                        APPENDIX

                  INPUT DATA FOR EXAMPLE

                       PROGRAM RUN

11111
111111111011000
1 1 1
INDUSTRY BRIEFING-500 MW
0 500 9000 10500 33 300 2 175 470 751
57.56 4.14 7.00 1.29 3.12 0.15 16.00 10.74 95 80 1 97 5 an
50 20 12.5 25 2 1.2 12 1 0.0 1 0 0 2.85 500
15 40 .2 40 30 60 1.2 7.0 1 100 0
2 3 4 5 51 .000001 42 10 1.35 142.1
1 12 9999 3500 25 25 5280 1 10 2.5
9 16 5 10 8 12 11.6 8 3 50 10 17.2 1.17
8 50 12.5 2.0 0.12 0.029 17 220.9 178.2 1977 1978

END
                           102

-------
                   OUTPUT OF EXAMPLE RUN
                    *** INPUTS ***


BOILER CHARACTERISTICS
MEGAWATTS =   500.

BOILER HEAT RATE =  9000, BTU/KUH

EXCESS AIR =  33. PERCENT,  INCLUDING  LEAKAGE

HOT GAS TEMPERATURE - 300.  DEC  F

COAL ANALYSIS, UT 7. AS FIRED  :

 C      H      0      N       S      CL     ASH     H20
57.56   4.14   7.00   1.29    3.12   0.15   16.00  10.74

SULFUR OVERHEAD =  95.0 PERCENT

ASH OVERHEAD =  80.0 PERCENT

HEATING VALUE OF COAL = 10500.  BTtl/LB
                         EFFICIENCY,     EMISSION,
FLYASH REMOVAL                %          |_BS/M  RTU
UPSTREAM OF SCRUBBER         97.5           0.30

UITHJN SCRUBBER              80.0           0.06

COST OF UPSTREAM FLYASH  REMOVAL EXCLUDED
ALKALI
LIMESTONE  ?
       CAC03        =   97.15  I.JT  7.  DRY  BASIS
       SOLUBLE  MGO  =    0.0
       INERTS       =    2,85
       MOISTURE  CONTENT =    5.00  LB H20/100 LBS DRY LIMESTONE
       LIMESTONE HARDNESS  WORK  INDEX  FACTOR = 10 00
       LIMESTONFT DEGREE OF GRIND  FACTOR =  1,35
FLY ASH  :
       SOLUBLE  CAO  =   0.0  U]  %
       SOLUBLE  MGO  •-   0.0
       INERTS       = 100.00

                        103

-------
RAW MATERIAL HANDLING AREA






NUMBER OF REDUNDANT ALKALI PREPARATION UNITS =






SCRUBBER SYSTEM VARIABLES
NUMBER OF OPERATING SCRUBBING TRAINS =   4




NUMBER OF REDUNDANT SCRUBBING TRAINS =   2




NUMBER OF BEDS =3




NUMBER OF GRIDS =   4




HEIGHT OF SPHERES PER BED =  5,0 INCHES




LIQUID-TO-GAS RATIO =  55. GAL/1000 ACF




SCRUBBER GAS VELOCITY = 12.5 FT/SEC




S02 EMISSION LIMIT = 1.20 LB S02/M BTU




STOICHIOMETRY RATIO     TO BE CALCULATED




ENTRAINMENT LEVEL = 0.10 WT X




EHT RESIDENCE TIME -  12.0 MIN




S02 OXIDIZED IN SYSTEM =  30.0 PERCENT




SOLIDS IN RECIRCULATED SLURRY =  15.0 WT %






SOLIDS DISPOSAL SYSTEM
COST OF LAND =  3500.00 DOLLARS/ACRE




SOLIDS IN SYSTEM SLUDGE DISCHARGE =  40,0 WT %




MAXIMUM POND AREA =  9999. ACRES




MAXIMUM EXCAVATION =  25.00 FT




DISTANCE TO PONP =  5280, FT




POND LINED WITH 10.0 INCHES CLAY
                       104

-------
STEAM REHEATER (IN-LINE)


SATURATED STEAM TEMPERATURE =  470. KEG F

HEAT OF VAPORIZATION OF STEAM =  751. BTU/LB

OUTLET FLUE GAS TEMPERATURE = 175. HEG F

SUPERFICIAL GAS VELOCITY (FACE VELOCITY) =  25.0 FT/SEC
  IT     SR       SROLD
   1     1,24     1.50
   2     1.24     1.24
   3     1.24     1.24
PAPER PRINT 01 r  POND  FEE
                          105

-------
                    *#X< OUTPUTS *.**
HOT GAS TO SCRUBBER

C02
HCt
S02
02
N2
H20
S02
MOLE PERCENT LB-HOLE/HR
12 . 3 15 0.2054E+05
0*011 0.1813E+02
0.238 0.3962l£i03
4,827 0.8050E+04
73.867 0.1232E+06
8.743 0.1458E+05
CONCENTRATION IN SCRUBBER INLET
FLYASH EMISSION = 0.30 LBS/MILLION
= 0.152 GRAINS/SCF

SOLUBLE CAO IN FLY ASH =
SOLUBLE MGQ IN FLY ASH =
LB/HR
0.9040E+06
0.6612E+03
0.2538E+05
0.2576E+06
0.3452E+07
0.2627E+06
GAS = 2376. PPM
BTU
(WET) OR 1371. LB/HR
0. LB/HR
0.
HOT GAS  FLOW  RATE  =   .1054E+07 SCFM (60 DEC FT  1 ATM)
                   =   ,1540E+07 ACFM (300.  DEG Fr 1 ATM)

CORRESPONDING COAL FIRING RATE =  .4286E+06 LB/HR

HOT PAS  HUMIDITY =  0.057 LB H20/LB DRY GAS

WET BULB TEMPERATURE = 127.  BEG F


UET GAS  FROM  SCRUBBER
MOLE PERCENT LB-MOLE/HR
C02
S02
02
N2
H20
11.683
0.047
4.480
68.969
14.820
S02 CONCENTRATION
FLYASH

EMISSION =
«
0.2087E+05
0.8430E+02
0.8003E+04
0.1232E+06
0.2647E+05
IN SCRUBBER OUTLET
0.06 LBS/MILLION
0.02R GRAINS/SCF
LB/HR
0.9185E+06
0.5400E+04
0.2561E+06
0.3452E+07
0.4769E+06
GAS = 472. PPM
BTU
(WET) OR 274. LB/HR
 TOTAL WATER PICKUP  =   439.   GPM
            INCLUDING    10.2  GPM  ENTRAINMENT

 WET GAS FLOW RATE =   .1128E-1-07 SCFM  (60  DEG  Fj  1  ATM)
                   -   .1274E+07 ACFM  (127.  DEG F»  1  ATM)

 WET GAS SATURATION  HUMIDITY  =  0.103  LB  H20/LB  DRY  GAS

                        106

-------
FLUE GAS TO STACK
        MOLE PERCENT

C02        11,665
S02         0.047
02          4.473
N2         68,860
H20        14,955
          LB-MOLE/HR

          0,20S7E-f-05
          0.8430E402
          0.8003E+04
          0.1232E+06
          0.2676E+05
        LB/HR

        0.9185E+06
        0.5400E+04
        0.2561E-f06
        0.3452E+07
        0.4820E+06
CALCULATED S02 REMOVAL EFFICIENCY =  78.8 %

SPECIFIED  S02 EMISSION =   1.20 POUNDS PER MILLION BTU

CALCULATED R02 CONCENTRATION IN STACK GAS =     471,
FLYASH EMISSION =  0.06 LBS/MILLION BTU
                -  O.O28 GRAINS/SCF 
-------
ATER BALANCE INPUTS
 RAINFALL(IN/YEAR)
 POND SEEPAGE(CM/SEC)*10**3
 POND EVAPORATION(IN/YEAR)
  51.
 100.
  42,
IATER  BALANCE  OUTPUTS
JATER  AVAILABLE

  RAINFALL
  ALKAL.I
      TOTAL
 658,  GPM
   4.  GPM
 662,  GPM
328778. LB/HR
  1995. LB/HR
330773. LB/HR
•JATER REQUIRED

  HUMIDIFICATION
  ENTRAINMENT
  DISPOSAL WATER
  HYDRATION UATER
  CLARIFIER EVAPORATION
  POND EVAPORATION
  SEEPAGE

  TOTAL WATER REQUIRED
 429,  GPM
  10,  0PM
 157,  GPM
  11.  GPM
   0,  GPM
 575,  GPM
 160.  GPM

1342,  GPM
214217.
  5103.
 78373.
  5307.
     0.
287557.
 80040.
        LB/HR
        LB/HR
        LB/HR
        LB/HR
        LB/HR
        LB/HR
        LB/HR
670596. LB/HR
!ET  UATER  REQUIRED
 680,  GPM
339823. LB/HR
                                   108

-------
SCRUBBER SYSTEM
TOTAL NUMBER OF SCRUBBING TRAINS (OPERATING+REDUNDANT) =  6

S02 REMOVAL =  78.7 PERCENT

PARTICULATE REMOVAL IN SCRUBBER SYSTEM =  80.0 PERCENT

TCA PRESSURE DROP ACROSS  3 BEDS =  8.6 IN. H20

TOTAL SYSTEM PRESSURE DROP = 14.8 IN..H20

SPECIFIED   LIQUID-TO-GAS-RATIO    =  55. GAL/1000 ACF

LIMESTONE ADDITION = 0.3990E+05 LB/HR DRY LIMESTONE

CALCULATED LIMESTONE STOICHIOMETRY  =  1.24 MOLE CAC03 ADDED AS LIMESTONE
                                           PER MOLE S02 ABSORBED

SOLUBLE CAO FROM FLY ASH =  0.0  MOLE PER MOLE S02 ABSORBED

TOTAL SOLUBLE MGO        =  0.0  MOLE PER MOLE S02 ABSORBED

TOTAL STOICHIOMETRY      =  1.24 MOLE SOLUBLE (CA+MG)
                                 PER MOLE S02 ABSORBED

SCRUBBER  INLET LIQUOR PH =  5.34

MAKE UP WATER =  680. GPM
                                109

-------
SYSTEM SLUDGE DISCHARGE
SPECIES
           H20
CAS04  .2H20
CAC03
INSOLUBLES
H20
CA++
MG-H-
S03—
S04—
CL-
LB-MOLE/HR   LB/HR
0.2182E+03
0.9272E+02
0.6629E+02

0.4350E+04
0.9937E+01
0.0
0.1254E4-00
0.8489E+00
O.1791E-H02
0.2817E+05
0. 1596E4-05
0.6635E+04
0.2234E+04
0.7837E+05
0.3983E+03
0.0
0.1004E4-02
0.8154E4-02
0.6351E+03
SOLID
COMF»
UT X

53.16
30.11
12,52
 4.22
                     LIQUID
                     COUP,
                     PPM
         5010.
            0.
          126.
         1026.
         7988.
TOTAL DISCHARGE FLOW RATE  = 0.1325E+06  LB/HR
                              200.      GPM

TOTAL DISSOLVED SOLIDS  IN  DISCHARGE  LIQUID =  14150.  PPM

DISCHARGE LIQUID PH =   7.24


SCRUBBER SLURRY BLEED
TOTAL FLOW RATE = 0.3533E+06 LB/HR
                     642.    GPM
TOTAL SUPERNATE RETURN
TOTAL FLOW RATE = 0.1820E+Q6 LB/HR
                     364.    GPM
                        110

-------
SUPERNATE TO WET BALL MILL
TOTAL FLOW RATE = 0.2461E+05 LB/HR
                      49.     GPM
LIMESTONE SLURRY FEED
TOTAL FLOW RATE = 0.6A51E+05 LB/HR
                =84.    GPM


SUPERNATE RETURN TO SCRUBBER OR EHT
TOTAL FLOW RATE = 0.1574E+OA LB/HR
                =    315.    GPM


RECYCLE SLURRY TO SCRUBBER
TOTAL FLOW RATE = 0.3859E+08 LB/HR
                =  70087.    GPM


FLUE GAS COOLING SLURRY
TOTAL FLOW RATE = 0.2807E+07 LB/HR
                    5097.    GPM
                       111

-------
                   POMD DFSIGM
 OPTIMIZED TO MINIMIZE TOTAL COST PLUS OVERHEAD
POND DIMENSIONS
DEPTH. OF POND
DEPTH OF EXCAVATION
LENGTH OF PERIMETER
LENGTH OF DIVIDER

AREA OF BOTTOM
AREA OF INSIDE WALLS
AREA OF OUTSIDE WALLS
AREA OF POND
AREA OF POND SITE
AREA OF POND SITE

VOLUME OF EXCAVATION
VOLUME OF SLUDGE TO BE
DISPOSED OVER LIFE OF PLANT
   20.fc2 FT
    3.24 FT
13193.   FT
 2403.   FT

 1074.   THOUSAND YD2
  129.   THOUSAND YD2
   9
-------
                       RAW  MATERIAL.  HANDLING  AND  PREPARATION
          INCLUDING  2 OPERATING  AND   1  SPAKE  PREPARATION UNITS
         ITEM
                                    DESCRIPTION
                                                        NO.  MATERIAL
                                                                        LABOR
 CAR  SHAKER AND  HOIST

 CAR  PULLER

 UNLOADING  HOPPER


 UNLOADING  VIBRATING FEEDER

 UNLOADING  BELT  CONVEYOR

 UNLOADING  INCLINE  PEl.T
 CONVEYOR

 UNLOADING  PIT MIST COLLECTOR


 UNLOADING  HOPPER


 UNLOADING  VIBRATING FEEDER

 UNLOADING  KELT  CONVEYOR

 UNLOADING  INCLINE  BELT
 CONVEYOR

 UNLOADING  PIT MIST COLLECTOR


 UNLOADING  PIT SUMP PUMP

 STORAGE"  BELT CONVEYOR

 STORAGE  CONVEYOR TRIPPER

 MOBILE EQUIPMENT

 RECLAIh  HOPPER


 RECLAIM  VIBRATING FEEDER

 RECLAIM  BELT CONVEYOR

 RECLAIM  INCLINE BELT CONVFYOR

 RECLAIM  PIT DUST COLLECTOR

 RECLAIM  PIT SUMP PUMP

 RECLAIM  BUCKET ELEVATOR

FEED BELT CONVEYOR

FEE!" CONVEYOR TRIPPER
 20HP  SHAKER  7.5HP  HOIST

 25HP  PULLER ,  5HP RETURN

 14FT  DIA,  10FT STRAIGHT
 SIDE  HT, CS

 3.5HP

 20FT  HORIZONTAL r f-HP

 310FT, 50HP
POLYPROPYLENE  BAGTYPE,
2200 CFM,7.5HP
J6FT niA, 10FT STRAIGHT
SIDE HT, CS

3.5HP

20FT HORIZONTAL* 5HP

310FTr 50HP



POLYPROPYLENE. BAGTYPE,
2200 CFM»7.5HP

60GPM, 70FT HEAD, 5HP

  200FT,   5HP

30FPM, 1HP

SCRAPPER TRACTOR
1
1
1
1
1
1
1
1
1
1
1
28582.
49345.
4160.
12134.
1 7527 .
AO&70.
5258.
4180.
12134.
17527.
60670,
1866.
1866.
7711,
1866.
0,
24875,
12438.
7711.
1866.
0.
24875.
7FT UIDE» 4.25FT HT, :>FT  2
WIDE BOTTOMt CS
3.5HP

  200FT.    5HP

193FTt 40HP

POLYPROPYLENE BAG TYPE

60bPM» 70FT HEAD, 5HP

90FT HIGH, 75HP

  60.FT HORIZONTAL 7.5HP'  1

30 FPM, IMP                j


       113
  5250.


  3371.

 57974,

 13482.

136171,

  1079.


 24268.

 40447,

 37750.

  5258.

  3371,

 80894.

 20223.

 13482.
12438.


  746.-

16167.

 2488.

    0.

 1741.


 3731.

 8706.

13930.

12438.

  746.

 1617.

 1368.

 2488.

-------
FEED BIN


BIN WEIGH .FEEDER

GYRATORY CRUSHERS

BALL MILL DUST COLLECTORS'


BALL MILL

MILLS PRODUCT TANK
MILLS PRODUCT TANK SlURRY
PUMP
SLURRY FEED TANK



SLURRY FEED TANK AGITATOR

SLURRY FEED TANK PUMPS




TOTAL EQUIPMENT COST
13FT DIAr 2JFT STRAIGHT
SIDE HT» COVERED. CS

14FT PULLEY CENTERS, 2HP

75HP

POLYPROPYLENE BAG TYPE
2200 CFKr 7.5HP
                  16179.
                            29851 .
135.HP
 1O.OTPH.
5500 GAL 10FT PIAf 10FT
HT. FLAKEGLASS LINED CS
MILLS PRODUCT TANK AGITATOR .  10HP
  42.GPM. 60FT HEAD*
   2.HP»  2 OPERATING
AND 1 SPARES
  44334.GAL» 19.6FT DIAi  1
 19.AFT HTt FLAKEGLASS-
LINED CS

  41.HP                   1

   21.GPM, 6,0 FT HEAD»    9
   l.HP.  4 OPERATING AND
 5 SPARE
3
3
3
3
3V
3
3
54603.
1597(45.
15774.
442088.
14561.
24673.
7719.
3731.
15672*
37313.
40784.
22761.
1119.
1493.
                  10621.



                  29197.

                  22342.
22762.



 2155.

 4478.
                                                           1412984.   298907.
                                    114

-------
                                   SCRUBBING
         INCLINING  A OPERATING AND  2 SPARE SCRUBBING TRAINS
        ITEM
                                   DESCRIPTION
                                                       NO. MATERIAL    LABOR
F.D, FANS


SHELL
RUBBER LINING
MIST ELIMINATOR
SLURRY HEADER AND NOZZLES
GRIDS
SPHERES

   TOTAL SCRUBBER COSTS

REHEATERS

SOOTBLOUERS

EFFLUENT HOLD TANK
EFFLUENT HOLD TANK AGITATOR

COOLING SPRAY PUMPS



ABSORBER RECYCLE PUMPS



MAKEUP MATER PUMPS




TOTAL EQUIPMENT COST
 14.SIN H20i WITH 1195.
HP MOTOR AND DRIVE
 6  2135508.   123075.
231288.GAL»  34.OFT DIAt
 34.OFT HTt FLAKEGLASS-
LINED CS

  63. HP

1274.GPM 100FT HEADf
  59.HPi 4 OPERATING
AND  6 SPARE

 8741.GPMr JOOFT HEAD?
 406.HP*  8 OPERATING
AND 10 SPARE

6
6
72
6
6
12
974744.
1439962.
442464.
376418.
566157.
210822.
4010566.
1256471.
485362.
109948.
415024.
141024.

334259.
51940.
358206.
206827.
183777.
20822.
18
 2549.GPM»  200.FT HEAD.  2
 715.HP.  1 OPERATING
AND  1 SPARE
     790391.
      19790.
                62O76.
                 1826.
                                                           9364080.  1343607.
                                   115

-------
                               WASTE DISPOSAL
        ITEM
                                   DESCRIPTION
                         NO. MATERIAL    LABOR
ABSORBER BLEED RECEIVING
TANK
ABSORBER BLEED TANK AGITATOR

POND FEED SLURRY POMPS



POND SUPERNATE PUMPS




TOTAL EQUIPMENT COST
 57761.GAL. 17.OFT DIA»
 34.OFT HTt FLAKGLASS-
LINED CS

  36. HP

  642.GPMf  130.FT HEAD
  39. HPi.  1 OPERATING
AND  1 SPARE
1

2
  364.GPM»  192.FT HEAD,  2
  29.HPf  1 OPERATING
AND  1 SPARE
14579.



20467.

14366.



 7144.
31229.



 1511.

 2718.



  659,
                               56557.    3&116.
                                   116

-------
LIMESTONE SLURRY PROCESS — BASIS:  500




PROJECTED CAPITAL  INVESTMENT  fcE
 UNIT.  1978 STARTUP



- INDUSTRY MHIEF1NG'



          INVESTMENT, THOUSANDS OF 1977 DOLLARS
                                                                                                           DISTRIBUTION



EOUTPMFNT
MATERIAL
LABOR
PIPING
MATERIAL
LAROR
DUCTWORK
MATERIAL
LA^OC)
FOUNDATIONS
MATERIAL
LAPOR
POND CONSTRUCTION
STRUCTURAL
M4TE4MI
LABOR
ELECTRICAL
MATERIAL
LAPOrt
INSTRUMENTATION
MATERIAL
L«POR
BUILDINGS
MATERIAL
LAPOR
SERVICES AND MISCELLANEOUS
SUBTOTAL DIRECT INVESTMENT
ENGINEERING DESIGN AN-:) SUPERVISION
CONSTRUCTION EXPENSES
CONTRACTOR FEE^
CONTINGENCY
SUBTOTAL FIXFO INVESTMENT
ALLOWANCE FOR STARTUP AND MODIFICATIONS
INTEREST DURING CONSTRUCTION
SUBTOTAL CAPITAL INVESTMENT
LAND
WORK-ING CAPITAL
TOTAL CAPITAL INVESTMENT
PAW MATERIAL
HANOI Ir»G AND
Mi-! PAR AT ION

1197.
255.

197.
79.

0.
0.

10*.
442.
0.

??7.
HS.

141.
261.

9*.
21.

28.
44.
103.
3300.
297.
*>?«.
Ib1^.
330.
4620.
370.
554.
5544.
7.
12H.
5680.


! SCRUBBING

7800.
1116.

2431.
70?.
V
19*9.
1347.

94.
281.
0.

1 95.
436.

457.
701.

743.
124.

0.
0.
595.
1H991.
1709.
3039.
950.
1%99.
26588.
?1?7.
31*1.
31906.
3.
738.
32647.

WASTE
DISPOSAL

46.
29.

729.
270.

0.
0.

10.
30.
398*.

1.
4.

83.
191 .

7.
2.

0.
C.
174.
5561.
500.
B90.
278.
556.
7785.
623.
934.
9342.
979.
216.
10538.


TOTAL

9042.
1400.

3357.
1052.

1969.
13*7.

208.
754.
3984.

423.
525.

681 .
1173.

844.
147.

28.
44.
872.
27852.
?507.
4<*5r>.
1 3-»3.
2765.
38993.
3119.
4679.
*6792.
989.
1083.
48864.
PERCENT
OF OHECT
INVESTMENT

32.5
5.0

12.1
3.8

7.1
4.6

0.7
2.7
14.3

1.5
1.9

2.4
4.2

3.0
0.5

C.I
0.2
3.1
100.0
•5.0
16.0
5.0
10.0
140.0
11.2
16.3
168.0
3.<,
3.9
175.*

-------
LIMESTONE SLURRY  PROCESS — BASIS--  -*oo -• UNIT.  i<»7H

PROJECTED REVENUE REQUIREMENTS - INOUSTrtY KR1EF 1*6-500
                                         DISPLAY SiFET FOR YEAH
                                       ANNUAL OPERATION K«—
i
7000
                            27.44 TONS PtP
                                    TOTAL FIXEO INVESTMENT
     DRY
          D1BECI.CDSIS
                                                       l*fS.3  K  TONS
                                                         0.0  *>  TONS
      B.OO/TON
     50.00/TON
                 SUBTOTAL RAW MATERIAL
                              AND
UTILITIES
                                                     25990.0 HAN-HR

                                                    6*1550.0 K L«
                                                    21*81*0.0 K GAL
     12.50/HAN-HP

      2.00/K LB
      o.l2/< GAL
                                                      3760.0 rl»
      17.00/MP
      PROCESS hATEP
      ELECTRICITY
    MAINTENANCE
      LA40H ANO MATERIAL
    ANALYSES

       SUBTOTAL CONVERSION COSTS

       SUBTOTAL DIRECT COSTS

1MQ15ECJ-CDSIS

  DEPRECIATION
  COST OF CAPITAL AND TAXES*  17.20* OF  UNDEPRECIATED INVESTMENT
  INSURANCE *, INTERIM REPLACEMENTS.   1.17*  OF  TOTAL  CAPITAL  INVESTMENT
  OVE3HEAO
    PLANT,  50.o* OF CONVERSION COSTS LESS  UTILITIES
    ADMINISTRATIVE. RESEARCH, AND SERVICE.
      10.0* OF OPERATING LA^OR AND SUPERVISION

       SUBTOTAL INDIRECT COSTS

       SUBTOTAL ANNUAL REVENUE REOUIHEMtiMT

       SLUDGE FIXATION COSTS              192100.0  TONS

       TOTAL ANNUAL REVENUE REUUIftEMENT

       EQUIVALENT UNIT REVENUE PE
-------
VO
L1MF5TONE SLURRY P^OCFSb  —  ?USIS:   500  Mi  UNIT.  1<»7* STARTUP




PROJECTED LIFETIME REVENUE RtOUlREMtNTS  - INOUSTKY HwlEFlNf»-SOO "id




                                                 TOTAL CAPITAL INVESTMENT:   *
                                                                                    488o4000
SULFUR
YEAPS ANNUAL POtaFR UNIT POWfP UNIT F)Y
AFTER OPERA- HEAT FUtL POLLUTION
POWER TION, REQUIREMENT, CONSUMPTION* CONTROL
UNIT KW-HR MILLION flTu TONS COAL PROCESS,
STAPT^ /KK /YEAR /Y£A*> TONS/YKA^
1 7000
2 70PO
3 7000
4 7000
— 5.11_jn2fi
6 7000
7 7000
8 7000
9 7000
1.2 7&U2
11 5000
12 5000
13 5000
14 5000
.ll.ll.5Q2a;
14 3500
'.7 3500
1« 3500
19 3500.
?] 1500
23 1500
23 15CO
2* . 1500
?6 1500
27 1500
28 1500
29 1500
ltt-ll_15J2fl".
OT 1?7500
LIFETIME




EVNUE «EOL>
LEVEL IZEO


31500000 1^00000
31500000 1500000
3 1 o o 0 r. i o ' Iriooooo
31500000 1500000
315DQ0.2P. 150010.0.
31500000 1500000
31500000 1500000
31510000 1500000
31500000 1500000
3.l2°°222 1§2°220
22500000 1071400
22500000 1071400
22500000 1071400
2P500000 10714CO
15750000 75001)0
15750000 75fiOO)
15750QOO 7">0000
1^750r,QO 750000
6750000 321400
•S7DOfiOO 3314QO
A750000 J?1'00
*>7SOOOO 3?1400
6/60000 S/Jl^OO
67-.0000 3?MOO
«-.7bflOOO 321400
6750000 321400
£ 7 s o o o o 32i'*l!^

573750r.0fl 27321000
AVERAGE I'.CREiSE IN UNIT *FVENJF
DOLLARS PER TON OF COAL
MILLS PFR KILOWATT-HOUR
CFNTS PFP MILLION ^Tll MF
DOLLARS PER TON OF bULFn
I&rMfNT DJSCOUNTtO IT 11. ft* TO
I'. CREASE IN UMIT REVENCF Rtt-^l»
HOLLARS PER TON OF COtL i
MILLS PC& K ILO*-TT-«OU-<
3r>000
3SOOO
35000
3SOOO
35QOO
35000
35UOO
35000
35000
350.00
2SOOO
25000
?5000
I7»no
17500
17500
17^00
75UO
7500
7300
7600
7500
7500
7500
7500
BYPROOUCT
ADJUSTED GROSS
ANNUAL REVENUE
•*AT^ • SLUUGE REOUIWEMENT TOTAL NET ANNUAL CUMULATIVE
EQUIVALENT FIXATION FEE EXCLUDING ANNUAL INCREASE NET INC«ElSf
TONS/YEAR WTON SLUDRE SLUDGE IN TOTAL IN TOT*1
FIXATION FIXATION REVENUE REVENUE
DRY Dhy COST, COST, REQUIREMENT, REQUIREMENT,
SLUIXit SLUDGE S/YE4R $/YEA« $ §
iv!i!S
1^2100
103 100
1^2100
192100
192100
I1' ''I 00
137200
1.17200
137200
13720ft
96000
96000
96000
9^22
41200
4 1 ? 1 0
41200
41*00
41200
41200
41200
12.00
12.00,
12.00"'
12.00

'~12.00
12.00
12.00
12.00
15 QQ
12.00
12.00
12.00
12.00
12.00
12.00
12.00
12.00
~12*00
12.00
12.00
12. UU
~12*00
12.00
12.00
12.00
18078900
17bl(>600
175^2400
17274100

"~16737500~
1646V300
16201000
15932700
i "jhf4^(30
"136^2300"
13354000
13085flOO
12017500
10b77300
1060SOOO
10340700
10U72500
~~747l700~
7203500
6935200
b t* (5^>^0 0
h J *^^ *S 00
6130400~
5862100
5593SOO
5325600
2305200
2305200
2305200
2305200

2305200
2305200
23C5200
2305200
2305200
Ib4b400
1646400
1646400
1 6** 'i^t 0 0
J 646400
1152000
115200C
11520CO
1152000
494400
494400
494400
494400
494400'
494400
494400
494400
20384100
201 15SCO
19847600

19042700
18774500
18506200
18237900

1526H/00
15000400
14732200
K463900
12029300
1 17olOOO
11492700
11224500
79661i>0
7697900
742960U
7161300
	 6624dOO 	
6356500
6046200
5820000
20384100
60347500
' ' -\ "- -»
^ 22 jZ^QQ.
1 1 H dh 050 0
137055000
ISSSfrlc-OO
173799100

"20/037500~
222037900
23ft770! 00
2512J4COO
2774bd900
2"92 ' 9930
3007! ?f> j.l
311937100
33G039-.30
3365S7300

^'•!*:'"::.'
36666^000
373 0?2500
j7y 1 1 0700

	 lifiQ 	 £1222 	 12*0.0 	 T-StSJjflO. 	 49.J4JJP. 	 55512(12 	 __3.iU4d2*0.2_
637^)00
t>tOuI«FM£
dUWNtO

IT INPUT
R wtMilVErt
34V9000
NT










INITIAL YEA«, HOLLARS
t^tNT ECUI
rJU^'lti>

v«Lf.NT TO OISCOUNTEO


REQUIREMENT


348494400

12.76
5.47
bn.74
546. TC
125705100
OVf< LIFE
11.80
5.06
CENTS PER MILLION HIU HFAT INPUT 56. 21

DOLLARS PFR TON OF iULFi"
< MfMOVFD


505.86
419A8000

1.53
0.66
7.32
65.86
16366000
OF PO»ER
1.54
0.66
7.32
65. 8*
390482400

14.29
6.13
68 • Of)
V fc» . V -f
612.52
142071100
UNIT
13.34
5.72
63.53
571.71













-------
            INCLUDING  1 OPERATING AND  1 SPARE PREPARATION UNITS
           ITEM
                                      DESCRIPTION
                         NO. MATERIAL    LABOR
   CONVEYOR FROM CALCINATION
   PLANT
                                 1500FT HORIZONTAL* 30HP
                              153293.
46144,
                                                                                        CD  ~O
NJ
C
H-  STORAGE SILO ELEVATOR

   CONCRETE STORAGE SILO



   STORAGE SILO HOPPER BOTTOM

   RECLAIM VIBRATING FEEDER

   RECLAIM BELT CONVEYOR

   FEED BIN ELEVATOR

   FEED BIN
  29,FT HIGH» 50 HP

   659.FT3r 8.2FT DIA »
 12.4FT STRAIGHT SIDE
STORAGE HT

60 DEGREEr CS

3.5HP .

  83.FT HORIZONTAL, 5HP

SOFT HIGH» 50HP

10FT DIA»  15FT STRAIGHT
SIDE HT» COVERED, CS
   BIN.VIBRATING FEEDER

   BIN WEIGH FEEDER

   SLAKER

   SLAKER PRODUCT TANK

   SLAKER PRODUCT TANK AGITATOR  10HP
                                 3, 5HP

                                 12FTr  12IN  SCREWr  IMP

                                    O.TPHt     O.HP
1
1

1
1
1
1
2
2
2
2
2
2
32047.
3161.

401,
12134.
15232.
52311.
5393.
9168,
11864.
24960.
12134.
14291 .
650,
7624.

1194.
1866.
3953,
995.
9950,
3234.
1244,
2303.
18905.
746.
i > i **->
5 3
r- o
CO
— 1 3>
yo











-------
LIME SYSTEM DUST COLLECTORS
SLAKER PRODUCT TANK SLURRY
PUMPS
SLURRY FEED TANK



SLURRY FEED TANK AGITATOR

SLURRY FFiED TANK PUMPS
~  TOTAL  EQUIPMENT COST
                                 POLYPROPYLENE BAG TYPE
                                 2200 CFhi7.5HP

                                    l.GPMi  60FT HEAD»
                                    O.HPi   1 OPERATING
                                 AND 1 SPARES

                                     680,GALF  4.9FT DIA»
                                   4.9FT HTf FLAKEBLASS-
                                 LI NED CS

                                    1 .HP

                                     O.GPMt  60 FT HEADi
                                    0,HP»   4 OPERATING AND
                                  5 SPARE
 1

 9
21032,


 4448,



  656,



 2867.

19444.
                                   SCRUBBING
         INCLUDING  4 OPERATING AND  2 SPARE SCRUBBING TRAINS
49751,


  995,



 1406.



  212,

 4476,
                                                            394838.    155649,
                                                                                     CO
                                                                                     o  o
                                                                                     o  :z
                                                                                     z  rn
                                                                                     c:  a>
        ITEM
                                      DESCRIPTION
NO, MATERIAL    LABOR

-------
                                     WASTE DISPOSAL
              ITEM
                                   DESCRIPTION
                         NO, MATERIAL-    LABOR
to
to
      ABSORBER BLEED RECEIVING
      TANK
ABSORBER BLEED TANK AGITATOR

THICKENER FEED PUMP



THICKENER


THICKENER OVERFLOW PUMPS



THICKENER OVERFLOW -TANK


SLUDGE FIXATION FEED PUMP
 52510. GAL f 16.5FT DIAi
 32.9FT HTi FLAK GLASS-
LINED CS

  34, HP

  665.GPMr 60FT HEAD.
  lO.HPf  1 OPERATING
AND  1 SPARE

 18248.SQ.FT.rl52.FT DIAi  1
  B.6FT HT

  438.GPMF  75, OFT HEAD*   2
  14. HP*   1 OPERATING
AND  1 SPARE
                                                             13682.    29306,
1
2
1


18967.
13078.
536285 .


1400.
2752.
366827,




m
•— •
3


O
-a
—i
o
CO


CO
rn
o
CO
  7224, GAL i
  B.6FT  HT
                                                 12. OFT DJAr   1
                                      207.GPMr 5 OFT HEADr
                                       6, HP,  1 OPERATING
                                    AND  1 SPARE
                                                                    5913.
1637,
                                10575.
3V27,
          1816,
      TOTAL EQUIPMENT COST
                                                             600137.    406574.
                                                                                       m
                                                                                       CO
                                                                                              CO

-------
                                      WASTE. DISPOSAL
               ITEM
                                     DESCRIPTION
                                                               NO*  MATERIAL    LABOR
to
       ABSORBER BLEED RECEIVING
       TANK
 ABSORBER BLEED TANK AGITATOR

 THICKENER FEEH PUMP



 THICKENER


 THICKENER OVERFLOW PUMPS



 TH 1CKENER 0VERFl 0U TANK


 FILTER FEED SLURRY PUMP



 FILTER


 FILTRATE  PUMP  (PER FILTER)



 FILTRATE  SURGE TANK


FILTRATE-  SURGE TANK PUMP
  52510.GALr 16.5FT DIAr   1
  32.9FT  HT, FLAKGLASS-
 LINEP CS

   34.HP                    1

   652.GPMF  60FT HEAD,     2
   IB.HP,   1 OPERATING
 AND   i SPARE:

  17913.SQ.FT.,151.FT MIA, 1
   8.5FT  HT

   430.6PMr   75.OFT HEADr   2
   14.HP,   1 OPERATING
 AND   J SPARE

   7092.GAL,  11.9FT D1A,   1
   8.5FT  HT

   l()2.GPMr  SOFT HEADi      3
   3.HP,  2 OPERATING
AND   1 SPARE

 269.SQ FT  FILTRATION      2.
AREA

  45.GPM,   20.OFT  HEAD,    4
   O.HP»   2  OPERATING
AND  2 SPARE
                                       1482. GAL,
                                       6.3FT  HI
              6.3FT DIA,   1
13682.



18967.

13053.
                                       90.GPM,   85.OFT HEAP)'  2
                                       3.HP,'  1  OPERATING
                                    AND  1 SPARE
                                                                      5877.
                                                                      1618.
                                                                    10750.
                                                                   180058.
                                                                     6211.
  594.
                                 3874,
29306.



 1400,

 2734.
                                                                    532294.    362329.
            542.
           3880 .
           2114,
          16611 .
            573.
 1424.
                                                                           357,
                                                                                                CO
                                                                                         (ZD
          ^^  C^  C3
          ~o  -o  •—•
          13  —I  CO
          m  *—i  ~o
          s:  CD  o
          —I  2:  co
          r^  -t  i—
          CO     .T>

-------
                    FIVE PROCESS STUDY OPERATING PROFILE
60
20
 0
          ' '
0
               l
               10
T
20       30        40
     BOILER ME - YEARS
                           50
1  '  I
  60
                                                                I I
70

-------
            FPC OPERATING PROFILE
ALL BOILERS AVERAGE CAPACITY FACTOR  vs. BOILER AGE-
             BASED ON 1969-1973 FPC DATA
                   BOILER AGE-YEARS
                        125

-------
                        NATIONAL ECONOMIC  RESEARCH  ASSOCIATE'S SENSITIVITY STUDY
NJ
                 160
                 120
                 100
             GO
                  80
                  60
                          300-MW UNIT
                          LIMESTONE SYSTEM
                                          I
                                                              COAL SULFUR
                                                              CONTENT, LB
                                                              S02/MBTU
                                         ,8         1,2      1,6
                                    SULFUR  REGULATION,  LB S02/MBTU
2,0

-------
                        NATIONAL ECONOMIC RESEARCH ASSOCIATE'S SENSITIVITY STUDY
ro
                 110
                               1
                         300-MW UNIT
                         LIME SYSTEM
              cr>
                 120
              LU
                 100
                  80
                  60
COAL SULFUR
CONTENT, LB
S02/MBTU
                                        ,8        1,2       1,6
                                    SULFUR REGULATION, LB S02/MBTU
      2,0

-------
oo
                     NATIONAL ECONOMIC RESEARCH ASSOCIATE'S SENSITIVITY  STUDY

                                                   	1	
                           )-W
600-MW UNIT
LIMESTONE SYSTEM
               130
            CD
               110
                90
           C/5
                70
               50
                                                             COAL SULFUR
                                                             CONTENT,  LB
                                                             S02/MBTU
                                       ,8       1,2         1,6
                                 SULFUR REGULATION, LB S02/MBTU
                                            2,0

-------
          NATIONAL ECONOMIC  RESEARCH  ASSOCIATE'S  SENSITIVITY  STUDY
130
                 1

             600-MW  UNIT
             LIME SYSTEM
   110
01
    90
CO
    70
    50
                                                COAL SULFUR CONTENT
                                                LB S02/MBTU
                                                            8
                          ,8       1,2        1.6       2,0
                      SULFUR REGULATION, LB S02/MBTU

-------
                           COSTS OF  REDUNDANCY
100
 90
 70
        UJOOfW
        8 OPERATING TRAINS
     SYSTEM
SULFUR
 60
                           234
                              REDUflATfT TRAINS

-------
o
00
                                                       T
I-]
M
S



w
M
o*

w
z
[I]
                                             ONSITE POND  OPTION
                                             3.5%  S COAL
                                             1.2 LB S09/MBTU
                                             INCLUDING^FLYASH
                                             DISPOSAL
                                              OPERATING PROFILE
H  4
                         1
                                   1
 I
             200
                       400        600        800

                           POWER  UNIT  SIZE,  MW
1,000
            LIMESTONE PROCESS  -  EFFECT OF POWER UNIT SIZE
          AND OPERATING PROFILE  ON UNIT REVENUE REQUIREMENT

-------
                                                          T
                               T
CO
N5
            O
            CO
            ^
            2
            H
            Z
            w
             w
             z
             w
                     ONSITE POND OPTION
                     3.57. S COAL
                     1.2 LB S02/MBTU
                                                                   LIMESTONE
                                       I
                                I
                           200
400       600       800

 POWER UNIT SIZE, MW
1,000
                             EFFECT OF POWER UNIT SIZE AND  PROCESS
                                 ON UNIT REVENUE REQUIREMENT

-------
CO
u>
o
oo
ON
•-*  7
». j  *
us
CO


I  6
              H
              z
              w
              3
              2
              H
              M
              2
                                                          ONSITE POND OPTION
                                                          3.5% S COAL
                                                          1.2 LB S02/MBTU
                                 INCLUDING FLYASH
                                  DISPOSAL
                                           EXCLUDING FLYASH
                                            DISPOSAL
                          200
                      400         600       800

                          POWER  UNIT  SIZE, MW
1,000
                         LIMESTONE PROCESS - EFFECT OF POWER UNIT SIZE
                            AND FLYASH ON UNIT REVENUE REQUIREMENT

-------
CO
              120
            z
            1100
en

>
M

H
M
Z
               80
                          200
                         400
600
                                                         T
                                                       T
                                                         ONSITE POND OPTION
                                                         3.57.  S COAL
                                                         INCLUDING FLYASH
                                                          DISPOSAL
800
                                         POWER UNIT SIZE,  MW
                                                                        VAL
                                                                    I
1,000
                          LIMESTONE  PROCESS  -  EFFECT OF POWER UNIT SIZE
                       AND % SO- REMOVAL  ON UNIT INVESTMENT REQUIREMENT

-------
              O
              oo
     r        i          r

ONSITE POND OPTION
3.5% S COAL
INCLUDING FLYASH DISPOSAL
CO
Ln
                                         7. S02 REMOVAL
                                                 I
                                   I
  I
                            200
              400       600       800


                  POWER UNIT SIZE, MW
1,000
                           LIMESTONE PROCESS - EFFECT OF POWER UNIT SIZE

                          AND 7.  S02  REMOVAL ON UNIT REVENUE REQUIREMENT

-------
    140
•co-
H
Z
120
    100
     80
 H
 M
 2
                            I
                                  I
                200
                                                         T
                                            ONSITE POND OPTION
                                            1.2 LB S02/MBTU
                                            INCLUDING FLYASH
                                             DISPOSAL
I
I
                      400       600       800

                        POWER UNIT SIZE, MW
         1,000
               LIMESTONE PROCESS - EFFECT OF POWER UNIT SIZE
               AND S IN COAL ON UNIT INVESTMENT REQUIREMENT

-------
o
oo
Cfl
h-J
1-1
W
w
G
2
W
H
M

3
                                               T
                               T
                     ONSITE  POND  OPTION
                     1.2 LB  S07/MBTU
                     INCLUDING  FLYASH
                       DISPOSAL
                                                              2.0__
                           I
           1
I
I
               200
400       600       800

    POWER UNIT SIZE, MW
         1,000
              LIMESTONE PROCESS - EFFECT OF POWER UNIT  SIZE
                AND S IN COAL ON UNIT REVENUE REQUIREMENT

-------
         COAL CLEANING VERSUS  100%  FGD

CASE 1  ROM COAL •*• 2000 MW •*• FGD
CASE 2  ROM COAL •» PREP PLT ->  2000  MW + FGD
CASE 3  ROM COAL •* CHEM COMM & PREP PLT + 2000 MW •> FGD
                       138

-------
                          CASE  1	CASE  2	CASE 3
CAPACITY FACTOR              45            56            56
GENERATION  (KWH/YR)     7.88  x  109    9.81 x  109    9.81 x 109
PRODUCTION COSTS
 (MILLS/KWH)                15.2         12.7         12.9
FGD REVENUE REQUIREMENT
 (MILLS/KWH)                 7.1          6.1          5.8
TOTAL GENERATION COSTS
 (MILLS/KWH)                22.3         18.8         18.7

-------
       LANDFILL AND PONDING CONCEPTS

           FOR FGD SLUDGE DISPOSAL
                        by
Jerome Rossoff,  Paul P. Leo, and Richard B. Fling
            The Aerospace Corporation
              El Segundo, California
                  Presented at the

      U. S.  Environmental Protection Agency
   Industrial Environmental Research Laboratory
          Industry Briefing Conference on
   Technology for Lime/Limestone Wet Scrubbing
      Research Triangle Park, North Carolina
                 August 29,  1978
                    140

-------
                    ABSTRACT
This paper is concerned with the environmentally
sound disposal of flue gas desulfurization (FGD)
sludges.  The environmental considerations and
the technology and costs associated with the dis-
posal of FGD sludges by landfilling and ponding
are summarized.  Concepts discussed are lined
ponds, unlined ponds equipped with underdrainage,
chemical treatment  and landfilling, and conversion
to gypsum.   The need for environmental control is
reviewed.  The capabilities  of each concept to pre-
vent water pollution and the  environmental consid-
erations that require site maintenance are discussed.
Bearing strengths associated with  landfill concepts
are included, and the status of developments of non-
operational concepts,  i. e. ,  ponding with under-
drainage and the disposal of FGD gypsum,  are
discussed, Additionally,  disposal  site volume
requirements and estimated disposal costs are
given.
                     141

-------
Introduction

This paper summarizes current concepts of landfilling and ponding for the
environmentally sound disposal of flue gas desulfurization (FGD) sludges.  The
techniques discussed herein represent the results of studies and assessments
performed by The Aerospace Corporation under contract to the Industrial
Environmental Research Laboratory of the  U.S. Environmental  Protection
Agency (EPA),  Research Triangle Park,  North Carolina.   These techniques
do not constitute endorsement or approval by the EPA, but are presented as
the authors' assessments of the  best available methods for the disposal of
FGD sludges by landfilling  or ponding.

With the passage  of the Resource Conservation and Recovery Act (RCRA),
public law 94-580, October 1976, guidelines and criteria are forthcoming  for
application to FGD sludges.  Determinations will be made by the EPA as  to
whether these sludges are  to be  considered hazardous, and, depending on
those determinations, criteria will be developed for FGD  sludge disposal.

Without federal criteria applicable  specifically to FGD sludges,  almost all
studies and developments up to this  time  have  used drinking water criteria
as the basis for establishing requirements for disposal.  Because the trace
element and salt content  of most samples analyzed exceeded  the drinking
water  criteria at  least for  some of  the constituents, the general approach
taken has been to dispose of FGD sludges such that no direct  discharge to  any
water  supply would be permitted, that any seepage would be minimized or
perhaps totally eliminated,  and that runoff would be controlled.  Additionally,
a  strong effort has been made by industrial and government agency develop-
ment and evaluation programs to determine disposal techniques  that would
not only be  environmentally sound from the standpoint of water quality control
and, when practical, would also result in reclamation of the land area selected
for the disposal site.  As a result,  all disposal techniques that have been de-
veloped for FGD sludge are intended for the control of water  quality, but not
all of them produce  reclaimable disposal sites.  The techniques  discussed
herein consist of  the following:  (a)  ponding of untreated sludges, (b) disposal
of untreated sludges in ponds equipped with underdrainage systems,  (c) chem-
ical treatment and landfilling, and (d) conversion to gypsum and  subsequent
disposal.

The basic characteristics of each of these approaches are discussed as to
protection of water supplies, land reclamation, and disposal  costs.


Water  Quality Criteria

A  comparison of chemical constituents from a large number of analyses of
sludge liquors in a discharge stream with the National Interim Primary Drink-
ing Water Regulations (40 CFR 141)  is given in Table I as  a ratio of  constituent
concentration to water criteria.  These ratios are given for the range of con-
stituents from a composite  of data for ten eastern and western sludges, with
and without fly ash,  and for the ten  independent samples.  It should be noted
that the values used in this  comparative analysis represent the initial concen-
trations that would seep from the base of an untreated sludge  pile.

                                142

-------
                                  Table I.  Comparison of Sludge  Liquors  with Water Criteria
NIPDWR
Drinking
Criteria,
mg/t
As 0. 05
Cd 0.01
Cr 0.05
Pb 0.05
Hg 0.002
Se 0. 01
F 2
TDS 500
pH (actual
values)*3
Concentration -f- Criteria (Nondimensional)

All
Samples
< 0. 8 - 2. 8
0.4 - 11
0.2Z - 5
0.2 - 6.6
0.03 - 2.5
0.28 - 20
< 0. 5 - 5
6.6 -48.5
6. 7 - 12.2
Sample
A

0.6
5.0
5.0
0.8
2.5
10.0
_ _
36
6.7
B

0.4
1.2
0.8
3.0
_ _
3. 3
0. 5
6.6
6.8
C

2.0
0.4
1.8
4.6
	
10.0
3. 3
30.0
8.0
D

0.04
_ _

<0. 2
<0. 1
4. 2
- -
13.4
12.2
E

0.4
11
0.6
6.6
< 0. 5
< 2
1.7
18.8
8.7
F

1. 2
1. 3
0. 2
0. 2
< 0.001
7. 8
1
20. 5
8.0
G

2.8
--
--
< 0.2
< 0. 1
20
--
28
7.8
H

0. 1
--
--
< 0. 2
< 0. 1
14
--
18.4
7. 3

I

0.8
5
--
0.8
0. 1
2.8
5
8.4
10.7

J

0.2
2. 5
1. 1
< 0. 1
0.03
0. 3
< 0. 5
48. 5
8.9
*-
OJ
       Sample data are as follows:
         Sample          Station
            A
            B
            C
            D
            E
            F
            G
            H
            I
            J
Mohave
Cholla
Shawnee
Shawnee
Shawnee
Shawnee
Shawnee
Shawnee
Duquesne Phillips
LG&E Paddy's Run
       EPA-proposed secondary regulation

      GForced-oxidized to gypsum.
    Absorbent

  Limestone
  Limestone
  Limestone
  Limestone
  Lime
  Lime
  Lime     c
  Limestone
  Lime
  Carbide lime

is 6. 5 to 8. 5.
                                          Ash
Sampling Date
3
59
40
6
40
6
6
6
60
12
Mar
Nov
Jun
Jan
Jun
Sep
Oct
Aug
Jun
Jul
1973
1974
1974
1977
1974
1976
1976
1977
1974
1976

-------
 In Table I, the ratios for the  range of constituent concentrations of the com-
 posite data show that all elements  analyzed,  as well as the total dissolved
 solids (TDS)  and pH, exceed drinking water criteria.  However, in observing
 the ratios for the ten independent samples  shown in the table, it can be seen
 that,  except for selenium in two samples and cadmium in one, no trace ele-
 ment exceeds the criteria by  a factor greater than 10.  (Water criteria for
 barium,  nitrate, and silver are  1,  10, and 0. 05 mg/.e,  respectively.   Limited
 field evaluation leachate data  show maximum concentrations for these elements
 to be about 5, 1, and 0.5 times the criteria,  respectively.)   The TDS are high
 for most of the samples,  and  the pH is excessive for two of the samples.
 Although trace elements are  not eliminated as a matter of concern for some
 sites by these data,  there are indications that in many cases the concentra-
 tions are quite low and that,  generally,  the concern may be for the concen-
 tration of dissolved  solids and,  in  some cases, pH.  Chemical oxygen demand
 (COD) was considered somewhat differently.  Values  of  COD in fresh sludge
 ranged between 40 and  140 mg/£, but, because of the  rapid oxidation charac-
 teristic of sulfite sludge, the  COD after one pore volume displacement by
 leaching was  10 mg/f or less  and rapidly decreasing.  Therefore,  because
 COD is significant only for fresh sludge and because the sludge is not dis-
 charged directly to streams,  it was concluded that COD is not a critical
 parameter.

 Because of depletion of the material with leaching time,  cation exchange and
 adsorption in the soil,  and dilution  between the disposal site and the consumer
 tap,  it is difficult at this time to specifically quantify the degree of pollution
 potential at a given site.  Therefore, because of the  comparatively large  con-
 centration of  dissolved  solids  and the identification of random values  of high
 concentrations of trace elements, methods  for disposal of these materials to
 prevent their access to public water supplies were assessed.


 Results

                             Operational Modes

 By the end of 1977,  SO2 scrubbers  were operating at 22 power stations having
 a  scrubbing capacity of approximately 10,375 MWe;  20 have  nonregenerable
 scrubbers, and 2 have regenerable scrubbers.  In tne nonregenerable category
 7  stations  (scrubbing 4460 MWe) use chemical treatment disposal processes,
 and 13 stations (scrubbing 5680 MWe) dispose of the sludge untreated.  The  two
 regenerable systems have a total capacity of  235 MWe.   A breakdown of the
 disposal modes is as follows:

                         Treated,  Untreated,  Gypsum, Stabilized,21 Untreated,
               Treated, Lined    Unlined    Lined     Unlined      Solar
               Unlined    Pond      Pond      Pond      Pond     Evaporation

 No. of Plants     61714           i

 Total MWe      4293      167      3250       1420      635         375
Stabilized, e.g., dewatered with fly ash addition; not necessarily the final
 disposal mode.

                                  144

-------
                           Disposal Alternatives

The general categories of disposal and the considerations required for environ-
mental control are shown in Table II.  In each case,  seepage of rainwater
through the sludge and eventual contamination of groundwater pose an environ-
mental concern for all disposal methods.  Runoff is a potential source of en-
vironmental pollution for landfill sites because these sites are open and do not
necessarily return water to the scrubber.  Only  in the case  of ponding is it
clear that the disposal site is not directly  amenable to land reclamation efforts,
although even in some of these cases  it may be possible upon retirement to  air-
dry» cap,  and vegetate the site.   Consideration of each of these effects are
    n in the following discussions.
      Ponding.  In general,  the simplest and the least cost (though not neces-
sarily the most environmentally sound) approach to FGD sludge disposal is
ponding.   This method requires that, if the pond does not contain a base mate-
rial considered to be impermeable,  a liner must be added to prevent seepage.
Operationally, sludge ponds exist today which contain either naturally imper-
meable soils or clay liners  transported and placed in the base and on the slopes
of the pond.  Because of the highly thixotropic nature of these sludges,  ponds


          Table II.  Environmental Effects of Disposal Alternatives
Type of
Disposal
Pond


Basin

Landfill


Condition
of Waste
Untreated3-
or
chemically
treatedb
Untreated3-
or
conditioned0
Conditioned0
or
chemically
treated0
Primary
Drainage
Supernate

Supernate
Underdrainage

Runoff


Environmental Effect
Seepage
Yes

Yes
Yes

Yes


Runoff
No

No
No

Yes


Land Reuse
No

Yes
Yes

Yes


 aUntreated waste refers to FGD sludges as emitted from primary or secondary
  dewatering equipment.

  Chemically treated sludges refer to the waste treated by one of several com-
  mercial processes that make these wastes  suitable for  landfill disposal.

 cConditioned waste refers to  sludge treated  by techniques other than chemical
  treatment and includes  oxidation to gypsum and dewatering by mixing with dry
  fly ash or other agents  that allow the material to be handled in a manner
  similar to  that for  soils.
                                 145

-------
 are nonstructural sites and generally are not considered amenable to recla-
 mation, except possibly in areas  of low rainfall and high evaporation.  Also,
 if ponded  sludges are not dewatered, larger land areas are needed to contain
 the m ate rial.

       Ponding with Underdrainage.  This approach to ponding is still under
 evaluation and is not being used operationally at this time.   Underdraining and
 collection of all seepage for return to the scrubber system maintains control
 of leachate at all times and has been shown at small  scale evaluation sites to
 produce a material structurally capable of supporting personnel and construc-
 tion equipment.   Present evaluations1 are being made to determine the feasi-
 bility of such an alternative regarding (a) site reclamation and (b) relaxation
 of requirements on the degree of  water-tightness of the  base material, inas-
 much as no  appreciable hydraulic head exists.  A site of this type collects
 rainfall via  the seepage system and returns it to  the  scrubber.  As  a result it
 is  necessary to limit the size of each disposal basin to maintain an  acceptable
 •water balance in  the scrubber loop.  This would be accomplished by dividing
 the site into sections of approximately 35 to 50 acres at a depth of about
 30 feet.  Figure 1 shows an underdrained, untreated pond supporting a general
 purpose farm tractor within one day after a 3-inch rainfall.

      Chemical Treatment.  The  stabilization of  FGD sludges by chemical
 treatment offers the most  positive solution to the disposal problem.  It con-
 verts the sludge to a structural material;  decreases its coefficient of perme-
 ability to a range of  approximately 10-5 to 10~7 cm/sec,  which as a minimum
 is one order of magnitude  better than untreated sludges; reduces the concen-
 tration of  salt constituents in the  leachate by approximately  50%; is amenable
 to  subgrade  or above-grade landfilling; and allows the disposal site  to be
 reclaimable. Chemically  treated sludges have  not been  shown to appreciably
 reduce concentration of trace elements in leachate,  and, even though the  con-
 centration of major species is reduced,  leaching  of chemically treated sites
 should  be avoided unless it can be assured that the leachate can be diluted by
 local groundwater and  streams.  A general procedure for managing rainfall
 runoff from  a chemically treated site is to collect the runoff in a peripheral
 ditch which directs the  water to a  settling pond.   Depending on the quality of
 the water in this pond,   it can be decanted  to a stream or  returned to the
 scrubber system.

 Chemically treated sludges have solids content  of approximately 45  to 65 wt%
 (or possibly higher depending on the dewatering potential and the treatment
 process used) and attain load bearing strengths in the range  of 75 to  300 psi
 (5.4 to  21.6  tons /ft2\

 Test ponds containing chemically treated sludges  are pictured in Figures  2
 and  3.

      Gypsum.  The  forced oxidation of sulfite sludges to gypsum or the pro-
 duction of high sulfate sludge from the use of western coal results in a waste
 material which is readily dewatered by vacuum filtration or by centrifuging
 to a solids content in the approximate range of 75 to 85 wt%.   Leachate from
 gypsum is  similar to that of sulfite sludges and therefore should be prevented
from entering water  supplies.  Because  gypsum tends to form a protective

                                 146

-------
Figure 1.  Untreated,  unstabilized
  sludge ponded with underdrainage,
    one day after 3-inch rainfall.
Figure 2.  Chemically treated sludge
  (IU Conversion Systems process).


Figure 3.  Chemically treated sludge
          (Dravo process).

               147

-------
 surface scale capable of shedding rainwater,  tests are currently being con-
 ducted to determine the applicability of the disposal of gypsum on the ground
 without the added benefit of liners or impoundment dikes.   Limited results
 have shown that gypsum sludges crack badly under freeze-thaw conditions,
 thereby allowing rainwater to enter into the material.  Additionally, gypsum
 sludge slumps in its freshly deposited condition when exposed to  rainfall
 and produces a runoff containing potentially high concentrations of dissolved
 solids from the  sludge,  as well as a condition which requires  machinery to
 replace the material on the disposal site.  These preliminary results indicate
 that considerable site maintenance may be required on an operational scale
 to reconfigure the  disposal pile after weathering (freeze-thaw and erosion)
 and to control the runoff.   Tests are continuing for the determination of what
 control (if any)  should be exercised at the site during and after disposal,  A
 gypsum test pile before and after weathering is pictured in Figures 4 and 5.

                          Sludge Volume Prediction

 Landfill volume  requirements are strongly affected by the solids  content of
 sludges.  A comparative analysis  of sludge production in acre feet annually is
 shown for a 500-MW plant in Figure 6.   (This figure neglects  the approximate
 25% increase in  acreage requirements to account for berm slopes and access
 roads.)   If it is  assumed that an untreated sludge settles to approximately
 50% solids, the acre feet produced in one year for this case would be 250.  The
 advantage for gypsum in this regard (neglecting other environmental factors)
 would be that approximately 155 acre ft would be produced,  providing that the
 sludge  is dewatered to a solids content  of 80%.  In the case of chemical treat-
 ment, if it is assumed that the material is disposed of at a solids content in
 the range of 60 to 70%, the volume to be disposed of would be  in the range of
 165 to  190 acre ft.

                          Disposal Cost Estimates

 Cost estimates for ponding and chemical treatment for landfilling have been
 made and reported by The Aerospace Corporation on several occasions.  Dur-
 ing studies associated with the EPA  Shawnee field disposal evaluation project,
 Aerospace cost estimates were made of chemical treatment disposal  and were
 reported in the initial report on that study. ^  The Aerospace estimates for
 lined-pond costs were presented in the initial and second progress reports on
 sludge disposal^'^ and at EPA flue gas  desulfurization symposiums. 5, 6, 7 ^11
 estimates  have been updated in a report to EPA** on new source performance
 standards  on a July 1977 basis; these cost estimates are summarized in
 Table III.
Conclusions

Constituent concentrations  of FGD sludges require disposal controls to prevent
direct discharge,  seepage, or runoff to water supplies.  The methods  used
operationally today are (a)  disposal of untreated sludges in ponds with  highly
impermeable liners or (b) chemical  treatment prior to sub-grade or above-
grade landfilling.  Other methods being evaluated are (a) disposal of untreated
sludges in ponds equipped with underdrainage and (b) conversion to  gypsum
for disposal.
                                 148

-------
Figure 4.  Gypsum filter cake immediately
    after placement, September 1977,
           Paducah, Kentucky.
 Figure 5.  Gypsum filter cake after first
       winter season, March 1978,
           Paducah,  Kentucky.
                   149

-------
                     300 r
                     280
                     260
                   £ 240
                   a
                   O
                   cc
                   * 220

                   3
                   ^ 200
                     180-
160-
                     140
                      90%S02 REMOVAL
                   L!ME UTILIZATION    90%
                   LIMESTONE UTILIZATION 80%
                                              GYPSUM (45% ash)
                      40    50     60     70     80
                                   SOLIDS CONTENT, wt %
                               90
100
                     Figure 6.  Sludge produced annually
                      (500-MW plant,  3.5% sulfur coal,
                           12,000 Btu/lb,  14% ash).


 Untreated sludge ponds have the disadvantage of not being reclaimable.  Those
 equipped with underdrainage maybe reclaimable, depending on evaluations
 now in progress.

 Chemical treatment improves the impermeability of sludges by one order of
 magnitude or more, reduces  the dissolved solids concentration by about 50%,
 and attains a bearing strength greater than 5 tons/ft2.  Chemically treated
 sites must be  maintained to control seepage or runoff,  depending on the process
 used.   Above-grade sites generally require maintenance for runoff control only.

 Gypsum sludges dewater readily to 75 to 85 wt% solids.  These materials when
 stacked have exhibited severe surface cracks after freeze-thaw cycling. There-
 fore, piling  or stacking gypsum without considerable site maintenance may not
 be a feasible disposal method, on the  basis of preliminary field tests.  Further
 testing is under way.

 Volume production for a 500-MW eastern plant,  on the  average,  is approxi-
 mately 250,  175,  and 155 acre ft annually for untreated,  chemically treated,
 and gypsum  sludges, respectively.  For a 1000-MW plant, these values would
 be increased by about 93%.  Landfill requirements for these volumes are in-
 creased by approximately 25% to account for berm slopes and access roads.

 Disposal cost estimates in mills per kilowatt hour (July 1977 dollars) for pond-
 ing on  indigenous  clay, ponding with liner  added,  and chemical treatment are
0.55, 0.80,  and 1.05,  respectively, fora 1000-MW plant burning  typical
eastern coal.

                                 150

-------
               Table III.  Disposal Cost Comparison'
Cost Basis ,
Mid- 1977 $
Mills/kWh
$ /ton of
sludge (dry)
S / ton of
coal
Ponding
Indigenous
Clay
0. 55
4. 90
1. 50
Liner
Added
0. 80
7.25
2. 20
Landfill,
Chemical
Treatment
1.05
9.70
2.95
Gypsum
1. 10
10. 30b
3. 10
Notes:
     Dollar base:

     Plant characteristics;
     Coal burned:

     Annual average
      operating hours:

     Plant and  disposal site
      lifetime:
         removal,  with
      limestone absorbent:

     Limestone utilization:
     Sludge generated:


     Average annual capital
      charges, 30-yr average:

     Cost of land used for
      disposal:

     Land depreciation:


     Disposal site:
July 1977

1000 MW,  8700 Btu/kWh
(0. 73-lb coal/kWh)

3.5% sulfur, 12,000 Btu/lb, 14% ash


4380 hr/yr (30-yr average)


30 yr

90%

80% of all cases except for gypsum,
which is 100%

4. 8 X 10  short tons/yr untreated
waste (dry) including ash


18% of total capital investment

$5000/acre; all land assumed pur-
 chased initially; sludge depth,  30 ft

Total depreciation in 30-yr; straight-
 line basis
Within one mile of the plant
Cost of forced oxidation and disposal of gysum sludge converted to
cost/ton  of equivalent quantity of nonoxidized sludge.  Divided by
1.08 to convert to gypsum cost.  Includes fly ash; disposal is in an
indigenous clay pond.
                          151

-------
                              REFERENCES
1.     R.  B. Fling etal..  Disposal of Flue Gas Cleaning Wastes:  EPA
      Shawnee Field Evaluation:  Second Annual Report7 EPA-600/7-78/024,
      U.S. Environmental Protection Agency, Research Triangle Park,
      North Carolina,  Feb 1978.

2.     R.  B. Fling _et_al_. ,  Disposal of Flue Gas Cleaning Wastes;  EPA
      Shawnee Field Evaluation:  Initial Report, EPA-600 /2-76-070, U.S.
      Environmental Protection Agency, Research Triangle Park, North
      Carolina,  March 1976.

3.     J.  Rossoff and R. C.  Rossi,  Disposal of By-Products from Non-
      regenerable Flue Gas Desulfurization Systems;  Initial Report,
      EPA-650/2-74-037a,  U.S.  Environmental Protection Agency,
      Research  Triangle Park, North Carolina, May 1974.

4.     J.  Rossoff et al. , Disposal of By-Products  from Nonregenerable  Flue
      Gas DesulfurTzation Systems: Second Progress Report, EPA-6QO-7-
      77-052,  U.S. Environmental Protection Agency, Research Triangle
      Park, North Carolina, May 1977.

5.     J.  Rossoff and R. C.  Rossi,  "Flue Gas Cleaning Waste Disposal, EPA
      Shawnee Field Evaluation,  "presented at the EPA Flue Gas Desulfuriza-
      tion Symposium, New Orleans, Louisiana, March 1976.

6.     J.  Rossoff etal. , "Disposal  of By-Products from Non-Regenerable
      Flue Gas Desulfurization Systems:  A Status Report, " presented at
      the EPA Flue Gas Desulfurization Symposium, Atlanta,  Georgia,
      Nov 4-7,  1974.

7.     P.  P. Leo,  R.  B.  Fling, and J. Rossoff, "Flue Gas Desulfurization
      Waste Disposal Field Study at the Shawnee Power Station, " presented
      at  the EPA Symposium on Flue Gas Desulfurization, Hollywood, Florida,
      Nov 8-11, 1977.

8,     P.  P. Leo and J. Rossoff, Controlling SO?  Emissions from Coal-Fired
      Stream Electric Generators;  Solid Waste Impact, EPA-600/7-78-044b,
      Vol.  II,  U.S. Environmental Protection Agency, Research Triangle
      Park, North Carolina, March 1978.
                               152

-------
        COMPARATIVE ECONOMICS OF FGD WASTE DISPOSAL
                       J.  W.  Barrier
           Emission Control Development  Projects
      Office  of  Agricultural  and  Chemical  Development
                Tennessee  Valley  Authority
                 Muscle Shoals,  Alabama
              Prepared for Presentation at
              Industry Briefing Conference
Results of EPA Lime/Limestone Wet Scrubbing Test Programs
  Sponsored by the U.S.  Environmental Protection Agency
     Royal Villa Motel in Raleigh, North Carolina
                    August 29,  1978
                           153

-------
             COMPARATIVE ECONOMICS OF FGD WASTE DISPOSAL
                            J. W. Barrier
                Emission Control Development Projects
           Office of Agricultural and  Chemical Development
                     Tennessee Valley  Authority
                       Muscle Shoals,  Alabama
                               ABSTRACT
Several series of studies to evaluate the economics of various systems
associated with the control of fly ash and sulfur dioxide emissions from
power plant flue gases are being conducted by the Tennessee Valley
Authority (TVA) for the U.S. Environmental Protection Agency  (EPA).  One
group of studies involves the preparation of economics for the comparison
of flue gas desulfurization (FGD) sludge disposal alternatives.  Two
studies are complete—one report is published and one report  is being
reviewed by EPA before publication—and a third study is underway.  The
results of the two completed studies are described in this report.

Six disposal alternatives have been evaluated to date.  A base case for
each process was established and complete conceptual designs of the
systems were prepared for use as a cost estimating basis.  Cost estimates
and conceptual designs are based on common premises used for all TVA-EPA
studies.

The six alternatives evaluated are (1) untreated ponding, (2) Dravo
Corporation's process, (3) Chemfix process, (4) IU Conversion Systems'
process, (5) untreated sludge - fly ash blending, and (6) gypsum landfill.
For each alternative total capital investments and annual revenue require-
ments were estimated for the base case and major case variations.
                               154

-------
             COMPARATIVE ECONOMICS OF FGD WASTE DISPOSAL
INTRODUCTION

The U.S. Environmental Protection Agency (EPA) is sponsoring an exten-
sive research and development program to evaluate, develop, and demonstrate
sludge disposal alternatives that are environmentally and economically
acceptable to the utility industry for flue gas desulfurization (FGD)
sludge  (1).  A major program area that involves the field testing of
potential processes for commercial-scale use is The Aerospace Corporation's
study being conducted at the Shawnee power plant of the Tennessee Valley
Authority (TVA).   All of the alternatives evaluated at Shawnee are also
considered in the TVA economic studies for sludge disposal options.

Two general categories of FGD processes are available for use by the
utility industry:  nonregenerable or throwaway processes which produce a
waste material for disposal and regenerable or recovery processes that
produce a saleable byproduct.  Many processes are available in both cate-
gories; however,  most utilities are selecting the lime or limestone
process which produces a throwaway sludge (2).  Two categories of waste
disposal processes are being used:  wet and dry.  Wet processes normally
involve pond disposal and dry processes usually involve landfill of
sludge  (3,4).  The alternatives evaluated by TVA are representative of a
range of disposal options and include both wet and dry disposal processes.

The six alternatives evaluated are (1) untreated ponding, (2) Dravo
Corporation's process, (3) Chemfix process, (4) IU Conversion Systems, Inc.,
(IUCS) process, (5) untreated sludge - fly ash blending, and (6) gypsum
disposal.  For each process considered, a base case was established and
definitive estimates of total capital investments and total annual revenue
requirements were calculated.  The estimates were all made using a set of
carefully defined common premises and are directly comparable.  Cost esti-
mates are based on process background information, flowsheets, material
and energy balances, equipment and system requirements, and raw material,
labor, and utility costs.  All estimates of capital investment are
projected to mid-1979 and revenue requirements to mid-1980.
                                   155

-------
BACKGROUND AND DESCRIPTION OF ALTERNATIVES

Many power plants with nonregenerable FGD systems that are now in opera-
tion in the United States use a sludge disposal method involving some
form of onsite ponding or impoundment of untreated material.  This method
of disposal, although popular, will not necessarily be the best option
for future installations.  Drawbacks, such as potential disposal regula-
tions and limited land availability, have made necessary the development
of other disposal options for FGD wastes (1).  Several other treatment
options involving chemical and physical stabilization are available to
the utility industry.

The technology associated with the six disposal alternatives evaluated in
the TVA-EPA work and discussed in this paper is assumed to be proven, but
in many cases is in the development stage and is not actually proven in
full-scale application (i.e., forced oxidation, simultaneous sulfur
dioxide (S02) and fly ash removal, etc.).  The primary emphasis of the
work was to evaluate the economics of the disposal alternatives rather
than the process technology  (5).

Untreated Ponding

As stated earlier, the untreated ponding option is the alternative
selected most often by the utility industry.  Effluent from the scrubber
system is pumped directly to a pond and allowed to settle.  Excess water
is recycled to the scrubber  system.  Very few items of equipment are
required if this option is used, but the capital investment for the dis-
posal pond is very high (7).

Dravo Process

Dravo offers two basic processes for FGD sludge disposal  (pond or land-
fill).  Although the pond or impoundment aJternative was the base case
for TVA studies, the more recently promoted  landfill process may be more
economically attractive.  Effluent from the  scrubber system is partially
dewatered using a thickener  before mixing with Dravo"s fixation additives
(Thiosorbic lime and Calcilox).  The treated material is then pumped to
an impoundment area where the material settled and is eventually stabi-
lized.  Excess water is recycled to the scrubber system.  Dravo's fixation
agents and their entire fixation process are patented (Synearth process)
(6).

IUCS Process

The IUCS system is called the Poz-0-Tec process and involves chemical
stabilization of calcium-based waste materials by mixing with lime and
fly ash.  Scrubber system effluent is dewatered using a thickener and
rotary drum filter.  The dewatered material  (containing about 60£ solids)
is then mixed with fixation  additives (lime  and fly ash) and trucked to a
disposal site for landfill disposal.  Fly ash is a necessary ingredient
for stabilization and can be blended with the sludge cake at the additive
                                 156

-------
mixing stage of processing or included with the sludge following removal
simultaneously with the S02 in the preceding and scrubbing stages of the
FGD process.  IUCS reports that the stabilized material is claylike and
can be easily handled, transported by truck, placed, and compacted, and
that the landfill is structurally suitable for future reclamation (7).

Chemfix Process

Chemfix has been applying their technology for sludge stabilization to
wastes generated by metal finishing, automotive assembly, and electronics
operations for several years (8).  Chemfix offers a process that yields
a treated stabilized sludge that is reported to be suitable for landfill
disposal.  Effluent from the scrubber system is dewatered using a thickener
and rotary drum filter.  The filter cake is mixed with two chemical addi-
tives (Portland cement and sodium silicate), during which stabilization of
the sludge is achieved.

Untreated Sludge - Fly Ash Blending

Many power plants are meeting particulate emission requirements for fly
ash by installing equipment for dry fly ash collection.  Dry fly ash can
be used in many cases as an additive for blending with dewatered scrubber
sludge to yield a physically stable material.  This process of sludge
treatment would allow the utility to dispose of both fly ash and scrubber
wastes in one operation and also to produce a waste product that is
suitable for landfill disposal.

Effluent from the scrubber system is dewatered using a thickener and
rotary drum filters.  The filter cake is mixed with dry fly ash which is
pneumatically conveyed from the fly ash collection system to the sludge
disposal facility.  The blended material Ls transported by truck to a
landfill disposal site.  TVA studies inui  ite that this procedure can be
used to produce a product suitable for I;1  Mill disposal and that handling
with trucks and earthmoving equipment is :   .?ible (9).
The lime and limestone FCiJ processes can be modi I Led  m  include a
processing step to force i he oxidation of calcium sulfite  sludge  (the
normal product ol these processes) to gypsum.  Gypsum is a more desirable
waste product because of improved settling properties (settling rate is
about 10 times greater than CaS03) and therefore a  reduced volume of
material can be attained through dewatering.  The landfill disposal of
gypsum can be accomplished without the use of blending or  mixing  equip-
ment and fixation additives.  Underflow from the scrubber  system  is
dewatered, using a thickener and rotary drum filter,  before  it is hauled
by truck to a landfill disposal site.  Tests conducted at  Shawnee power
plant indicate Lhat SO^ .:iul fly ash can be removed  simultaneously in the
scrubber system and therefore the equipment for  dry fly  ash  collection
is not needed (10,11).
                                 157

-------
EVALUATION OF ALTERNATIVES

A comparative economic evaluation of several processes requires that the
basis for the capital investment and revenue requirement estimates be the
same.  All TVA studies are made using a predetermined set of design and
economic premises for the power plant, fuel, FGD system, and estimate
calculation procedures.  These premises that allow the comparison of
estimates are summarized in the following paragraphs.

Base Case Design Premises

Power Plant—

   1.  The plant is newly constructed and has a 30-year life.

   2.  The single coal-fired unit has an output of 500 MW.

   3.  The total operating life is 127,500 hours with an average
       annual capacity of 4,250 hours.

   4.  The power unit heat input requirement is 9,000 Btu/kWh.

   5.  The coal heating value is 10,500 Btu/lb.

   6.  The coal contains 3.5% (by wt) sulfur (dry) and 16%  (by wt) ash.

FGD System—

   1.  A limestone scrubbing process is used for S02 removal.

   2.  S02 and fly ash are removed to meet NSPS.   [EPA issued Federal
       Standards of Performance for New Stationary Sources  (often called
       "new source performance standards" or NSPS).]  The allowable S02
       emission is 1.2 Ib/MBtu heat input and the particulate emission,
       0.1 Ib/MBtu heat input.

   3.  Eighty-five percent of the ash present in the coal is emitted as
       fly ash.

   4.  Ninety-five percent of the sulfur in the coal is emitted as S02.

   5.  Effluent from the scrubber system contains 15% solids.

   6.  All storage facilities have a 30-day capacity and feed bins,
       intermediate storage tanks, etc., have an 8-hour capacity.

Untreated Ponding—

   1.  Effluent (15% solids) from the scrubber system is pumped to a
       clay-lined disposal pond.
                                  158

-------
   2.  The pond is located 1 mile from the scrubber facilities.

   3.  The sludge settles to 50% solids in the pond and excess water
       is recycled to the scrubber system.

   4.  The FGD process stoichiometry is 1.5 raols calcium oxide per mol
       S02 removed.

   5.  Fly ash and S02 are removed simultaneously in the scrubber system;
       therefore the sludge contains both fly ash and calcium wastes.

   6.  Fifteen percent of the S02 removed is converted to gypsum and the
       remaining 85% calcium sulfite.

Dravo Process—

   1.  Thickened sludge (35% solids) is treated with Dravo additives:
       Calcilox (7% of dry solids) and Thiosorbic lime (1% of dry solids).

   2.  Treated sludge is pumped 1 mile to a clay-lined pond for disposal.

   3.  Stabilization as a soillike material occurs over a 2- to 4-week
       period.  Fixed sludge is 50% solids and excess water is recycled
       to the scrubber system.

   4.  The FGD process stoichiometry is 1.5 mols calcium oxide per mol
       S02 removed.

   5.  Fly ash and S02 are removed simultaneously in the scrubber system;
       therefore the sludge contains both fly ash and calcium wastes.

   6.  Fifteen percent of the S02 removed is converted to gypsum and the
       remaining 85% calcium sulfite.

IUCS Process—

   1.  Dewatered sludge (60% solids) is treated with lime (4% of dry solids).

   2.  Trucks are used to transport the treated material to a landfill
       disposal site located 1 mile from the scrubber facilities.

   3.  Treated sludge is assumed to have claylike properties and can be
       placed and compacted in a landfill with typical earthmoving equip-
       ment.

   4.  The FGD process stoichiometry is 1.5 mols calcium oxide per mol S02
       removed.

   5.  Fly ash and S02 are removed simultaneously in the scrubber system;
       therefore the sludge contains both flyash and calcium wastes.

   6.  Fifteen percent of S02 removed is converted to gypsum and the
       remaining 85% calcium sulfite.

                                 159

-------
Chemfix Process—

   1.  Thickened sludge (35% solids) is transported by pipeline (1 mile)
       to the disposal site where additional dewatering, mixing with
       fixation additives, and landfill placement occurs.

   2.  Dewatered sludge (60% solids) is stabilized by mixing with two
       Chemfix additives:  Portland cement (7% of dry solids) and sodium
       silicate (2% of dry solids).

   3.  Treated material is placed and compacted as landfill using
       typical earthmoving equipment.

   4.  The FGD process stoichiometry is 1.5 mols calcium oxide per mol
       S02 removed.

   5.  Fly ash and S02 are removed simultaneously in the scrubber system;
       therefore the sludge contains both fly ash and calcium wastes.

   6.  Fifteen percent of the S02 removed is converted to gypsum and the
       remaining 85% calcium sulfite.

Untreated Sludge - Fly Ash Blending—

   1.  Dewatered sludge (60% solids) is blended with dry fly ash to
       yield a physically stable material.

   2.  Fly ash is removed from the flue gas to meet NSPS using an
       electrostatic precipitator  (ESP) and pneumatically conveyed to
       the sludge treatment area for blending with sludge.

   3.  The blended material (about 75% solids) is transported to a
       landfill disposal  site by truck (1 in Lie).

   4.  Typical carthmovinR equipment is used for placement  and compac-
       tion  in a landfill.

   5.  The FGD process stoichiometry is 1.5 mols calcium oxide per mol
       S02 removed.

   6.  Fifteen percent of the S02  removed is converted  to gypsum and
       the remaining 85%  calcium sulfite.

Gypsum—

   1.  The limestone FGD  process is modified to  provide  forced oxidation
       of calcium sulfite sludge to gypsum.  The FGD  process stoichi-
       ometry  is 1.1 mols calcium  oxide per mol  S02 removed.

   2.  Ninety-five percent of the  S02  removed  is converted  to gypsum
       and the remaining  5% calcium sulfite.
                                  160

-------
   3.  Fly ash and S02 are removed simultaneously in the scrubber loop
       to meet NSPS;  therefore,  sludge contains both gypsum and fly ash.

   4.  Dewatered gypsum (about 80% solids) is transported by truck
       (1 mile) to the landfill  disposal  site.

   5.  Typical earthmoving equipment is used for placement and compaction
       of the gypsum in the landfill.

Economic Premises

A midwestern plant location was  selected because of coal availability
for the large number of coal-fired plants in this region.  Other economic
assumptions are summarized as follows:

   1.  All capital cost estimates are based on Chemical Engineering cost
       indices (labor index - 237.9, material index - 264.9).  Capital
       costs are project-i to mid-1979 using these indices.  Construction
       on the project is assumed to have started in mid-1977 and to be
       completed in mid-1980.

   2.  Direct capital costs cover process equipment, piping and insula-
       tion, transport lines, foundations and structural, excavation and
       site preparation, roads and railroads, electrical instrumentation,
       buildings, and trucks and earthmoving equipment.  Material and
       labor (fabrication and installation) costs for each of these items
       were estimated.  These estimates are based on costs obtained from
       vendors and on related literature information.

   3.  Indirect capital costs include engineering design and supervision,
       architect and  engineering contractor expenses, construction
       expenses, contractor fees, contingency,  allowance for startup and
       modifications, and interest during construction.  Two other capital
       costs not included as indirect costs, but in the total capital
       investment, are working capital and land.  These estimates are
       based on current industry practice and authoritative literature
       sources.

   4.  Direct costs for revenue  requirements include raw materials, labor,
       electricity, equipment fuel and maintenance, and analyses.  These
       costs are projected to mid-1980.

   5.  Indirect costs for revenue requirements are capital charges and
       overheads.

   6.  Capital charges are based on regulated utility economics.

   7.  Revenue requirements are  projected for an annual 7000 hr/yr (first
       year) operation.  Other estimates are made for lifetime revenue
       requirements that are based on the declining operating profile
       of the plant.
                                 161

-------
 Case  Variations in  Design  Premises

 The base  case  design  premises  were  altered  for  selected variables  in
 order to  evaluate the effects  of  changes  in operating conditions and
 site-specific  design  factors.   Several  of the variations which were
 considered  are as follows:

    1.   Plant size:  200  and  1500  MW (the  1500-MW plant is assumed  to
        be three 500-MW units).

    2.   Coal composition:   Sulfur  content, 2.0%  and  5.0%; ash content,
        12%  and 20%.

    3.   Remaining life of an  existing  plant:  25,  20, and 15 years.

    4.   Distance to  disposal:   5 and 10  miles.

    5.   Availability of land  for disposal  site.   Construction:  50% and
        75%  of  optimum.
RESULTS

Two TVA-EPA studies to evaluate the economics of six FGD sludge disposal
alternatives are complete.  The capital investments and revenue require-
ments of the base cases and major case variations for the six options
are discussed in this paper.  Additional details concerning the cost
estimates can be obtained by reviewing the two TVA-EPA reports (1)
(one of the two reports is not yet published, but details are available
from the author of this paper).

Total System Costs

The total cost of S02 and particulate emission control can be obtained
by combining the cost estimates of the FGD system with waste disposal
system costs.  Estimates of FGD system costs (total capital investment
and annual revenue requirements) are available from other TVA-EPA
studies (5,13) and are suitable for combining with the waste disposal
system costs discussed in this report.  These costs are summarized in
Tables 1 and 2.  The FGD costs presented in this paper apply only to the
base case conditions and therefore cannot be used with waste disposal
systems other than the base cases.

Unit Revenue Requirements

Unit revenue requirements for the base case system and several major
case variations are shown in Table 3.
                                162

-------
                                  TABLE 1.  SUMMARY OF CAPITAL INVESTMENTS FOR

                                    COMBINED FGD AND SLUDGE DISPOSAL SYSTEMS
                      Disposal process'
                                                            Total capital investments
                   Variation
   Site
description
  FGD
 system
         Disposal
          system
k$
$/kW
k$
$/kW
                Combined
                 system
k$
u>
               Untreated             Ponding
               Dravo                 Ponding
               Dravo                 Landfill
               Chemfix               Landfill
               IUCS                  Landfill
               Gypsum                Landfill
               Untreated sludge -    Landfill
                fly ash blending
$/kW
              36,368C    72.8   17,211   34.4   53,579   107.2
              36,368C    72.8   24,114   48.2   60,482   121.0
              36,368C    72.8   12,670   25.3   49,038    98.1
              36,368C    72.8   13,531   27.2   49,899    99.8
              36,368C    72.8   10,717   21.4   47,085    94.2
              38,671c'd  77.3    5,411   10.7   44,082    88.2
              45,982e    92.0    8,605   17.2   ^4,587   109.2
               a.   Dewatering equipment for all cases included in the disposal system.
               b.   The amounts shown are for the base case (mid-1979 costs).
               c.   Costs are for an FGD system which removes both S02 and fly ash in the scrubber
                   loop.
               d.   Cost includes additional equipment required for forced oxidation ($2,300,000).
               e.   An electrostatic precipitator (ESP) is used to remove fly  ash and its installed
                   cost is  included ($9,614,000).

-------
TABLE 2.  SUMMARY OF REVENUE REQUIREMENTS FOR




  COMBINED FGD AND SLUDGE DISPOSAL SYSTEMS

Revenue requirements3
Disposal process


Variation
Untreated
Dravo
Dravo
Chemf ix
IUCS
Gypsum
Untreated
fly ash






sludge -
blending
Site
description
Ponding
Ponding
Landfill
Landfill
Landfill
Landfill
Landfill

FGD system
Total
annual $
ll,841,500b
ll,841,500b
ll,841,500b
ll,841,500b
ll,841,500b
12,846,800b
13,816,500d

Mills/
kWh
3.38
3.38
3.38
3.38
3.38
>c 3.67
3.94


Total
annual
3,280,
6,701,
6,620,
6,988,
5,291,
3,117,
3,735,

Disposal system

$
000
000
000
000
000
500
000

Mills/
kWh
0.94
1.91
1.89
2.00
1.51
0.89
1.07

$/ton
dry solids
8.08
15.32
15.16
16.51
12.55
7.86
9.20

Combined systems
Total
annual
15,121,
18,542,
18,461,
18,829,
17,132,
15,964,
17,551,


$
500
500
500
500
500
300
500

Mills/
kWh
4.32
5.30
5.27
5.38
4.90
4.56
5.01


a. The amounts shown are for the
b. Costs
c. Cost
are for
includes
d. An ESP is used
base case (mid-1980 costs).
an FGD system which removes
that associated
to remove fly
with forced
both S02 and fly ash in
oxidation
ash and its associated
equipment
($1
the scrubber loop.
,005,300).

operating costs are included ($1,975

,000).





-------
                               TABLE 3.   UNIT REVENUE  REQUIREMENTS  - ALL PROCESSES
Disposal process
Untreated

Base case*3
Variation from base case
200 MW
1500 MW
Existing, 25-year life
Existing, 20-year life
Existing, 15-year life
12% ash in coal
20% ash in coal
2% sulfur in coal
5% sulfur in coal
5 miles to disposal
10 miles to disposal
Constrained acreage
(50% of optimum)
Constrained acreage
(75% of optimum)
Mills/
kWh
0.94

1.44
0.64
0.55
0.45
0.38
0.83
1.03
0.75
1.10
1.58
2.14
1.18

0.96

$/dry
ton
8.08

12.12
5.55
4.69
3.80
3.19
8.68
7.44
9.37
7.35
13.61
18.48
10.15

8.29

Dravo
Mills/
kWh
1.91

2.60
1.36
1.32
1.21
1.16
1.69
2.12
1.52
2.29
2.32
2.67
2.60

2.25

$/dry
ton
15.32

20.41
10.87
10.30
9.50
9.04
16.43
15.58
17.45
14.08
18.57
21.39
20.82

18.06

IUCS
Mills/
kWh
1.51

2.55
0.99
1.01
1.02
1.04
1.30
1.71
1.33
1.77
1.85
2.14




S/dry
ton
12.55

20.68
8.23
8.24
8.26
8.43
13.05
11.84
15.57
11.29
15.40
17.73




Chemf ix
Mills/
kWh
2.00

3.24
1.37
1.40
1.41
1.43
1.78
2.17
1.70
2.36
2.48
2.86




S/dry
ton
16.51

26.14
11.31
11.36
11.39
11.59
17.86
15.00
20.19
15.01
20.49
23.63




Untreated sludge -
fly ash blending
Mills/
kWh
1.07

1.96
0.65
1.07
1.06
1.06
1.02
1.11
0.91
1.20
1.25
1.39




S/dry
ton
9.20

16.51
5.64
9.01
8.97
8.94
10.77
8.03
11.26
7.88
10.81
11.96




Gypsum
Mills/
kWh
0.89

1.79
0.47
0.88
0.88
0.88
0.86
0.92
0.77
0.93
1.06
1.22




S/dry
ton
7.86

15.42
4.17
7.63
7.62
7.61
9.23
6.75
9.74
6.45
9.37
10.80





a.   Basis
      Midwest plant location, mid-1980 costs; 7,000 hr/yr plant on-stream time;  S02  and fly ash removed to meet NSPS.
b.   Base case
      New 500-MW plant with 30-year life.

-------
 Sludge Disposal System Costs

 Major case variations and their effects on costs are discussed in the
 following section of the paper.  The costs shown in these tables represent
 only the costs associated with the sludge disposal area.

 Power Plant Size—

 The power plant size has an almost direct effect on sludge disposal
 costs.   A slight economy of scale is seen for the plant sizes evaluated.
 Table 4 is a summary of sludge disposal process costs for the alternatives
 evaluated.

 Coal Composition—

 The sulfur and ash contents of the coal also have a direct effect on
 the quantity of sludge for disposal.   Cost estimates were made for
 variable sulfur and ash percentages of the coal.   These estimates are
 summarized in Table 5.

 Remaining  Plant Life—

 In many cases existing power  plants (5-15 years old)  are required to
 install FGD systems to meet emission  regulations.   Several cost  estimates
 were made  to evaluate  the  sludge  disposal costs for plants with  remaining
 operating  times of  less than  30 years (15,  20,  and  25 years).  Capital
 investments for these  cases were  considerably less  if the disposal
 alternative involved ponding.   Unit revenue requirements were  increased
 because the depreciation of capital was  taken over  a  shorter period  of
 time.   Table 6 summarizes  the  remaining  life case variation estimates.

 Distance  to Disposal Site—

 Case variations were considered to  determine the effect  of the distance
 to the  waste disposal  site on  capital investment and  revenue require-
 ments.   These  results are  summarized  in  Table 7.  The  capital  investment
 and  revenue  requirements increase rapidly with  the  increasing  distance
 to disposal  for alternatives using  pipelines for slurry  transport.   Costs
 increase for alternatives  for  using trucks  for  transport,  but  not  as much
 as  the  pipeline transport alternatives.

Availability of  Land—

The  quantity of  land available  for construction of a disposal pond for
untreated sludge can be a significant factor in selecting a disposal
alternative.   Several cost estimates were made to evaluate the effect of
land availability on costs.  Estimates are normally made in TVA studies
by determining  the minimum total pond cost by optimizing between land
cost and construction costs.  The quantity of land is therefore the
amount that should be used to obtain the lowest overall pond cost.
                                 166

-------
       TABLE 4.  TOTAL CAPITAL INVESTMENTS AND ANNUAL REVENUE REQUIREMENTS

                          FOR PLANT SIZE CASE VARIATIONS


                    Total capital investment, k$a   Annual revenue requirements, k$
Disposal process
Untreated
Dravo
IUCS
Chemfix
Untreated sludge -
fly ash blending
Gypsum
Power
200b
9,800
13,942
7,193
9,259
6,126

3,988
plant size, MW
500C
17,211
24,114
10,717
13,531
8,605

5,411
1500b
36,455
48,235
20,105
24,104
18,282

9,826
Power
200b
2,014
3,643
3,567
4,529
2,742

2,502
pj.ant sizet MW
500°
3,280
6,701
5,291
6,988
3,735

3,118
1500D
6,746
14,264
10,411
14,362
6,867

4,961

c.
    New plant with 30-year life; Midwest plant location; mid-1979 capital costs;
    mid-1980 revenue requirements; 7,000 hr/yr on-stream time; coal analyses
    (by wt):  3.5% sulfur (dry basis), 16% ash; fly ash and S02 removed to meet
    NSPS; 1 mile to disposal site.
    Base case premises except plant size.
    Base case.

-------
                             TABLE 5.  TOTAL CAPITAL INVESTMENT AND ANNUAL REVENUE

                                REQUIREMENTS FOR COAL COMPOSITION CASE VARIATIONS
oo
                                                                     Annual revenue requirement. k$
Sulfur in coal, %
Disposal process
Untreated
Dravo
IUCS
Chemf ix
Untreated sludge -
fly ash blending
Gypsum
2b
13,390
19,251
9,345
11,879
7,356

4,782
5C
20,655
28,523
11,957
14,192
9,534

5,884
Ash in
12^
15,031
21,466
9,025
11,123
7,917

5,042
coal, %
20e
19,055
26,028
12,283
14,854
9,309

5,707
Sulfur in coal, %
2b
2,639
5,314
4,654
5,935
3,186

2,707
5C
3,869
8,007
6,118
8,263
4,199

3,252
Ash in
12d
2,902
5,924
4,533
6,229
3,581

3,018
coal, %
20e
3,609
7,406
5,971
7,600
3,896

3,206
           a.  New plant with  30-year  life; Midwest plant  location; mid-1980  operating costs; mid-1979
              capital  costs;  7,000  hr/yr  on-stream time;  fly ash  and  S0a  removed  to meet  NSPS; 1 mile
              to  disposal  site.
           b.  Base case premises  except percent  sulfur  in coal  and coal  heating value  (10,700 Btu/lb).
           c.  Base case premises  except percent  sulfur  in coal  and coal  heating value  (10,400 Btu/lb),
           d.  Base case premises  except percent  ash  in  coal and coal  heating value  (11,100  Btu/lb).
           e.  Base case premises  except percent  ash  in  coal and coal  heating value  (9,900 Btu/lb).

-------
         TABLE 6.  TOTAL CAPITAL INVESTMENT AND ANNUAL REVENUE REQUIREMENT

                    FOR REMAINING POWER PLANT LIFE CASE VARIATIONS
Total capital investment,
, k$a
Remaining power plant life, jrear
Disposal process
Untreated
Dravo
IUCS
Chemf ix
Untreated sludge -
fly ash blending
Gypsum

17
24
10
13
8

5
3Qb
,211
,114
,717
,531
,605

,411

14
21
10
13
8

5
25c
,578
,416
,591
,400
,528

,174

11
18
10
13
8

5
20C
,399
,281
,402
,204
,381

,115
15C
&, 822
15,553
10,269
13,077
8,276

5,076
Annual
revenue
Remaining power
30b
3,280
6,701
5,291
6,988
3,735

3,118
25C
2,906
6,377
5,402
7,152
3,739

3,097
requirements , k$a
plant life, year
20C
2,135
5,941
5,430
7,191
3,724

3,091
15C
2,130
5,728
5,559
7,359
3,712

3,087

a.  Midwest plant location; mid-1979 capital costs;  mid-1980 revenue requirements;
    7,000 hr/yr on-stream time; fly ash and S02 removed to meet NSPS;  1 mile to dis-
    posal site.
b.  Base case.
c.  Same as base case except remaining plant life and boiler heat rate (9,200 Btu/kWh).

-------
•-J
o
                       TABLE 7.  TOTAL CAPITAL  INVESTMENT AND ANNUAL REVENUE REQUIREMENTS

                                  FOR DISTANCE  TO DISPOSAL  SITE  CASE VARIATIONS

Q
Total capital investment, k$
Distance to disposal site, mile
Disposal process
Untreated
Dravo
IUCS
Chemfix
Untreated sludge -
fly ash blending
Gypsum
lb
17,211
24,114
10,717
13,531
8,605

5,411
5C
26,836
30,994
11,377
18,313
8,969

5,750
10C
37,420
37,765
11,891
20,227
9,334

6,514
rt
Annual revenue requirement, k$
Distance to disposal site, mile
lb
3,280
6,701
5,291
6,988
3,735

3,118
5C
5,527
8,124
6,490
8,675
4,389

3,719
10C
7,504
9,360
7,475
10,003
4,855

4,286
               a.   New plant with  30-year  life; Midwest  plant  location; mid-1979 capital costs; mid-
                   1980 revenue  requirements;  7,000  hr/yr  on-stream  time;  fly ash and S02 removed to
                   meet NSPS.
               b.   Base case.
               c.   Same as  base  case  except  distance to  disposal  site.

-------
Estimates shown in Table 8 are for systems with disposal ponds constructed
on a less than optimum acreage.  Although total land costs are less for
these cases, pond construction costs are much higher than for the optimum
ponds.

Other Variations—

Several other variations from the base case design and economic premises
were considered in TVA sludge studies.  Since these case variations had
a lesser effect on the costs than the variations discussed, the results
are not included in this paper.

Lifetime Revenue Requirements

Estimates of the total revenue requirements of waste disposal processes
over the 30-year system life were estimated.  These costs, as shown in
Table 9, are cumulative over the 30-year plant life.


CONCLUSIONS

Several conclusions can be derived from the results generated by the
TVA-EPA sludge disposal economic studies.

   1.  The base case sludge disposal system requiring the lowest capital
       investment and annual revenue requirement was gypsum disposal.
       This alternative requires a much smaller investment for equipment
       than any other alternative except untreated ponding which requires
       a very expensive disposal pond.  The selection of this alternative
       would require that a typical limestone FGD system be modified to
       include the forced oxidation of sulfite (S03) compounds to gypsum.
       This requires an additional capital investment of $2,300,000.

   2.  In all case variations,  the gypsum process had the lowest total
       capital investment and annual revenue requirements.

   3.  The alternatives involving pond disposal (untreated and Dravo)
       required the highest capital investments.  All other processes
       were for landfill disposal.

   4.  The three processes involving chemical treatment (Dravo, IUCS,
       and Chemfix) all had higher  annual revenue requirements than the
       three processes involving no chemical treatment.

   5.  Both unit capital investment and unit revenue requirements were
       slightly lower for large plant size.

   6.  Capital requirements and revenue requirements vary almost directly
       in proportion to the quantity of sludge for disposal.   A slight
       economy of  scale is seen.   Cases involving coal, ash,  and sulfur
       content variations are examples of this effect.
                                171

-------
                TABLE  8.   TOTAL CAPITAL INVESTMENTS AND

           ANNUAL REVENUE REQUIREMENTS  FOR  LAND AVAILABILITY

            CASE VARIATIONS  FOR UNTREATED PONDING  DISPOSAL
Case
variation
Optimum landc
75% optimum land"
50% optimum landd
Land
requirement,
acre
407
305
204
Total capital
investments, k$
17,211
17,985
22,676
Annual revenue
requirements , k$
3,280
3,365
4,119
b


a.  New 500-MW plant with 30-year life;  Midwest plant location;  mid-
    1979 costs; fly ash and S02 removed  to meet NSPS; 1 mile to
    disposal site.
b.  Same as footnote "a" except costs are mid-1980.
c.  Base case for untreated disposal option.
d.  Same as base case except acreage and cost of disposal pond.
                                 172

-------
                       TABLE 9.   SUMMARY  OF LIFETIME  REVENUE  REQUIREMENTS  FOR  ALL  PROCESSES*
10
       Disposal process
                   Actual cumulative
                       lifetime
                revenue requirements, $
Lifetime average
  unit revenue
  requirements,
    mills/kWh
                                                                           Discounted
                                                                           cumulative
                                                                         lifetime revenue
Levelized unit
   revenue
 requirements,
  mills/kWhc
Untreated
Dravo
IUCS
Chemfix
Untreated sludge -
fly ash blending
Gypsum
97,757,800
175,764,900
131,224,200
167,942,300
96,526,800

78,072,400
1.53
2.76
2.06
2.63
1.51

1.22
33,612,100
62,052,600
45,381,700
59,099,300
32,801,900

216,513,400
1.35
2.50
1.83
2.38
1.32

1.07
Basis
  New plant with 30-year life; Midwest plant location; mid-1980 costs; fly ash and S02
  removed to meet NSPS; operating profile:  7,000 hr/yr for 10 years, 5,000 hr/yr for 5 years,
  3,500 hr/yr for 5 years, 1,500 hr/yr for 10 years;  coal analysis (wt %) - 3.5% sulfur (dry),
  16% ash.
Discounted to initial year at 10%.
Equivalent to discounted process cost over life of power plant.

-------
   7.   The remaining life of  a power plant has a significant effect on
       the relative ranking of capital investments for the two alterna-
       tives involving pond disposal.   As the plant life is reduced,
       these alternatives become more favorable.

   8.   The distance to the disposal site greatly increases the capital
       investments for the untreated,  Dravo, and Chemfix alternatives.
       These increases are primarily due to the additional costs for
       pumps and pipelines (other alternatives involve truck trans-
       portation to disposal site).

   9.   Case variations for disposal of untreated sludge in ponds con-
       structed on less than the optimum acreage have higher total
       capital investments than the base (optimum acreage) case.  These
       variations illustrate the potential problems for plants with a
       limited quantity of land available for pond construction.

  10.   Alternatives involving truck transport and landfill disposal
       generally had higher revenue requirements, but lower capital
       investments than the alternatives involving pipeline transport
       and pond disposal.

The results presented in this paper do not take into account site-specific
waste disposal conditions that a utility may encounter when selecting a
system for installation.  Results are based only on predetermined design
and economic premises and should not be interpreted to represent a sLte-
specific disposal situation.
                                  174

-------
REFERENCES

 1.  Jones, J. W.  Research and Development for Control of Waste and
     Water Pollution from Flue Gas Cleaning Systems.  In:  Proceedings
     of Symposium on Flue Gas Desulfurization, Vol. II, New Orleans,
     Louisiana, March 8-11, 1976.  EPA-600/2-76-136b (NTIS PB 262 722),
     May 1976.  pp. 579-604.

 2.  Crowe, J. L., and H. W. Elder.  Status and Plans for Waste Disposal
     from Utility Applications of Flue Gas Desulfurization Systems.  In:
     Proceedings of Symposium on Flue Gas Desulfurization, Vol. II,
     New Orleans, Louisiana, March 8-11, 1976.  EPA-600/2-76-136b  (NTIS
     PB 262 722), May 1976.  pp. 565-577.

 3.  Fling, R. B., W. M. Graven, F. D. Hess, P. P. Leo, R. C. Rossi,
     and J. Rossoff.  Disposal of Flue Gas Cleaning Wastes:  EPA Shawnee
     Field Evaluation -  Initial Report.  EPA-600/2-76-070  (NTIS PB
     251 876), March 1976.  221 pp.

 4.  Leo, P.  P., and J.  Rossoff.  Control of Waste and Water Pollution
     from Power  Plant Flue  Gas Cleaning  Systems:  First Annual R and D
     Report.   EPA-600/7-76-018  (NTIS PB  259 211), October  1976.

 5.  Barrier,  J. W., H.  L.  Faucett, and  L. J.  Henson.  Economics of
     Disposal of Lime-Limestone  Scrubbing Wastes:  Untreated and
     Chemically  Treated  Wastes.  TVA Bull. Y-123, EPA-600/7-78-023a,
     February 1978.  452 pp.

 6.  Selmeczi, J.  G.  Flue  Gas Desulfurization and  Stabilization.
     Dravo Lime  Company, Pittsburgh, Pennsylvania, May 1975.

 7.  Poz-0-Tec Process for  Economical and Environmentally  Acceptable
     Stabilization of Scrubber  Sludge and Ash.   IU  Conversion  Systems,
     Inc., Philadelphia, Pennsylvania.

 8.  Conner,  J.  R.  Ultimate Disposal of Liquid  Wastes by  Chemical
     Fixation.   In:  Proceeding  of  29th  Annual Purdue  Industrial Waste
     Conference, Purdue  University, West Lafayette,  Indiana, May 7-19,
     1974.  pp.  906-922.

 9.  Kelso, T. M.   Monthly  progress report.   Tennessee Valley  Authority,
     Plant Operations Section,  Emission  Control  Development Projects,
     Muscle Shoals, Alabama, November-December 1976 and  January-
     February 1977.

 10.  Bechtel  Corporation.   Progress report  for work conducted  at  EPA
     Alkali Scrubbing Test  Facility at  TVA Shawnee Steam Plant,
     Paducah, Kentucky,  March  1977 to May  2,  1977.

 11.  Borgwardt,  R. H.   Sludge  Oxidation in Limestone FGD Scrubbers.
     EPA-600/7-77-061,  June 1977.
                                  175

-------
12.   McGlamery, G.  G.,  R.  L.  Torstrick, W. J. Broadfoot, J. P. Simpson,
     L.  J. Henson,  S.  V.  Tomlinson, and J. F. Young.  Detailed Cost
     Estimates for  Advanced Effluent Desulfurization Processes.  TVA
     Bull. Y-90, EPA-600/2-75-006 (NTIS PB 242 541), January 1975.
     418 pp.

13.   Torstrick, R.  L.,  L.  J.  Henson, and S. V. Tomlinson.  Economic
     Evaluation Techniques, Results, and Computer Modeling for Flue Gas
     Desulfurization.   In:  Proceedings of Symposium on Flue Gas
     Desulfurization,  Vol. I, Hollywood, Florida, November 8-11, 1977.
     EPA-600/7-78-058a, March 1978.  pp. 118-168.
                                 176

-------
                               TECHNICAL REPORT DATA
                         (Please read Instructions on the reverse before completing)
1. REPORT NO.
 EPA-600/7-79-092
                          2.
                                                     3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Proceedings: Industry Briefing on EPA Lime/Lime-
 stone Wet Scrubbing Test Programs (August 1978)
                               5. REPORT DATE
                                March 1979
                               6, PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
John E. Williams, Conference Chairman
                                                     8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
                                10. PROGRAM ELEMENT NO.
                                INE624A
See Block 12
                                11. CONTRACT/GRANT NO.

                                N.A. (Inhouse)
12. SPONSORING AGENCY NAME AND ADDRESS
 EPA, Office of Research and Development
 Industrial Environmental Research Laboratory
 Research Triangle Park, NC  27711
                                13. TYPE OF REPORT AND PERIOD COVERED
                                Proceedings: 8/29/78	
                                14. SPONSORING AGENCY CODE
                                 EPA/600/13
15.SUPPLEMENTARY NOTES BERL-RTP project officer is John E.  Williams, MD-61, 919/541-
2483.
16. ABSTRACT
 The proceedings document presentations made during the August 29, 1978 industry
 briefing conference which dealt with the status of EPA/IERL-RTP's flue gas desul-
 furization (FGD) research, development,  and application programs. Subjects con-
 sidered included: lime/limestone scrubbing test results,  forced oxidation, process
 cost and energy requirements, by-product disposal options, and future test plans.
 The conference provided developers, vendors, users, and those concerned with
 regulatory guidelines with a current review of progress made in lERL-RTP's FGD
 technology development program.
17.
                             KEY WORDS AND DOCUMENT ANALYSIS
                DESCRIPTORS
                                          b.lDENTIFIERS/OPEN ENDED TERMS
                                            c. COSATI Field/Group
 Pollution
 Flue Gases
 Sulfur Oxides
 Desulfurization
 Scrubbers
 Calcium Oxides
Calcium Carbo-
 nates
Oxidation
Waste Disposal
Power
Operating Costs
Pollution Control
Stationary Sources
Forced Oxidation
Energy Requirements
13B
21B
07B        07C
07A,07D
131         14G
        14A,05A
18. DISTRIBUTION STATEMENT
 Unlimited
                                          19. SECURITY CLASS (ThisReport}
                                          Unclassified
                                                                  21. NO. OF PAGES
                                                180
                    20. SECURITY CLASS (Thispage)
                    Unclassified
                                            22. PRICE
EPA Form 2220-1 (9-73)

-------