United States
Environmental Protection
Agency
Air and Energy Engineering
Research Laboratory
Research Triangle Park NC 27711
Research and Development
EPA/600/S7-87/003 Apr. 1987
Project  Summary
Evaluation  of  Sulfur  Capture
Capability of a  Prototype Scale
Controlled-Flow/Split-Flame
Burner
J. Vatsky and E. S. Schindler
  This  report describes large pilot
demonstration of sulfur capture using
copulverization of limestone with a high
sulfur eastern bituminous coal and com-
bustion of the mixture  using Foster
Wheeler's commercial  Controlled-
Flow/Split-Flame  (CF/SF) Low  NOX
burner. Optimization of the sulfur
capture was attempted through the use
of overfire air and two proprietary flame
temperature control methods. Addi-
tionally, the  effects of excess air
changes, load changes,  and  different
calcium/ sulfur mole ratios (Ca/S) were
evaluated. The CF/SF  burner  was
chosen because of its internal staging
and proven low IMOX capabilities; its use
in combination with two flame  tem-
perature reduction methods could re-
duce the flame temperature to minimize
dead burning  of limestone and  thus
enhance SO2 capture. Although the use
of flame  temperature reduction and
overfire air improved the SO2 capture,
the optimum SO2  capture of 29% at a
Ca/S of 2.15 was low. Operation under
optimum SO2 capture mode resulted in
measured NO, emissions of 0.19 Ib/106
Btu*; CO was less than 25 ppm at an
excess oxygen level of 3.0%. The testing
was done at a 42 x 106 Btu/hr heat
input horizontally fired pilot plant con-
figured like a conventional pulverized-
coal-fired boiler.
' Readers more familiar with metric units may use
 the factors listed at the back of this Summary to
 convert to that system.
  This Project Summary was developed
by EPA's Air and Energy Engineering
Research Laboratory, Research Triangle
Park, NC, to announce key findings of
the research project that Is fully docu-
mented In a separate report of the same
title (see Project Report ordering in-
formation at back).

Introduction
  This report summarizes a joint Environ-
mental Protection Agency (EPA)/Foster
Wheeler  Energy Corporation (FWEC) test
program  to  evaluate the m-situ SO2
reduction capabilities of limestone injec-
tion with a low NOX internally staged
burner when the limestone and coal are
copulvenzed and injected through the
coal nozzle. The tests  were performed
between April 13 and May 16, 1983. The
burner used is Foster  Wheeler's com-
mercial  Controlled-Flow/Split-Flame
(CF/SF) burner (Figure  i). The test pro-
gram was based on EPA's concept that, if
the limestone is intimately mixed with
the coal during the pulverization process
and burned under low NO* conditions,
high S02  capture levels can be obtained.
When this method of limestone injection
is combined with the low flame tempera-
ture characteristics of the CF/SF burner
and FWEC's proprietary flame tempera-
ture reduction methods, the total S02
capture may be  enhanced. Successful
achievement of the Limestone Injection
Multistage Burner (LIMB) process may
result in S02 reductions at a much lower
cost than with conventional wet removal
methods. Although this technique may

-------
                                                                                  Table 2.    Analysis of Test Limestone
Figure 1.    Controlled-flow/split-flame (CF/SF) burner
 not replace wet methods of S02 reduction,
 it would  be appropriate for retrofits of
 existing uncontrolled boilers firing high
 sulfur coals.
   The relatively short flame produced by
 the CF/SF burner is especially favorable
 for retrofits where the depth of the fur-
 nace is limited. Flames do not extend into
 the upper furnace which would increase
 the furnace exit gas temperature (FEGT).
 Increasing FEGT can cause fouling and
 slagging as well as uncontrolled steam
 temperatures and reduced efficiency.
   The tests were run at FWEC's Japanese
 licensee  Ishikawajima  Harima Heavy
 Industries Co.,  Ltd. (IHI) Aioi Works in
 Japan where IHI has a large coal com-
 bustion test facility. The fuel used is a
 high  sulfur western Pennsylvania  bitu-
 minous coal from  the Middle Kittaning
 Seam with  a sulfur content of  about
 3.1%. An analysis of the fuel is shown in
 Table 1.
   This fuel  was chosen because  it is
 typical of the fuels used in older boilers
 that  may be susceptible  to acid rain
 control legislation. The limestone chosen
 isVicron from California's Lucerne Valley.
 It is a high calcium limestone and was
 chosen because it had been used before
 in other EPA test  programs and would
 allow more relevant comparison of S02
Table 1.    Analysis of Test Fuel
Fuel Name
  Origin
Middle Kittaning
  Western PA
Proximate
  Fixed Carbon,'
  Volatiles, %
  Ash.%
  Moisture, %

Ultimate
  Carbon, %
  Hydrogen, %
  Oxygen, %
  Nitrogen, %
  Sulfur, %
  HHV. Btu/lb

Operating
Conditions

Fuel Rate
  kg/hr

Heat Input
  106 Btu/hr
          50.1
          34.6
           9.5
           S.8
          68.7
           4.6
           7.1
           1.2
           3.1
      12,818
       1.500
          42.3
capture among EPA test programs.  An
analysis of the  limestone is shown  in
Table 2.

Test Facility Description
  IHI's test facility is designed to evaluate
fuels and combustion  systems  on a
Name
Origin
CaCO3, %
MgC03. %
SiO2. %
AI2O3, %
Fe203
Moisture, %
Surface, %
Inherent, %
Vicron
Lucerne Valley, CA
98.1
0.9
0.11
0.01
0.01

0.03
0.1
prototype scale (up to 50 x 106 Btu/hr).
Functionally  useful  steam is not gen-
erated  so  that operation and design
changes do not affect the steam supply to
industrial or power generation equipment.
This provides an atmosphere conducive
to testing without interruption.
  The pulverized coal system differs from
that which is in current commercial prac-
tice on  pulverized coal-fired boilers. An
indirect storage system is used and allows
wide variations in  air/coal ratios. The
limestone bunker supplies, via a feeder, a
Foster  Wheeler vertical pulverizer. The
coal and limestone are mixed and pul-
verized  in the mill to  a minimum coal
fineness of 70% through 200 mesh. The
pulverized fuel is carried pneumatically
to a cyclone separator where the fuel is
separated from the carrier  air and  fed
into a  pulverized fuel  bin; a baghouse
filters the air before exhausting to  the
atmosphere, and collects the fines which
are also fed  into the fuel bin. A screw
feeder at the botton of the fuel bin  feeds
the pulverized fuel over a weighing device
and  into a  fuel/primary air mixer. This
allows great  flexibility in controlling  the
primary air to fuel ratio.
  The facility is fired by a single burner
which  simplifies burner flame  studies
since flame  interactions do  not occur.
The furnace is refractory lined to simulate
utility size  furnace  heat release  rates.
Nine view  ports along  each side of  the
furnace at  the burner  level, along with
five others at upper elevations, allow the
operator to observe  the flame and take
temperature  measurements at different
points along the flame's length. Overfire
air ports are available for staging tests.
  The combustion air takes the following
path through the system. A forced draf'

-------
ifan supplies atmospheric air to the shell
side of a tubular air preheater where it is
heated up to the range of 536 to 653°F.
Hot air is mixed with cold tempering air
to obtain the desired primary air temper-
ature.  The  remaining  hot  air  is  then
supplied to the wmdbox.  Combustion
products pass out of the furnace, through
a  convection section  and  through the
tube side of the preheater, after which it
is cleaned of  paniculate  matter  in a
multiclone  and then a baghouse. After
the baghouse, an induced draft fan forces
the combustion products to the  stack.
Ttibiu  3 summarizes   basic  system
parameters.

Test Methodology
  The intention of  the test program was
to evaluate various operating modes for
their potential to improve SO2 reduction
obtainable by copulverization of coal and
limestone.  A number  of  variables were
evaluated:
   Overfire Air
   Furnace Excess Oxygen
   Calcium to Sulfur Mole Ratio
   Two Proprietary Flame Temperature
    Reduction Methods
   Load
   Overfire Air Injected Higher
    in the Furnace
   Burner Parameters
  These variables were thought to have
the greatest potential  in  improving S02
capture.  This was  especially true of the
two flame temperature reduction methods
where,  in the  past, peak flame temper-
ature reductions of 70 to 90°F were seen
singly and over 200°F was obtained when
these methods were combined
  The in-depth evaluation consisted of a
complete full  factorial matrix  of tests:
testing each variable in combination with
every other combination of other  vari-
ables  Furnace excess oxygen  was  an
exception in that only a half factorial was
planned
  Simultaneously with the determination
of the  effect each  variable  has on S02
reduction, S03, NOX, CO, and total hydro-
carbons were  measured. The  intention
was  to observe the effect each variable
had on other emission species to evaluate
the overall  emission characteristics of
each  combination  that  improved SO2
reduction.

Major Results and Conclusions

Gaseous Emission Levels
  • S02 Emissions
      The  addition of limestone to the
    fuel at a Ca/S  of 2 15 resulted in an
Table 3.
System Specifications
Furnace
Burner
Coal Handling
Pulverizer
         Width
         Depth:
         Height.

         Coal
         Overfire Air
         Heat Liberation-
        Elevator
        Bunker-
        Table Feeder-

        Type:
        Capacity.
        Fineness:
Tubular Air Preheater Air Flow Rate:
                  Air Temp. Inlet:
                  Air Temp Outlet:

Paniculate Collection Type:
Equipment         Gas Flow Rate •
                  Oust Loading
                    Inlet:
                    Outlet:

Limestone Handling  Bunker.
                  Feeder:
3100 mm (10.2 ft)
4500 mm (148 ft>
11,000 mm (36 ft)

200 kg/h (4.400 Ib/h)
As Necessary
Max 111 x. 10ekcal/m3hx
  (12.5x 103 Btu/ft3h)

1.  5 T/h (11 x 103 Ib/h)
1.  10m3 (350 ft3)
2.  15 T/h (33 x 103 Ib/h) max.

IHI-FW Ring & Roller Mill MBF-16
8 T/h (17 xlO3 Ib/h)
70% through 200 mesh

31 T/h(68x103lb/h)
20°C(70°F)
320°C(610°F)

Baghouse following a multiclone
20.000 Nrrf/mm (12440 scfm)

36. g/Nm3 (87 gr/scf)
0.1. g/Nm3 (0.242 gr/scf)

225 kg/h (500 Ib/h) Max.
20-320 kg/h (50-700 Ib/h)
    optimum emission reduction of 28%.
    This is an improvement over the 22-
    23% found without any changes in
    operation of the burner-furnace. This
    optimum was found with a combi-
    nation of 3% excess O2, 20% overfire
    air, and with FW's proprietary flame
    temperature  reduction  method #2
    (FTRM#2). Another combination of
    operating variables (5% excess O2
    and FTRM #2)  resulted in higher
    SO2 reduction at the same Ca/S,
    but it also increased NOX emissions
    such that the total of acid forming
    emissions of SO2 and NOX was
    higher than for the optimum case.
      Increasing the limestone addition
    rate during otherwise normal oper-
    ating conditions (i.e., optimum NOX
    burner settings, 3% excess 02, no
    limestone, no overfire air, full load,
    and no flame temperature reduction
    methods in use), to a Ca/S of 3.26
    resulted in  33.4%  reduction. This
    increase in S02 reduction is  es-
    sentially linear up to Ca/S = 3.26. If
    the optimum S02 control method is
    extrapolated  to  Ca/S  = 3.26,  the
    SO2 reduction would  increase to
    43%. Although it is generally con-
    ceded that  S02 reduction is  not
    linear  with Ca/S, a linear relation-
    ship was found  up to  a Ca/S of
    3.26, the maximum value tested.
                                  However, this optimum occurred
                                  with overfire air ports open, resulting
                                  in slagging.
                                • NOX Emissions
                                     In general,  adding  limestone  to
                                  the fuel reduced NOX by 10%. Under
                                  normal operating conditions, the
                                  NOX emission rate measured for the
                                  CF/SF burner was 0.32 lb/106 Btu.
                                     This represents a 60% reduction
                                  from the predicted uncontrolled NOX
                                  emission rate of 0.8 lb/106 Btu using
                                  this fuel and a pre-NSPS burner at
                                  this test facility. Under the conditions
                                  of optimum SO2 reduction, the NOX
                                  decreased to 019 lb/106 Btu, an
                                  additional 41% reduction from 0.32
                                  lb/106  Btu  (or a total  of  a 76%
                                  reduction from  the  uncontrolled
                                  level). About 25 to 30% of this addi-
                                  tional NOX reduction can be attributed
                                  to overfire air  (OFA), 10%  can be
                                  attributed to adding limestone to the
                                  fuel, and 1% is attributed to the use
                                  of FWEC's proprietary FTRM #2. The
                                  negligible NOX reduction due to the
                                  FTRM #2 is expected since the peak
                                  flame temperature is already sub-
                                  stantially below 2900°F.
                                • CO Emissions
                                     In general, adding limestone at a
                                  Ca/S of 2.15  reduced CO concen-
                                  trations at the economizer outlet to
                                  below 35 ppm corrected to 0% excess

-------
    02,  with one exception. Under
    normal operating conditions, the CO
    averaged 39 ppm. Under the opti-
    mum S02 reduction test conditions,
    the CO concentration dropped to 24
    ppm, a 38% reduction. Half of this
    reduction can be attributed to lime-
    stone addition; the remainder can
    be attributed to FIRM #2.
  • SO3 Emissions
      In general, adding  limestone to
    the fuel  at a Ca/S = 2.15 always
    reduced S03 concentrations to below
    20  ppm corrected to 0% excess O2
    regardless of the initial concentra-
    tion. Under normal operating condi-
    tions S03 concentrations averaged
    28  ppm.  Under conditions of opti-
    mum SO2 reduction the SO3 con-
    centrations were reduced to 8 ppm,
    a 71% reduction. About equal per-
    centages of this reduction  can be
    attributed to OFA  and limestone
    addition.
  • Total Hydrocarbons (THC)
      Limestone addition in many cases
    decreased the THC emissions, but
    there were  many exceptions  The
    only operating variable that had a
    consistent effect on THC was FTRM
    #1, and  it increased THC. Neither
    excess oxygen nor overfire air had a
    consistent  effect on  THC. Under
    normal operating conditions the THC
    concentrations averaged 3 ppm cor-
    rected to 0%  excess O2. Under
    conditions of optimal S02 reduction
    the THC was reduced to 1.7 ppm, a
    43% reduction The reduction is at-
    tributed to a  synergistic effect of the
    combination of OFA, FTRM #2, and
    limestone addition  since  none of
    these (alone) consistently  reduced
    THC


SO2 Capture in the Baghouse
      No significant S02 or SO3 reduc-
    tion was  measured across the bag-
    house with  or without limestone
    addition.  SO3 reduction was mea-
    sured across the air  heater. SO3
    change across the air heater cannot
    be  explained;  additionally, the  re-
    duction is virtually  independent of
    the presence of limestone. S03 con-
    centrations dropped from an average
    8 ppm to  about 0.3 ppm when lime-
    stone was  being added; S03 was
    reduced from 18.3 to 0.6 ppm when
    limestone was not being added, and
    was further  reduced  to  0 4 ppm
    across the baghouse. The air heater
    is tubular with an exit temperature
Table 4,    Ash Fusibility Temperature
                                              Ash Fusion Temperatures, °F

                                              Test 42             Test 43
Ca/S

Oxidizing Atmosphere
Deformation
Softening
Hemisphere
Flow
         0
       2372
       2408
       2507
       2561
                         2.15
            2426
            2453
            2516
            2705
Reducing Atmosphere
Deformation
Softening
Hemisphere
Flow
1940
1958
1976
2453
2156
2246
2345
2552
    of about 500°F. Consequently, there
    should be no S03 condensation prior
    to the baghouse. At these tempera-
    tures the SO3 level should  remain
    constant unless absorption is occur-
    ring on some surface. All test results
    are corrected to 0% excess oxygen.

Effect on Equipment
  • Furnace and Slagging Potential
      No detrimental side effects were
    noted. The addition of limestone to
    the fuel did not increase the slagging
    potential of the coal. The coal was
    considered to be of medium  to high
    slagging  potential. During  normal
    combustion,  both with and without
    limestone, slagging was not evident;
    but, when  overfire air was used,
    slagging was evident, both with and
    without limestone This  was  fully
    expected based on an analysis of the
    ash constituents  and oxidizing/
    reducing fusion temperatures shown
    in Table 4. These results show that
    the ash  fusion  temperatures  are
    higher when  limestone  is being
    added at a Ca/S mole ratio of 2 15.
      All ash fusion temperatures in-
    crease as the furnace conditions
    change from reducing to  oxidizing.
    In this case the ash softening tem-
    perature increases by 450°F without
    limestone addition; and  by  207°F
    with limestone. Also all ash fusion
    temperatures  increase   with the
    addition of limestone The increase
    is largest under reducing  atmo-
    sphere. The reducing ash  softening
    temperature increases 288°F, and
    the reducing hemisphere  tempera-
    ture increases by 369°F when lime-
    stone is added to the fuel.
  • Baghouse
      The baghouse  operated normally
    during the test program. No increase
    in pressure drop was seen. The daily
    start-up/shutdown cycle did  not
    precipitate any bag blinding.
  • Burner
       No detrimental side effects were
    noticed on  the burner,  flame, or
    combustion in general.

Metric Conversion
  Readers more familiar with metric units
may use the following factors to convert
to that system
Nonmetnc
                Times
                         Yields Metric
Btu/hr
Btu/lb
°F
Ib/W6 Btu
  1 054
  232
  5/9(°F-32)
430
kJ/hr
J/g
°C
ng/J

-------
     J.  Vatsky and E. S. Schindler are  with Foster Wheeler Energy Corporation,
       Livingston, NJ 07039.
     Charles C. Masser is the EPA Project Officer (see below).
     The complete  report,  entitled "Evaluation of Sulfur Capture Capability of a
       Prototype Scale Controlled-Flow/Split-Flame Burner," (Order No. PB 87-168
       670/AS; Cost: $18.95, subject to change) will be available only from:
             National Technical Information Service
             5285 Port Royal Road
             Springfield, VA 22161
             Telephone: 703-487-4650
     The EPA Project Officer can be contacted at:
             Air and Energy Engineering Research Laboratory
             U.S. Environmental Protection Agency
             Research Triangle Park, NC 27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S7-87/003
                0000329   PS

-------