United States
                     Environmental Protection
                     Agency
Air and Energy Engineering
Research Laboratory
Research Triangle Park NC 27711
                     Research and Development
EPA/600/S7-85/025  Aug. 1985
SERA          Project  Summary

                     Bench Scale Studies  of
                     Limestone  Injection  for  S02
                     Control
                     P. L Case, L Ho, M. P. Heap, R. Payne, and D. W. Pershing
                       This report describes research carried
                     out in one  task of an  EPA program
                     entitled. The Development of Criteria
                     for Extension of Applicability of Low
                     Emission, High Efficiency Coal Burners.
                     The task involved using  a series of
                     bench-scale facilities to determine the
                     process parameters controlling the cap-
                     ture of sulfur species by calcium-based
                     sorbents when pulverized coal is burned
                     under low NO* conditions.
                       This Project Summary was developed
                     by EPA's Air and Energy Engineering
                     Research Laboratory. Research Trian-
                     gle 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 infor-
                     mation at back).


                     Objectives
                      Although coal  is the most abundant
                     source of fossil fuel energy in the U.S., its
                     use poses several serious problems for
                     society, not the  least of which  is the
                     emission  of atmospheric pollutants. The
                     combustion of coal results in the forma-
                     tion of sulfur and nitrogen  oxides (SO«
                     and NOx), which have been identified as
                     precursors of acid precipitation. Although
                     evidence is not conclusive, it is generally
                     recognized that the environmental dam-
                     age caused by acid precipitation can be
                     mitigated by reducing emissions of these
                     precursors from coal-fired power plants.
                     Consequently, there is a need to develop
                     efficient and cost effective control tech-
                     niques for these pollutants for retrofit
                     applications. This report addresses the
                     use of calcium based sorbents for in situ
                     sulfur capture under combustion condi-
tions that minimize NOx formation from
pulverized coal combustion.
  The use of calcium based sorbents to
remove S02 from the combustion prod-
ucts of coal and oil flames is not a new
concept. Considerable effort was expend-
ed during the late 1960s and early 1970s
to develop the dry limestone  injection
process. These activities were  curtailed
because sulfur capture was not particu-
larly  efficient and injecting limestone
caused considerable operating problems
(e.g., fouling in the convective sections
and plugging of the air heater). Recent
studies in Germany indicate that this
process could be a viable means of reduc-
ing SOa emissions if the  sorbent were
injected under low NOx conditions. In the
earlier studies, poor sulfur capture was
attributed to thermal deactivation of the
sorbent (deadburning) and poor sorbent
dispersion throughout the boiler. These
problems may be overcome if the sorbent
is injected with one of the air or fuel
streams in a low NO, burner. This would
solve the problem of dispersion since all
the fuel and air must mix if combustion is
to be completed. Thermal deactivation
could be reduced because low NOX condi-
tions minimize peak flame temperature.
Objectives of  the  bench  scale studies
described in this report were:

  1.   To  determine if the  reduced peak
     flame temperatures and extensive
     fuel-rich zones typical of low NO*
     combustion conditions are condu-
     cive to sulfur capture by dry sorbent.
  2.   To determine the optimum time/-
     temperature  stoichiometry history
     of the sorbent for sulfur capture
     during coal combustion.

-------
 3.  To determine if captures reported
     in several pilot scale tests are pos-
     sible in the current boiler popula-
     tion.
 4.  To provide preliminary information
     on the impact of calcium based sor-
     bents on boiler operability.

Approach
  When the task was initated,  insuffi-
cient information was available to allow a
complete definition of the bench  scale
studies. The questions raised in previous
experiments and the large number  of
process variables  suggested  that  a
phased  approach was necessary. This
method of planning and experimentation
allowed the investigation  of processes
and parameters which proved most impor-
tant and enabled a selection of test condi-
tions which would develop  an under-
standing of the sulfur capture process  in
the most cost effective manner. The pro-
gram organization is shown schemati-
cally in Figure 1. The task was initated by
conducting a series of program definition
tasks in an existing refractory tunnel fur-
nace, concurrent  with the design and
construction of a boiler simulator furnace
(BSF). Most of the experimental investi-
gations were carried out in the BSF. Fig-
ure 1 shows that the BSF experiments
were divided into three  categories: (A)
combustion parameter  screening, (B)
thermal history/sulfur chemistry studies,
and (C) fuel and sorbent studies. The data
from these experiments were used  to
plant the pilot scale studies carried out  in
other tasks of the same program and  to
define the needs for future experiments.
  The BSF, simulating the thermal his-
tory  of the products of combustion in a
wide variety of pulverized coal fired boil-
ers, was designed to satisfy the following
criteria:

• Independent control of wall tempera-
   ture throughout the furnace and con-
   vective sections.
• Fuel consumption rate  nominally 80
   Ib/hr (36 kg/hr).
• Accommodate various firing systems.
• Variable preheat on all combustion air
   streams.
• Simulation  of particulate dropout
   throughout the radiant and convective
   sections.
• Ability to rapidly and easily clean inte-
   rior sections.
• Potential for near adiabatic operation
   by minimizing heat loss.
      Program
      Definition
                       Preliminary
                       Shakedown
                       Experiments
                         BSF*
                        Design
                                                BSF
                                            Construction
     Boiler
     Simulator
     Furnace.
                          I
    (A)
Combustion
Parameter
Screening
                          \
                          (B)
                     Thermal History,
                     Sulfur Chemistry
                        Studies
                          i
                          (C)
                         Fue/&
                        Sorbent
                        Studies
                                      Results from
                                         Other
                                     Investigations
                                      \\
                                         Pilot
                                        Scale
                                       Studies
                                             Identification
                                             of Controlling
                                              Processes
Figure 1.   Organization of bench scale sulfur studies.
•  Potential  to extract  heat at  various
   locations.
•  Ease of access for sampling probes.
However,  foremost in the design was
recognition that the times and tempera-
tures experienced by the sorbent should
cover as wide a  range  as possible to
encompass those  encountered in actual
boilers. Figure 2 is a schematic of the BSF
system, showing  the sorbent and coal
feed and air supply systems and the gen-
eral arrangement of thefurnacesections.
                 The BSF was designed to simulate the
                 various parts of a boiler. Thus, the radiant
                 furnace simulates the main heat release
                 zone; the postflame cavity, the region
                 above the main heat release zone before
                 the furnace exit; and air-cooled tubes, the
                 superheater, reheat, and air heater sec-
                 tions of the convective pass.

                 Significant Results
                   Task results are in three groups: com-
                 bustion parameter screening, sulfur chem-
                 istry and thermal environment, and fuel
                 and sorbent effects.

-------
           Sorbent
           Feeder
O                                                                                      Standard Gas Sample Locations.
                                                                                      Continuous NO, SOz. Oa CO, CO2
                                                                                      Laminar Flow Elements (LFE)
                                             4/r Heaters. Propane Fired

                                             Probe Access— 6" to 10" Spacing,
                                             Entire Furnace Axis
Loss in Weight
 Coal Feeder
           Propane or
                                                                                                      Baghouse
                                                     Removable Water-
                                                      Cooled Pipes (3)
                                      Convective
                                       Section
 Figure 2.    Functional schematic of boiler simulator furnace.
 Combustion Parameter
 Screening Studies (Series A)
   The screening experiments defined the
 role of various combustion parameters in
 the sulfur capture process. The experi-
 ments were conducted in two sections,
 according to the method of combustion
 air staging used to simulate low NOX
 combustion systems. In the first series of
 experiments (internally staged), a sub-
 scale distributed mixing burner was used
 and the parameters investigated included
 firing rate, heat extraction, burner zone
 stoichiometry,  excess  air level, sorbent
 injection location (with the fuel or staged
 air), and staged air velocity. The second
 series of experiments (externally staged)
 were designed to investigate sulfur cap-
 ture  under fuel-rich conditions using a
.system which  provided for a relatively
' long (1 -2 sec) residence time under fuel-
rich conditions. In this second series of
tests, the parameters investigated in-
cluded firing rate, sorbent injection, heat
extraction rate, rich zone stoichiometry,
excess air level, and tertiary air location.
  The results of the  screening studies
indicated that thermal environment had
the major impact on sulfur capture by dry
sorbents injected directly into the fur-
nace. The burner-staged experiments
showed the influence of thermal envi-
ronment most clearly: a reduction in peak
furnace temperature due to cooling in the
radiant furnace or to a reduction in load
increased sulfur capture significantly.
Changes in other parameters (e.g., sor-
bent  location  or  injection velocity) had
secondary effects on sulfur capture. The
externally staged experiments indicated
that sulfur capture was less dependent
on overall heat extraction rate.
                                                              A main point of interest in the param-
                                                            eter studies was to establish the impor-
                                                            tance of rich zone capture, since it was
                                                            speculated that reactions  involving re-
                                                            duced sulfur species (e.g., H2S or COS)
                                                            could be captured more easily than SO2
                                                            Figure 3 shows overall sorbent utilization
                                                            as a function of the stoichiometry of the
                                                            fuel-rich zone for coal and propane doped
                                                            with H2S to give the same exhaust S02
                                                            level. I n this instance, the sorbent (Vicron)
                                                            was added with air in the fuel-rich zone at
                                                            the base  of the postflame section. Rich
                                                            zone stoichiometry is expressed in terms
                                                            of the gas phase for both the gaseous and
                                                            solid fuel. It can be argued that the differ-
                                                            ence in effectiveness with the two fuels
                                                            is associated with coal:  under fuel-rich
                                                            conditions, much of the sulfur remains
                                                            with the char and is not available for cap-
                                                            ture. When all sulfur is in the gaseous

-------
    30
    20
 .§'

 I
a
    10
                                              Q Coal


                                              /\ Propane + HiS
      0.4
Figure 3.
                     0.6             0.8             1.0

                     First-Stage Stoichiometry Ratio (Gas Phase)
1.2
           Capture and retention of sulfur under external staged conditions for coal and
           propane doped with H*S, sorbent added with air at base of postflame section.
           uncooled.
stage, calcium utilization under fuel-rich
conditions approaches 30 percent.

Sulfur Chemistry and Thermal
Environment Studies (Series B)
  These experiments clarified the effect
of thermal environment on sulfur cap-
ture, utilizing two series of experiments
to:

  1.  Investigate  sulfur evolution  and
     speciation in the fuel-rich zone of a
     staged combustion system.
  2.  Define the influence of a wide range
     of thermal environments on sulfur
     capture under  both internal  and
     external staged conditions.

Results from these tests include:

• Under fuel-rich conditions, sulfur spe-
   cies appear to be far from equilibrium.
• Even with gaseous fuels, sulfur cap-
   ture in the fuel-rich zone does not
   appear to significantly impact  overall
   sulfur capture.
                                        • With coal,  capture  in  the  fuel-rich
                                          zone is limited because of the reduced
                                          sulfur driving force when a large frac-
                                          tion of the sulfur remains in the solid
                                          phase.


                                       Fuel and Sorbent Effects
                                       (Series C)
                                         The influence of fuel and sorbent prop-
                                       erties on sulfur capture was investigated
                                       in the third series of bench-scale experi-
                                       ments. The sorbent parameters, which
                                       were varied, included particle  size and
                                       composition. Sulfur capture was meas-
                                       ured  with a  coal/water slurry  and a
                                       lignite, as well as the baseline bituminous
                                       Indiana coal.
                                         Figure 4 compares sulfur capture for
                                       the four fuels investigated with one sor-
                                       bent. The capture is worst with the Indi-
                                       ana coal and very similar for the lignite,
                                       doped propane, and the coal/water slur-
                                       ry. The dotted line shown in Figure 4 cor-
                                       responds to 50 percent capture at a calci-
                                       um/sulfur ratio of 2 with 2,500 ppm SO2.
The influence of sorbent type on sulfur
capture was fuel dependent. With the
Indiana coal, dolomite was by far the
most  effective  sorbent; whereas, with
doped propane, captures with  dolomite
and Vicron were similar. The reasons for
the interaction of fuel and sorbent effects
are not now known.


Conclusions
  Results of the experiments in the boiler
simulator furnace  indicated that the
parameters  of major  importance are
thermal  environment,  calcium/sulfur
ratio,  and sorbent composition. Thermal
environment (local temperature) was
shown to have a strong effect on the utili-
zation of dry sorbents injected directly
into  the furnace. The burner-staged
experiments  showed  the influence  of
temperature most clearly: a reduction in
peak flame temperature, due to cooling in
the radiant zone or to a reduction in load,
increased sulfur  capture significantly.
Changes in other combustion parameters
(e.g., sorbent location and injection veloc-
ity) had  secondary  effects on  sulfur
capture.
  The effect of thermal history under
physically-staged  conditions  was less
definitive. Changes in  both  load and
thermal cooling in the radiant furnace
were  complicated  by changes in furnace
temperature due to the physical staging
itself, since  physically staging the coal
flame significantly altered the furnace
temperature profile. With external stag-
ing near stoichiometric conditions, the
first (fuel rich) stage, which included the
entire radiant furnace, and the second
stage region (the  first part of the post-
flame cavity) were hotter than under
burner-staged  conditions. First  stage
Stoichiometry and sorbent location both
influence sulfur capture under externally-
staged conditions; however, the results
suggest that these influences may be
more  strongly related to thermal than to
chemical changes. Further, the data do
not support the concept that sorbent utili-
zation is much higher  under  fuel-rich
conditions.
  The influence of sorbent type (Vicron
45-3, Vicron 15-15, Michigan marl, hy-
drated lime [Ca(OH)2], and dolomite) on
sulfur capture was studied at five differ-
ent combustion conditions: Indiana No. 3
coal fired in the distributed mixing burner
with and without  external cooling, Indi-
ana No. 3 coal under externally-staged
conditions, and propane doped with HaS.
In general, dolomite gave good capture
under all conditions. Hydrated lime ap-

-------
     80
     60
 c
 u
 I
 I
 I
 
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S7-85/025
          0000329   PS

          U  S  ENVIR PROTECTION AG£NCY
          REGION  3  LIBRARY
          230  S  DEARBORN  STREFT
          CHICAGO             IL

-------