United States
                   Environmental Protection
                   Agency
Air and Energy Engineering
Research Laboratory
Research Triangle Park NC 27711
                   Research and Development
EPA/600/S9-86/029 Mar. 1987
&EPA          Project Summary

                    Proceedings:  1986 Joint
                    Symposium on Dry SO2 and
                    Simultaneous SO2/NOX
                    Control Technologies
                   Richard D. Stern and George R. Offen
                     The 1986 Joint Symposium on Dry
                   SO2 and Simultaneous SO2/NOX Con-
                   trol Technologies was held in Raleigh,
                   North Carolina, June 2 through 6,1986.
                   Forty-nine papers were presented by
                   EPA  and EPRI staff  members, repre-
                   sentative of utility companies, equip-
                   ment manufacturers, sorbent suppliers,
                   and  research and  development com-
                   panies under contract to EPA, EPRI,
                   and other research groups. Additionally,
                   researchers from Austria, Canada, Fin-
                   land, the Federal Republic of Germany,
                   and the Netherlands presented informa-
                   tion on world wide technological devel-
                   opments in dry SO2  control. Specific
                   topics discussed included enhanced
                   sorbent utilization through thermal pro-
                   treatment of sorbents or use of hydrated
                   or promoted  calcium-based  sorbents,
                   effects of temperature and other injec-
                   tion  parameters on sulfur capture ef-
                   ficiency, economics of dry sorbent
                   injection, post-furnace injection for SO2
                   control, effects of dry sorbent injection
                   processes on  electrostatic precipitator
                   performance,  waste management op-
                   tions, and the global status of com-
                   mercial-scale dry  sorbent  injection
                   processes.
                     This Summary contains abstracts of
                   the 49 technical papers presented at
                   the Symposium and  3 unpresented
                   papers distributed at  the  meeting.
                   Abstracts denoted  by an asterisk (*)
                   describe  research wholly or partially
                   funded by EPA and have been reviewed
                   in accordance with EPA peer and ad-
                   ministrative review processes.
  This Project Summary was developed
by EPA's Air and Energy Engineering
Research Laboratory, Research Triangle
Park, NC, to announce key findings of
the symposium that Is fully documented
in two separate volumes (see Project
Report ordering information at back).
Introduction
  The 1986 Joint Symposium on Dry S02
and Simultaneous SO2/NOX  Control
Technologies was held in Raleigh, North
Carolina, June 2 through 6, 1986. This
symposium, jointly sponsored  by EPRI
and EPA, was the second meeting of its
kind devoted solely to the discussion of
emissions  control  processes based  on
dry injection  of  calcium  or  sodium
sorbents to meet S02 and NOX regulations
for coal-fired power plants. Specific topics
discussed  included enhanced  sorbent
utilization through thermal pretreatment
of sorbents or use of hydrated or promoted
calcium-based sorbents, effects of tem-
perature and other injection parameters
on sulfur capture efficiency, economics
of dry sorbent  injection, post-furnace
injection for SO2 control, effects of dry
sorbent injection processes on electro-
static precipitator  performance, waste
management options,  and the global
status of commercial-scale dry sorbent
injection processes.
  Forty-nine papers were presented by
EPA and EPRI staff members, representa-
tives  of utility  companies, equipment
manufacturers,  sorbent suppliers,  and
research and development companies

-------
under contract to EPA, EPRI, and other
research groups. Additionally, research-
ers from Austria, Canada, Finland, the
Federal Republic of Germany, and the
Netherlands  presented information  on
worldwide technological developments in
dry S02 control.
  Symposium cochairmen were George
R. Offen, Project Manager in  the Air
Quality Control Program of EPRI's Coal
Combustion  Systems  Division; and
Richard  D.  Stern,  Chief of the  LIMB
Applications  Branch of  EPA's Air and
Energy Engineering Research Laboratory.
Participants in the  introductory session
included keynote speaker John Riggs,
Staff Director of the U.S. House of Repre-
sentatives Subcommittee on Fossil Fuel
and  Synthetic  Fuels; and  William
Holewinski, Director of Generating Station
Engineering  at Wisconsin Power and
Light Company, who presented the utility
perspective on acidic deposition control.
Overviews  of  the EPA and EPRI acidic
deposition  control programs were pre-
sented by John Skinner, Director of EPA's
Office of Environmental Engineering and
Technology, and John  Maulbetsch, Pro-
gram Manager in the Air Quality Control
Program of EPRI's Coal Combustion Sys-
tems Division, respectively.
  The Proceedings of the 1986 Sympo-
sium have been compiled in two volumes.
Volume 1, Sorbents, Process Research,
and Dispersion, contains papers from the
first four sessions:
  •   Session I: Introduction
  •  Session II: Sorbents
  •  Session III: Process Research
  • Session IV: Mixing/Dispersion
Contained  in Volume  2. Economics,
Power Plant Integration, and Commercial
Applications, are papers presented in the
remaining four sessions:
  •   Session V: Economics
  •   Session VI: Post-Furnace SO2
                Removal
  •   Session VI: System Impacts
  •  Session VIII: Commercial Scale
                Applications

  Also included  in Volume 2 are three
unpresented papers submitted in con-
junction with the Symposium or associ-
ated with a poster session held during
the Symposium.

SESSION I: INTRODUCTION

Chairmen, Richard Stern,
EPA, AEERL, and
George Offen,
EPRI                       Paper 1A
KEYNOTE ADDRESS

John Riggs,
Fossil and Synthetic
  Fuels Subcommittee,
U.S. House of Representatives

  This paper presents four political issues
that are  of primary importance to  most
elected officials in Washington: energy,
security, the  environment, electricity
prices and the Federal budget deficit.
Following this, is a summary of how Con-
gress' attempt to reconcile these issues
brought about the creation of the Federal
Clean Coal Technology Reserve. Finally,
the impact of clean coal technologies on
the debate over acid rain legislation is
discussed.

                          Paper 1B

COAL COMBUSTION
OVERVIEW/TECHNICAL
PERSPECTIVE OF THE
EPA PROGRAM*

John H. Skinner,
Office of Environmental Engi-
  neering and Technology,
EPA
  This paper presents an overview of the
program being undertaken by EPA in the
area of control technology development
and evaluation for coal combustion pro-
cesses. It presents a perspective on the
Limestone  Injection Multistage Burner
(LIMB) program and where it may fit into
a National Acid Rain Control Strategy.
Some of the technical  issues which
require resolution are discussed.

                          Paper 1C

OVERVIEW/TECHNICAL
PERSPECTIVE OF THE
EPRI PROGRAM

John S. Maulbetsch,
Coal Combustion Systems
  Division,
EPRI
  This paper summarizes the technical
perspective of EPRI's program for dry SO2
control technology research and develop-
ment. The results of research conducted
since  1984 are discussed briefly. The
current EPRI view of the applicability of
dry SO2 control technologies is presented,
particularly in  light of proposed acid rain
legislation.

                          Paper 1D
UTILITY PERSPECTIVE

William Holewinski,
Wisconsin Power & Light
  Company
  This paper describes the perspective of
Wisconsin Power & Light Company, a
medium-sized utility, on emissions con-
trol regulations and  potential control
technologies used  to meet them.  Of
particular  concern  is development ol
cost-effective S02, control technologies
that  may be applied  to smaller, oldei
plants  with  cyclone-fired boilers. Dr>
sorbent injection is  considered a  prime
candidate for use in this application.

SESSION II, PART 1:
SORBENTS — SELECTION,
PREPARATION, AND
PERFORMANCE

Chairmen, Brian Gullet,
EPA, AEERL
                           Paper 2/

PRECALCINATION AND ITS
EFFECT ON SORBENT
UTILIZATION DURING UPPER
FURNACE INJECTION

V. P. Roman,
S. J. Bortz,
R. J. Yang,
KVB, Inc.
P. Flament,
International Flame Research
   Foundation, and
G. R. Offen, Electric Power
   Research Institute
  An experimental program using eigl
sorbents  was conducted to determir
optimum methods of producing high su
face area limes and to  assess the
potential for achieving  high sorbei
utilizations in the 1200°C-9OO°C sulfatic
window. The performance of a fixed b<
calciner operated at 750-800°C  and
plug flow reactor operated at 900-920*
were compared on the basis of the surfa<
area produced versus the amount
calcining air on N2  used. The plug flc
reactor proved capable of producing 4
m2/g surface areas from CaC03 sorber
with about 25g air/g sorbent, while
produce a similar surface area with t
fixed bed reactor required an air/sorbe
ratio of 100/1.
  The  surface areas obtained were  r
significantly affected by either the si

-------
bent's trace mineral matter or magnesium
content.
  When the high surface area precalcines
were reinjected into a flue gas environ-
ment without SOj at 1100°C, the pre-
calcine was found to quickly (100 ms)
sinter to the surface area generated
within 100 ms  when injecting the raw
(uncalcined sorbent) into the same condi-
tions. Sulfation experiments at 1100°C
and a residence time of 0.5 s showed
littler difference between the utilization
levels obtained with 40 mVg precalcines
or raw sorbents.
                          Paper 2B
PILOT SCALE FURNACE
EVALUATION OF HYDRATED
SORBENTS FOR SO, CAPTURE*
G. C. Snow and
J. M. Lorrain,
Acurex Corporation
S. L. Rakes,
U.S. Environmental Protection
  Agency
  Several candidate sorbent materials for
the full scale Limestone Injection Multi-
stage Burners  (LIMB) demonstration at
Edgewater Station in Ohio were evaluated
in the EPA Innovative Furnace. Measure-
ments of S02 capture were made while
firing at 13.8 kW (47300 Btu/hr) on both
a Pittsburgh No. 8 coal (2.6% sulfur) and
doped natural gas. The sorbent injection
point was at 1200°C. With  a  furnace
quench rate of 260°C/s,  the average
residence time  at reaction temperatures
was 1.3 s. At  molar Ca/S ratios of 2,
Vicron  45-3 limestone produced at 38%
reduction in S02emissionsfrom baseline
levels. Increased SO2 capture rates,
ranging from 60 to 69% were measured
for various calcitic atmospheric hydrates.
Dolomitic sorbents were able to achieve
captures of 73 to 75% and showed no
significant difference  between atmos-
pheric  and pressure hydrates. With the
addition of sodium bicarbonate (NaHCOa)
at molar Ca/Na = 15, captures of 83 and
88% were possible for the calcitic and
dolomitic hydrates, respectively.

                          Paper 2C

DRY SORBENT SO2 CONTROL
FOR NEW POWER PLANTS
BURNING LOW SULFUR
WESTERN COALS

D. P. Teixeria and
T. A. Lott, Pacific Gas and
  Electric Company
 L. J. Muzio,
 Fossil Energy Research
   Corporation
  The Pacific Gas and Electric Company
 is developing a dry calcium-based sorbent
 injection S02 control process capable of
 high levels of SO2 removal. The target is
 95% S02 removal for new power plants
 burning low sulfur Western coals. Sor-
 bent performance at low sulfur combus-
 tion conditions (500 ppm flue gas S02)
 has been studied  in a 105,000 kJ/hr
 (100,000 Btu/hr) gas-fired combustion
 reactor. Sorbent utilization  was  seen to
 decrease significantly with decreasing
 SO2 concentrations; gas-phase diffusion
 is probably the limiting factor.  Experi-
 mental atmospheric dolomitic and calcitic
 hydrates were prepared and tested. S02
 removals exceeded those of commercial
 materials; 60% removal was achieved at
 calcium to sulfur ratios of 2 to 1.
                          Paper 2D

 PHYSICAL PARAMETERS
 GOVERNING THE REACTIVITY
 OFCa(OH)2WITHSO2*
 D. A. Kirchgessner and
 B. K. Gullett,
 U. S. Environmental Protection
  Agency
 J. M. Lorrain,
 Acurex Corporation
  SO2  capture data  from  bench- and
 pilot-scale experiments show that, among
 the calcium-based sorbents, Ca(OH)2 is
 superior in performance to both CaCO3
 and CaO. The available data also show
 that hydrates  vary in their ability to
 capture SO2. A bench-scale atmospheric
 hydration process, in conjunction with a
 flow reactor for S02 reactivity testing, is
 being  used to determine  the process
 parameters and material characteristics
 that affect sulfur capture by hydrates.
 Thus far the effects of parent lime particle
 size and degree of burn, and the hydration
water temperature have been evaluated.
Variations in each of these parameters
 appear to be capable of altering the ability
 of Ca(OH)2 to capture SO2.

                          Paper 2E

ENHANCED UTILIZATION OF
FURNACE INJECTED
CALCIUM-BASED SORBENTS

G. F. Weber,
M. E.  Codings, and
M. H. Bobman
University of North Dakota
  Energy Research Center
  Simultaneous control  of  SOX/NOX
emissions is under investigation at the
University of North Dakota Energy Re-
search Center (UNDERC). Furnace injec-
tion tests have focused on the direct
injection of  pressure-hydrated lime fol-
lowed by particulate collection in a bag-
house operated at temperatures of 63 to
149°C.  S02 reduction as high as 80%
has been  observed  for  high-calcium
pressure-hydrated lime (PHL) injected at
a Ca/SO2 mole ratio of 2.0 with all of the
S02 capture occurring in the flue gas at
temperatures exceeding 760°C.
  Pilot-scale tests have shown that fur-
nace injection of high-calcium PHL at a
Ca/S02 ratio of 1.0 can  result in 65 to
70% S02 capture  if the baghouse is
operated at 71 °C,  approximately 11°C
above the dew point. Improved sorbent
utilization has been observed both for
high-calcium PHL and for additive en-
hanced high-calcium  PHL at baghouse
temperatures of 63 to 104°C. Of the 65
to 70% overall utilization, 45 to 50% of
the sorbent was utilized in the upstream
flue gas and an additional 15 to 25% in
the baghouse.
                          Paper 2F
REACTIVITY OF ATMOSPHERIC
AND PRESSURE HYDRATED
SORBENTS FOR SO2
CONTROL*
J. M. McCarthy,
S. L. Chen,
J. C. Kramlich,
W.  R. Seeker, and
D. W.  Pershing
Energy and Environmental
  Research Corporation
  The purpose of this study was to develop
an  understanding  of the factors  that
control the reactivity of hydrated sorbents
toward SO2 in coal fired furnaces. The
study focused on  the impacts  of the
hydrate properties such as particle size,
surface area, and chemical composition,
and the furnace temperature of the injec-
tion location. Bench- and laboratory-scale
hydrators were used to  produce atmo-
spheric and pressure hydrated sorbents,
with parameters pertinent to the hydration
process varied. The chemical and physical
properties were characterized for these
hydrates and several commercially avail-
able hydrates, and they were tested for
SO2  sorption reactivity on bench scale
(17.6 kW)  and  pilot scale  (300  kW)
furnaces.

-------
  The results of this study indicated that
pressure hydrates generated under well
controlled conditions were more reactive
than commercially produced atmospheric
hydrates. The important production and
operating parameters for the pressure
hydration process included the size and
composition of the quicklime, the hydra-
tion temperature and pressure, the water
addition, and  the pressure progression
during discharge. Comparison of  com-
mercial atmospheric hydrates significant
differences in reactivity, ranging from 40
to 58% at a calcium to sulfur molar ratio
of 2. All hydrates, atmospheric and pres-
sure,  exhibited  a strong dependence
injection temperature.
                           Paper 2G

SORBENT DEVELOPMENT AND
PRODUCTION STUDIES*

J. P.  Gooch, E. B. Dismukes,  and
R. Beittel,
Southern Research Institute
J. L. Thompson,
Dravo Lime Company
S. L.  Rakes,
U. S. Environmental Protection
  Agency
  Recent pilot-scale studies have shown
that furnace injection (at temperatures of
approximately 1200°C) of commercially
produced calcitic hydrated lime will pro-
vide calcium utilizations in the range of
25% at Ca/S ratios of 2 to 1. This paper
presents results obtained to date on an
EPA-sponsored project with the objective
of developing calcium-based  sorbents
which  obtain higher utilizations  than
those achieved with commercial hydrated
lime.
  Sorbent preparation  methods  con-
sidered include: 1) production of high
surface  area  calcines using  modified
kilns, 2) production of pressure-hydrated
calcitic limes,  and 3) modified  hydration
processes. Method  3 was chosen for
laboratory- and pilot-scale studies. Labo-
ratory experiments  have produced hy-
drated lime with uncalcined surface areas
in excess of 80 m2/g. Preliminary cal-
cination experiments suggest that these
high surface  area hydrates will sinter
more quickly than conventional hydrates
in a furnace  environment. Efforts are
underway to produce quantities sufficient
for testing in a  106 Btu/hr (106 kJ/hr)
pilot-scale combustor. Sulfation  studies
with coal-firing have been performed with
pressure-hydrated calcitic lime produced
by other laboratories. Results from these
tests are presented, together with results
obtained with several commercial atmo-
spheric hydrates.

SESSION II, PART 2:
SORBENTS - PROMOTERS
AND ADDITIVES

Chairmen, Blair Martin,
EPA, AEERL

                           Paper 2H

ENHANCED SULFUR CAPTURE
BY PROMOTED CALCIUM-
BASED SORBENTS*

D. M. Slaughter,
S. L Chen, and
W. R. Seeker
Energy and Environmental
   Research Corporation
  Injection of calcium-based sorbents are
currently being considered as a potential
method of reducing S02 in existing coal-
fired boilers. This study investigated the
ability of small concentrations of additives
to enhance sulfur capture with calcium-
based sorbents. The interaction between
the mineral ash (produced from coal
combustion) and the sorbent/additive
combination was also studied. Two  fur-
nace facilities and an X-ray diffractometer
were used in the experimental study. The
furnances used included a 300 kW down-
fired furnace capable of simulating time
and temperature profiles for a variety of
boilers,  and a bench-scale drop-tube
furnace.
  The results show that calcium-based
sorbents, when in contact with mineral
ash, remove sulfur from the gas phase.
Under combustion conditions envisioned
for application of this technology, con-
tacting between mineral ash and sorbent
was small; therefore, changes in reactivity
of the sorbent were small.
  Addition  of known promoters with
sorbents has effectively increased sulfur
capture. Both sodium and chromium react
with the calcium  to increase the acces-
sibility  of  CaO  sites  by particle frag-
mentation,  creation of large cracks (or
pores), and the presence of a liquid phase.
In general,  iron did not  promote  an in-
crease in sulfur capture under combustion
conditions in a coal-fired boiler or promote
physical changes  in the sorbent.
  In general the chromium eliminated by
ash interaction was small, while the ash
effectively prevented a significant portion
of the sodium from contacting and re-
acting with the calcium sorbent. If the
impact of the sodium  promoter  is to
change the physical characteristics, as
believed,  process methods before injec-
tion  may be implemented  that allow
interaction of the sodium without inter-
ference from the mineral ash.

                            Paper 21

THE EFFECTIVENESS OF
ADDITIVES FOR ENHANCING
SO2 REMOVAL WITH CALCIUM
BASED SORBENTS
L. J. Muzio,
Fossil Energy Research
   Corporation
G. R. Often,
Electric Power Research Institute
A. Boni,
Physical Sciences, Inc.
R. Beittel,
Southern Research Institute
  Additives incorporated with calciun
based sorbents  have been  reported t<
enhance  sulfur  capture by increasing
mass transport within the sorbent particle
Alkali metal additives can be added to th
sorbent as physical mixtures or incorpc
rated with the sorbent during the hydra
tion process.  Another class of additiv
compounds, alcohol water solutions, ca
be used to alter the hydration proces;
producing high  surface area hydrate;
This  paper focuses on  both types (
hydrates. Theoretical considerations <
thermochemistry and vaporization/cor
densation mechanisms helped define th
additives  used in the test program. Th
experimental work was carried out in tw
facilities:  a coal-fired pilot  scale con
bustor, and an intermediate scale con
bustor firing natural gas doped with SO
The experimental results discussed covi
the: 1) relative effectiveness of additive
(sodium, potassium, lithium, cesium, iror
2) method of incorporating the additiv
3) amount of additive, and 4) effect
coal ash. The SO2 removal characterise
of a high  surface area (40 mVg) Ca(OK
material produced using alcohol additk
during hydration are also discussed.
                           Paper:

ALTERNATIVE SO2 SORBENTS

D. O. Ham,
G. A. Simons, and
G. A. Moniz
Physical Sciences Inc.
  Sulfur sorption experiments have be
conducted in a fixed bed reactor in ore

-------
to screen potential mineral sorbents for
use downstream from convective super-
heater tubes. All tests were carried out
using a  simulated combustion effluent
(74.0%  N2,  12.1% C02, 4.6% 02, and
9.2% H20) at  100 kPa pressure, 3000
ppm S02, and in the temperature range
of 400 to 800°C. Four classes of sorbents
have been tested: calcium hydrate, alkali
based (trona), mixtures (alkali promoters),
and mixed cations (shortite and gaylus-
site). Shortite  is consistently  high over
the temperature range. Trona, gaylussite,
and Colton  hydrate  promoted by trona
are all very reactive at 800°C  but much
less  reactive  at  lower temperatures.
Promotion of  Colton hydrate by trona
depends on close physical contact.

SESSION II, PART 3:
SORBENTS —
FUNDAMENTALS

Chairmen, Dennis Drehmel,
EPA, AEERL
                          Paper 2K

EPA STUDY OF HYDROXIDE
REACTIVITY IN A
DIFFERENTIAL REACTOR*

R. H. Borgwardt,
U. S. Environmental Protection
  Agency
K. R. Bruce,
Acurex Corporation
  Laboratory investigations of some of
the  factors influencing the reactivity of
calcium hydroxides with SO2 are reported.
The materials examined included com-
mercial grade Ca(OH)2, pressure hydrated
Ca(OH)2, analytical grade  Ca(OH)2, and
hydroxides prepared  in the  laboratory
from calcined dolomite and limestone. In
the  latter case, the effects of  limestone
calcination conditions (time, temperature)
and hydration technique were evaluated.
The methods of hydration included steam
(105 and 200°C) and water slaking. The
effect of S02 partial pressure on reactivity
was measured at 800°C for three types
of Ca(OH)2 over a  range of 120 to 7000
ppm. Preliminary tests  at  65°C  are
reported for hydroxides exposed to SOz
and high humidity for times ranging from
5 to 120 seconds.

                          Paper 2L

FUNDAMENTAL  STUDIES OF
SORBENT REACTIVITY IN
ISOTHERMAL  REACTORS*
D. W. Pershing,
G. H. Newton, and
D. J. Harrison,
University of Utah
J. A. Cole,
J. C. Kramlich,
W. R. Seeker, and
G. D. Silcox,
Energy and Environmental
  Research Corporation
  Three aspects of sorbent reactivity were
investigated  in  fundamental studies
under isothermal conditions. The first
test series examined the relative reactivity
of a series of limestones, hydrated calcitic
limes, and hydrated dolomitic limes. The
second test series focused on the in-
fluence of particle size on sorbent utiliza-
tion.  Precalcined sized carbonates were
sulfated under isothermal conditions. The
results showed that the limitation to large
stone utilization was the pluggage of the
porous structure at the surface caused by
local overutilization of the sorbent. Below
2-5 Aim, all  portions of the stone are
equally accessible to the S02, and further
size reductions do not enhance utilization.
In the third series, carefully controlled
isothermal experiments were used  to
characterize the CaO + S02 reaction. The
tests were  performed  on sized, pre-
calcined hydrated lime. The results show
a rapid utilization within the first 100
msec. At longer times a reduction in the
sulfation rate was noted which appears
to coincide  with  blockage of the
microporosity.

SESSION III, PART 1:
PROCESS RESEARCH
Chairman, George Offen,
EPRI
                          Paper 3A
A COMPARISON OF
COMBUSTION FACILITIES AND
CALCIUM-BASED SORBENTS
IN TERMS OF THEIR SULFUR
CAPTURE PERFORMANCE*

G. D. Silcox,
R. Payne,
D. W. Pershing, and
W. R. Seeker
Energy and Environmental
  Research Corporation
  Sulfur capture data from combustion
facilities using dry sorbent injection were
compared. The comparisons were per-
formed in order to determine the similari-
ties and differences between facilities
and to understand the effects of various
reaction parameters. Different sorbents
were also  compared in terms of their
capability to capture sulfur. Gaps in data
were identified.
  The data and the comparisons were
interpreted with the help of mathematical
models of  the sulfation reaction and of
the calcination process. The calcination
model shows that the peak surface area
for hydrates occurs just milliseconds after
injection, while that for carbonates is on
the order of 0.2 second. This difference
may explain, in part, the higher reactivity
of hydrates. The sulfation model incorpo-
rates a  nonlinear dependence on SO2
concentration  which is  understood in
terms of a surface reaction  model. The
sulfation and activation models  are
combined and used to explain the compli-
cated dependence of capture on injection
temperature and time.
                          Paper 3B
THERMAL ENVIRONMENT IN A
TANGENTIALLY PULVERIZED-
COAL-FIRED UTILITY BOILER:
COMPARISON OF FIELD TESTS
WITH HEAT TRANSFER
MODELING*
B. M. Cetegen and
W. Richter,
Energy & Environmental Systems
D. G. Lachapelle,
U. S. Environmental Protection
  Agency
  Effective  implementation of the in-situ
desulfurization  process  in  coal-fired
boilers requires a detailed description of
thermal  characteristics of  the  boiler
furnace. This paper discusses the results
of the detailed field tests conducted at a
420 MWe tangentially fired utility boiler
and compares them with predictions
obtained from multi-dimensional furnace
heat transfer models. Results and com-
parisons are reported for two furnace
heat inputs: 1088 MW, (100% load) and
502 MW, (45% load). Sensitivity studies
with the model showed that significantly
accurate predictions of overall thermal
furnace  performance require only the
approximate description of furnace flow,
but a realistic  estimation of the overall
insulation effect of wall ash deposits.

                          Paper 3C

EFFECTS OFINJECTON
TEMPERATURE AND  QUENCH
ON SORBENT UTILIZATION*

-------
R. Beittel,
Southern Research Institute
S. J. Bortz,
KVB, Inc.
G. R. Often,
Electric Power Research Institute
D. C. Drehmel,
U. S. Environmental Protection
  Agency
  Two coal-fired pilot combustors were
used to evaluate the effects of injection
temperature and quench  rate on the
utilization of furnace-injected sorbents.
The quench rate was varied from 38 to
316°C/s by  load  and configuration
changes in one unit, and up to 593°C/s
in the second unit. Limestone and calcitic
and dolomitic hydroxides  were tested.
Injection temperature was found  to be
the primary effect.  Quench rates were
significant only for the calcitic hydroxide
or above 316°C/s;  however, increased
quench rate increased the optimum in-
jection temperature
                          Paper 3D

APPLICATION OF UPPER-
FURNACE SORBENT
INJECTION FOR SO2 CONTROL
IN COAL-FIRED CYCLONE-
EQUIPPED BOILERS

H. Farzan,
L. W. Rodgers, and
G. J. Maringo
Babcock and Wilcox Company
G. R. Often,
Electric Power Research Institute
  Upper furnace sorbent injection for SO2
capture was examined utilizing a 6 x 106
Btu/hr  (1  7  MWt),  cyclone-equipped
boiler simulator The facility  has a con-
vective pass to simulate full-scale time/
temperature histories, including quench
rates  within the sujfation temperature
window (1260 - 871 °C) of approximately
649°C/s.
  Two limestones, one calcitic hydrated
lime, a pressure-hydrated dolomitic lime,
and an atmospheric-hydrated dolomitic
lime were injected at the furnace exit at
1232°C. In-furnace  surface area and
porosity measurements were performed
on the sorbents. No correlation between
sulfur capture and  calcine surface area
was observed. Despite the rapid quench
rate, more  than 50% sulfur capture was
achieved injecting the dolomitic materials
at a Ca/S of 2, independent of the hydra-
tion  process  Sulfur  capture increased
with time due to tube and wall deposits.
The deposits could be removed with in-
creased sootblowing pressure, resulting
in similar recovery of heat flux as obtained
without sorbent.
SESSION III, PART 2:
PROCESS RESEARCH
Chairman, George Often,
EPRI

                          Paper 3E

CONTROLLING POLLUTANT
EMISSIONS FROM COAL AND
OIL COMBUSTORS THROUGH
THE SUPPLEMENTAL USE OF
NATURAL GAS*

S. L Chen,
J.  M.  McCarthy,
M. P. Heap,
W. R. Seeker, and
D. W. Pershing
Energy and Environmental
   Research Corporation
F.  R. Kurzynske,
Gas Research Institute
  Natural gas can be used to enhance
the economic  acceptability of eastern
coals for power generation  under more
stringent environmental regulations. The
sulfur  and nitrogen in coal can lead to
emissions of acid rain precursors such as
SO2 and NOX, while the non-hydrocarbon
components in coal  affect power plant
performance because of  problems as-
sociated with the ash. The application of
technologies which  take  advantage of
the clean burning qualities of natural gas
offers  an attractive  and  cost effective
option for acid rain precursor control. The
study discussed in this paper focuses on
the use of natural gas as a reburning fuel
combined with  sorbent  injection for
combined NOX/SO2 control. The work is
supported by the Gas Research Institute
and builds upon DOE/EPA research ef-
forts. Bench and pilot scale tests have
been performed on the natural gas re-
burning/sorbent injection  process.
Results obtained indicate that 60% NOX
reduction and 70% S02 reduction can be
achieved when 20% natural gas reburning
is combined with hydrate injection (Ca/S
-  2), under realistic time/temperature
furnace profiles. Reburning with natural
gas was found to have several distinct
advantages over reburning with coal. In
addition, this study has demonstrated the
potential of using natural gas to enhance
calcium utilization.
SESSION IV:
MIXING/DISPERSION

Chairman, Nicholas Kresovich,
EPA, AEERL

                          Paper 4A

INFLUENCE OF SORBENT
INJECTION AERODYNAMICS
ON SO2 CONTROL

B. M. Cetegen,
T. R. Johnson,
D. Moyeda, and
R. Payne
Energy and Environmental
   Research Corporation
  Mixing conditions produced by injection
of calcium-based sorbents  into large
boiler furnaces can influence the sulfur
removal efficiency. A practical means of
injecting  pulverized  sorbents into  the
furnace flow is by sorbent-laden gas jets
issuing from nozzles on combustor walls.
This paper describes studies on the trajec-
tory and mixing of sorbent jets with the
cross-flow stream of combustion gases
in the upper furnace region of a boiler
Several aspects of the problem are dis
cussed: bench scale isothermal jet mixinc
experiments in boiler upper furnace con
figurations,  modeling of jet mixing ir
such geometries, and pilot scale combus
tion tests  identifying the effect of mixim
on sulfur  removal. Key parameters con
trolling mixing are identified for furnaci
configurations, and  a simple integra
model based on mass, momentum, an<
energy conservation  is developed fo
predicting jet  injection  characteristics
The sensitivity of sulfur capture to mixini
conditions is presented for a calciur
hydroxide sorbent.

                          Paper 4

SORBENT INJECTOR DESIGN
FOR SULFUR OXIDE COA/TflOi
FROM TANGENTIALLY FIRED
FURNACES BY DRY SORBEN7
INJECTION*

R. W. Koucky,
D. K. Anderson,
P. R. Thibeault, and
J. L Marion
Combustion Engineering, Inc.
C. C. Masser,
U. S.  Environmental Protection
   Agency

-------
  Combustion Engineering, Inc., under
Environmental Protection Agency spon-
sorship, is conducting  a  program  to
develop design criteria for injection of dry
sorbents into tangentially fired furnaces
for the control of S02 emissions. The
program includes aerodynamic cold flow
testing and mathematical  modeling of
sorbent injection, demonstration testing
of SO2 emission control in a 14.6  MW,
(50 x  106 Btu/hr) tangentially fired test
facility, and pilot-scale testing to evaluate
furnace slagging and fouling behavior
during sorbent injection. This paper
describes initial results of the flow testing
and mathematical modeling.  Injector
development testing was conducted in a
large  scale cold-flow model  which ac-
curately simulates  the  geometry  of a
tangentially fired utility boiler including
the pendant heat transfer surfaces in-
stalled in the sulfur capture temperature
window of the  upper furnace. Several
sorbent injector configurations were
tested to optimize injection location and
injection method. The sorbent dispersion
mathematical model uses cold-flow  velo-
city and concentration data to aid in the
selection of injector designs for combus-
tion testing, to extrapolate cold-flow dis-
persion results to hot furnace geometries.

                           Paper 4C

ANALYSIS OF THE SORBENT
INJECTION PROCESS FOR
IN-FURNACE SO2 CONTROL

R. M. Himes and
L. J.  Muzio,
Fossil Energy Research
   Corporation
G. R.  Often,
Electric Power Research Institute
  The  effectiveness of  in-furnace SO2
control  is strongly dependent upon a
number of factors which primarily include
the sorbent material used, the injection
location in the furnace, and the degree of
mixing achieved with  the  combustion
products. The current study  is directed
toward establishing a framework for the
design of an in-furnace sorbent injection
system to optimize the dispersion of the
sorbent in  regions where mixing will oc-
cur under established temperature condi-
tions.  This paper focuses on the use of
empirical  correlations  and cold-flow
modeling to facilitate the design process.
As the empirical correlations have  been
typically developed under the controlled
conditions of a wind tunnel, the cold-flow
model is used to evaluate the correlations
under relatively uniform flow conditions
as well as a more representative furnace
flow field. The simplification of a two-
phase jet into an equivalent gas jet having
a modified density is also evaluated by
comparing the trajectories of single- and
two-phase jets, with identical momentum
ratio found to be the predominant para-
meter  governing the  penetration  and
mixing of the jet under these conditions.
The bulk flow patterns and turbulence
generated under simulated burner flow
conditions,  however, significantly  im-
pacted the trajectory and mixing of the jet
such that the empirical correlations were
no longer valid.
SESSION V: ECONOMICS
Chairman, Dan Giovanni,
Electric Power Technologies, Inc.
                           Paper 5A
CONTROL  COST MODELING
FOR SENSITIVITY AND
ECONOMIC COMPARISON*
N. Kaplan,
D. G. Lachapelle, and
J. Chappell,
U. S. Environmental Protection
  Agency
  Using  computerized  cost  models
developed both internally and under con-
tract to EPA  by  the  Tennessee Valley
Authority over a period of several years,
this paper presents costs for utility retrofit
applications of wet flue gas desulfuriza-
tion (FGD), dry FGD,  and the limestone
injection with multistage burners (LIMB)
process.  Capabilities  of  the  models,
premises, and limits  of the sensitivity
study are discussed. Tabular and graphical
presentations of the results are included.
For a base case of 300 MWe boiler size,
2.2%  sulfur coal, 50% capacity factor,
intermediate sorbent  cost ($71/ton lime
and $15/ton limestone), 1.6 retrofit factor
for FGD systems, and 15 year remaining
unit life, the unit control costs for wet
FGD,  dry FGD, and  LIMB  are $1337,
$1220, and $997/ton of SO2 removed,
respectively. Sensitivity of costs to each
of the above  parameters spanning the
base case is  presented and discussed.
Unit control costs are developed for all
cases, while capital costs and  levelized
annual revenue requirements  are also
presented for the base case.
                           Paper 5B
ECONOMICS OF FURNACE
SORBENT INJECTION FOR SO2
EMISSION CONTROL
 P. A. Ireland,
 G. D. Brown, and
 J. J. Sebesta,
 Stearns Catalytic Corp.
 M. W. McElroy,
 Electric Power Research Institute
  The concept of injecting various cal-
 cium-based sorbent materials into coal-
 fired utility boilers  is currently  under
 development as an option for reducing
 S02 emissions from  new and existing
 power  plants. The  major incentive for
 this development is the anticipated lower
 cost of furnace  sorbent injection com-
 pared to conventional flue gas treatment
 technologies and the potential future need
 for incremental SO2 control  at existing
 units.
  This paper examines the economics of
furnace sorbent injection for  S02 emis-
sion control and the major variables which
 impact  the  overall process  economics.
The economics examined focus primarily
on sorbent injection as a retrofit technique
to control SO2  emissions on  existing
plants. The major variables found to affect
the sorbent injection economics  are
sorbent performance (Ca/S ratio vs SO2
removal), sorbent cost, inflation rate, coal
sulfur content, plant capacity factor, plant
size, and sorbent transportation distance.
Furnace sorbent injection is  also com-
pared to wet FGD on a $/ton SO2 removal
overall control cost basis.

                           Paper 5C

 COMPARATIVE COSTS OF
FLUE GAS DESULFURIZATION:
ADVANTAGES OF FURNACE
INJECTION OF PRESSURE
HYDRATED LIME OVER DRY
SCRUBBING

 M. H. Bobman and
 G. F. Weber,
 University of North Dakota
   Energy Research Center
T. P. Dorchak,
 U. S. Department of Energy
  Costs of pressure-hydrated  lime injec-
tion (PHLI) for the control of  S02 emis-
sions were  developed and compared to
 lime spray dryer absorbers (SDA) oper-
ating under identical design conditions.
 Since 1981, development work,  spon-
sored by the U. S. Department of Energy
at the University of North Dakota Energy
 Research Center (UNDERC), has focused
on PHLI as a cost effective control method,
producing a dry readily disposable waste

-------
product. In the PHLI process, dry sorbent
is injected into the upper furnace region
and the sulfated paniculate collected in a
downstream baghouse  or electrostatic
precipitator. In SDA processes, increas-
ingly used for low sulfur coals, flue gas
S02 contacts lime slurries in large reactor
vessels with subsequent collection of the
dried paniculate in a  downstream bag-
house. Therefore PHLI can have decisive
economic advantage over SDA due to the
elimination of complex absorption vessels
and water chemistry,  and  reactor pres-
sure drop with overall reductions in labor
and water consumption.
  To determine  the possible cost  ad-
vantage of PHLI, costs were developed
for  both PHLI and SDA applied to  two
divergent baseline case scenarios. In the
first baseline case, a new 500 MW plant,
located in the upper Midwest, firing a low
sulfur (1%) lignite, was considered.  A
stringent 90% control target was specified
due to anticipated stricter regulations or
location in a Class I control region. In the
second baseline case, an existing  and
easily retrofitable 300 MW plant, located
in the central Midwest, firing a moderate
sulfur (2%) Eastern bituminous coal,  was
considered. An incremental 50% control
target was specified primarily in anticipa-
tion of acid rain  regulations. The impact
of sulfur content in the coal was assessed
at the 1, 2, and 3% sulfur levels, and for
PHLI, the effect of sorbent utilization rates
was determined  at the 25, 50, and 75%
levels For SDA, the sorbent utilization
rate remained fixed at 80% for all cases
evaluated.
  The evaluations demonstrated  PHLI's
cost-effectiveness over lime SDA  over
the entire range of operating conditions
investigated. Under the worst conditions
for PHLI (i e., operating at the larger  new
plant, firing a  1 % sulfur coal with a  25%
sorbent utilization), PHLI affords a  25%
savings in levelized revenue over SDA,
for the analogous retrofit conditions, PHLI
produces a 35% savings. When sorbent
utilization rises to 75%,  PHLI  savings
increase to about  65% of SDA costs for
both new and retrofit designs In absolute
terms,  capital costs for retrofit PHLI are
low, ranging closely  about $55/kW  at
the 50% sorbent utilization rate, for the
new plant, capital costs are more sensitive
to sulfur content, ranging from $33  to
$62/kW at a 50% utilization rate Under
all  circumstances, operation and main-
tenance costs dominate, due to the  high
sensitivity to sorbent consumption rates,
typically 70% of the total cost for PHLI.
Since the incremental  impact of either
process on paniculate collection equip-
ment is site specific, costs of additional
flue gas handling, baghouses, or precipi-
tators have not been entered  into the
evaluation.
                           Paper 5D

ECONOMICS OF SULFUR
REMOVAL USING "DRY"-
CARBONATED TRONA ORE
FGD SORBENT

J. P. Krieg,
G. W. Mortimer, and
L. W. Weiss,
Cogit Consulting Group
  A  great many potential  commercial
sources of sodium are available for use in
dry injection  FGD systems The sorption
performance of some of these materials
can be improved by processing prior to
injection  Three examples involving raw
trona and dry-carbonated trona  sorbents
are compared  in this study, serving  a
new 500 MW power generating unit and
starting with a selected trona ore f.o.b.
Green River, Wyoming, price of $50/ton
Dry carbonation at the utility site and by a
producer/supplier at  Green River are
evaluated separately against the direct
injection  of  trona ore. The  distance
between the utility site and the  Green
River source of trona ore was considered.
The results show significantly lower first
year operating costs and lower 30-year
levelized costs for dry-carbonated trona,
and this advantage increases  with rail
distance to the power plant from  Green
River. The key factors likely to affect this
outlook are discussed.

SESSION  VI:  POST-FURNACE
SO2 REMOVAL

Chairman,  Richard Rhudy,
EPRI
                           Paper 6A

LABORATORY TESTS ON
FURNACE LIMESTONE
INJECTION AND
DEVELOPMENT OF A SECOND
DESULFURIZATION STEP WITH
FLI ASH THROUGH DUCT
INJECTION

G. Staudinger,
Technical University  Graz
H. Schrofelbauer and
H. Meier,
Osterreichische Draukraftwerke
  AG
  Furnace conditions were simulated in
a laboratory setup by putting single lime-
stone particles with diameters  of 3 to
100 /um into a hot flue gas stream for 3 to
20 seconds. The specific surface area
was  found to increase with  calcination
and to decrease very quickly after passing
a maximum,  the  latter  process being
faster at a higher temperature. Further-
more, single particles were cut into slices
of 2 /jm and analyzed with a raster elec-
tron  microscope, and the sulfur distri-
bution throughout the particle was
measured.  If limestone particles  are
produced from one stone by milling and
fractionation on sieves, the particles with
40 jum diameter will have the smallest
B.E.T. surface area.
  Since the utilization of the Ca in  the
furnace limestone injection (FLI) ash is
only  10 to 15% and because the Austrian
emission  regulations  demand (for  old
plants) desulfurization rates of 80% and a
maximum SO2  concentration  of 800
mg/Nm3 (270 ppm) in the flue gas, simple
FLI will not be sufficient in the future. A
process was developed in which the FLI
ash is separated from the flue gas, milled,
treated with steam and  reinjected into
the flue gas duct upstream from the dust
collector. A pilot plant, which processed a
slipstream of 500 NmVh from a cogen-
eration plant was  operated  for  190C
hours.  The results  are satisfactory:  nc
technical  problems were encountered
One  project for installation of the process
is underway.
  Injection of sodium bicarbonate  intc
the flue gas  duct was thoroughly  in-
vestigated in a 15,000 mVhr pilot plant
more than 80% SO2-removal was founc
with a  stoichiometric ratio (Na2/SO2) o
1 The  retrofit of a  110 MW plant with
this process is under construction.
                           Paper 61
REACTIVATION OF SOLIDS
FROM FURNACE INJECTION
OF  LIMESTONE FOR SO2
CONTROL*
W. Jozewicz and
J. C. S. Chang,
Acurex Corporation
T. G. Brna and
C. B. Sedman,
U. S. Environmental Protection
  Agency
  Six  different  post-furnace injectio
solids and flyash from Hoot Lake Static
                                  8

-------
(Otter Tail Power Company) experiments
have been characterized using scanning
electron microscope (SEM), X-ray powder
diffraction, titration analysis for available
CaO,  and ion chromatography.  The
samples were tested in a bench-scale
reactor for the removal of S02 from a
mixture of N2 and SO2. Virtually no S02
removal was observed with untreated
solids. However, high SO2 capture oc-
curred when samples were hydrated and
dried  prior to S02 exposure. The  S02
capture by solids increased with increas-
ing time and temperature of hydration.
For the same time/temperature condi-
tions  of hydration, higher S02  capture
was achieved with solids of higher flyash/
sorbent ratio. New crystals  (ettringites)
were found in hydrated and dried samples.
The B.E.T. surface area of the solids was
significantly increased by the hydration
process. A possible mechanism of en-
hanced S02 capture by hydration of the
product solids is discussed.
                          Paper 6C

DRY HYDROXIDE INJECTION
AT ECONOMIZER
TEMPERATURES FOR
IMPROVED SO2 CONTROL

S. J.  Bortz,
V. P.  Roman, and
R. J.  Yang,
KVB, Inc.
G. R. Offen,
Electric  Power Research Institute
  Bench  and pilot  testing injecting
hydroxides into temperatures between
427 and 649°C has shown a significant
potential for removing SO2 from a  flue
gas stream. The reaction rates of calcium
hydroxide directly with both SO2 and CO2
at around 538°C are significantly faster
than the reaction of CaO with SO2 or CO2
near 1093°C.  Hydroxide injection, at a
Ca/S =  1-2  into a pilot  combustor
simulating a  typical large  utility boiler
quench rate of 538°C/s between  649
and 427°C, has achieved calcium utiliza-
tions  (predominantly as CaSO3) of 25-
35% with commercially available calcium
hydroxides and 35-45%  with specially
prepared high reactivity hydroxides.
  The utilization levels achieved in  250
ms residence time near  the optimum
injection temperature of 538°C appear to
be  determined  mainly  by the initial
porosity of the hydroxide and the ratio of
CaS03/CaC03 formed  during  the re-
action. Sorbent preparation methods that
increase the  initial hydrate porosity by
 1.5 over a normal hydrate have been
 identified and found to also yield a greater
 ratio of CaSO3/CaC03. Injection of these
 special hydrates resulted in a  60% in-
 crease in sorbent  utilization  (to 45%)
 compared to commercial hydrates.

                           Paper 6D

 FUNDAMENTAL STUDIES OF
 LOW-TEMPERATURE SULFUR
 CAPTURE BY DRY CALCITIC
 SORBENT INJECTION

 W. R. Seeker,
 S. L Chen, J. C. Kramlich,
 S. B. Greene, and
 B. J. Overmoe,
 Energy and Environmental
   Research Corporation
  Three experimental studies were per-
 formed to develop fundamental informa-
 tion on low-temperature or duct capture
 by dry injection of calcitic or dolomitic
 sorbents. These were performed for the
 Department of Energy's Pittsburgh Energy
 Technology Center. The first experiment
 simulated the surface of a baghouse upon
 which sorbent had collected. Fresh
 sorbent was exposed to varying concen-
 trations of S02 and water vapor flowing
 through a fixed bed. A bench-scale re-
 actor  was used to define the relative
 contribution of dispersed phase and filter
 capture. Finally, a  baghouse connected
 to a  pilot-scale furnace  was  used to
 observe  baghouse behavior when the
 sorbent  was  collecting  during the
 experiment.

                           Paper 6E

 SORBENT IMPROVEMENT AND
 COMPUTER MODELING
 STUDIES FOR COOLSIDE
DESULFURIZATION

 H. Yoon,
J. A.  Withum,
W. A. Rosenhoover, and
 F. P.  Burke,
 Conoco Coal Research Division
  Coolside desulfurization  is a retrofit
S02  control technology for  coal-fired
utility stations. It consists of dry sorbent
injection  and flue  gas  humidification
downstream of the air preheater. Boiler
lime/limestone injection (BLI) differs in
that  the  sorbent is injected  into the
furnace.  Field  tests  have  shown  that
Coolside  and BLI/humidification  can
achieve SO2 removals of 80 and  65%,
 respectively, from high sulfur coal flue
 gases. For both processes, sorbent utiliza-
 tion is  the  most significant factor  in
 process economics. In addition, both pro-
 cesses benefit from flue gas humidifica-
 tion because it enhances the SO2 capture
 reactions and the paniculate removal ef-
 ficiency if, as is commonly the case, the
 station is equipped with an electrostatic
 precipitator (ESP). This paper describes
 laboratory studies of methods to improve
 Coolside and BLI sorbent utilizations
 through changes in  sorbent properties
 and additives, and the  use  of sorbent
 recycle. This paper also describes model-
 ing studies conducted to determine nozzle
 and humidifier  design  and operating
 parameters,  and to identify key design
 issues for further development.

                           Paper 6F

 RESULTS OF 1.0  MM BTU/
 HOUR TESTING  OF HALT
 (HYDRATE ADDITION AT LOW
 TEMPERATURE)  FOR SO2
 CONTROL

 M. Babu and
 R. C. Forsythe,
 Dravo Corp.
 C. Runyon,
 Ohio Edison Company
 E. Evans,
 U. S. Department  of Energy,
 Pittsburgh  Energy Technology
   Center
 J. Thompson,
 Process Calx, Inc.
  SO2 removal efficiencies of 60-70%
upstream  of the baghouse can be
achieved for a 3.2%  S coal  using the
HALT process at a Ca/S of 2.0. Quality
control of the hydrate and uniform dis-
tribution of the sorbent across the duct
cross section is essential. Direct contact
water spray cooling of the gas with the
hydrate feed entering upstream gives high
removals. Removal efficiency with down-
stream injection  is more  sensitive to
approach to adiabatic saturation  and
Ca/S. Cold hydrate feed increases re-
moval efficiency.  Addition  of NaOH to
spray water is beneficial. Of the hydrates
prepared with additives to  hydration
water, methanol-sugar formulation per-
formed the best. Long term testing on a
larger scale is required to determine the
effect of approach to adiabatic saturation
temperature  on removal,  incremental
removal due  to  baghouse/ESP, and

-------
factors related to improving equipment
availability. These will be covered on a 5
MW pilot test unit program to follow.
                          Paper 6G
EPRI PILOT TESTING OF SO2
REMOVAL BY CALCIUM
INJECTION UPSTREAM OF A
PARTICULATE CONTROL
DEVICE
Q. M. Blythe,
Radian Corporation
R. Smith,
Fossil Energy Research
  Corporation
M. McElroy and
R. Rhudy,
Electric Power Research Institute
V. Bland and
C. Martin,
KVB, Inc.
  EPRI is developing a potentially  low-
cost FGD process that involves dry injec-
tion of calcium-based reagents directly
into humidified flue gas upstream  of a
paniculate  control  device. Currently,
process  development tests are being
conducted on a 2000 acfm (0.9 amVs)
pilot fabric filter. Tests using an ESP are
also planned. Pilot-scale tests are being
supported by laboratory-scale screening
of the effects of reagent type  and  pre-
paration  techniques. Future  process
development efforts will include process
economics, an engineering evaluation of
the feasibility of applying the process to a
larger scale, and process modeling.
  Initial process development tests have
been  completed, and results  are  pre-
sented.  These  tests confirm  that the
important  performance variables are
Ca/S  ratio  and approach  to  adiabatic
saturation Both high calcium and dolomi-
tic  hydrates have  been studied.  The
results also indicate that the use of addi-
tives,  the use of high  specific surface
area reagents,  and  sorbent reinjection
may promote greater S02 removals and
increased sorbent utilization.
SESSION VII: SYSTEM
IMPACTS
Chairman, Richard Stern,
EPA,  AEERL
                           Paper 7A
PARTICLE PROPERTIES
RELATED TO ESP
PERFORMANCE WITH
SORBENT INJECTION AND
GAS  CONDITIONING
J. L DuBard,
J. P. Gooch, and
R. Beittel,
Southern Research Institute
S. L Rakes,
U. S. Environmental Protection
  Agency
G. R. Often,
Electric Power Research Institute
  Processes employing dry sorbent  in-
jection for S02 control require additional
particulate collection capability in existing
electrostatic precipitators (ESPs). Both
EPA and EPRI are sponsoring research
with the following  objectives  1) to
quantify the impact on ESP performance
of various process plans, and 2) to develop
performance upgrading strategies de-
signed to provide cost-effective particulate
control within the constraints imposed by
existing plant configurations
  This paper provides results from pilot-
and laboratory-scale  measurements of
ash/sorbent particle properties related to
ESP  performance. The measurements
were made on particulate matter gen-
erated in a 106 Btu/hr (293 kW) combus-
tor fired on high-sulfur coal, with hydrated
lime injected as a sorbent. The measure-
ments include: 1) dust resistivity obtained
in situ with a point-plane probe, 2) particle
size distributions obtained with  cascade
impactors, and 3) the effects of increased
fine  particle loadings of ESP  voltage-
current relationships.  Water vapor and
SO3 were each evaluated as conditioning
agents  in situ and in the laboratory  A
mathematical model was used to project
ESP performance with flue gas condition-
ing, over a range of SCA values  up to 50
mVmVs.

                           Paper 7B

ELECTROSTATIC
PRECIPITATOR UPGRADES
FOR FURNACE SORBENT
INJECTION

D. J. Helfritch,
P. L. Feldman, and
B. Weinstein,
Cottrell Environmental Sciences
M. W.  McElroy,
EPRI
  The injection of calcium based sorbents,
such as limestone or calcium hydroxide,
directly  into the  furnace of coal-fired
boilers is currently under development as
an option  for reducing SO2  emissions
from existing power plants. The sorbents
react with SO2 ' i form calcium sulfate
The calcium compounds can increase the
oarticulate loading to a downstream ESP
oy a factor of 2 or 3 and can increase the
resistivity of the particulate by orders of
magnitude. Methods which can be used
to restore  ESP  performance  to pre-
sorbent injection levels are evaluated with
respect to feasibility, effectiveness, and
:ost  Methods which are evaluated in-
clude increased collection surface area,
flue gas cooling by water evaporation,
cooling by heat exchange, pulse energiza-
tion, and SO3 conditioning.
                           Paper 7C
DESIGN CONSIDERATIONS
FOR WASTE MANAGEMENT
SYSTEMS FOR ADVANCED SO,
CONTROL  TECHNOLOGIES

A. G. Ektund and
C. M. Thompson,
Radian Corporation
D. M. Golden,
Electric  Power Research Institute
H. Ness,
Department of Energy
  This paper is a  review of waste char
actenstics likely to affect waste handlmc
systems design determined in separate
studies of  a)  calcium-based  furnac<
sorbent injection wastes, and b) sodium
based  dry scrubbing wastes Chemica
composition,  leaching  properties, an<
physical characteristics were determine!
Dy standard methods Areas of  concert
that were identified for the calcium-basei
processes  include' special  handling re
quirements for powdery, alkaline material
high heats of hydration due to unreactei
lime, alkalinity of resultant  leachates
large volumes of waste per weight, an<
rapid,  flash setting  upon  exposure  t
moisture  Characteristics important fc
utilization potential for the calcium-base
wastes were identified  as follows: po;
zolanic capability, high surface for re
action if used as reuse sorbent, and hec
generation upon  hydration.  For th
sodium-based processes, the solubility <
the sodium compounds and lack of coh«
sive strengths were the principal source
of waste management concern.  Reger
eration of sodium compounds from tr
dry wastes  may be  a potential  area fi
utilization of by-products from  sodiun
based dry scrubbing systems.

                          Paper 7

ANALYSIS OF LIMB WASTE
MANAGEMENT OPTIONS*

-------
R. S. Dahlin and
C. L. Lishawa,
Southern Research Institute
N. Kaplan,
U. S. Environmental Protection
  Agency
  This paper presents the results of
laboratory  testing  and  engineering
analyses intended to assess various ap-
proaches to  the  management of ash/
sorbent  waste products from  the LIMB
process. The feasibility of conventional
landfill disposal  was assessed through
leachate analysis (pH  and  metals),
permeability measurements,  and  re-
activity (temperature-rise) studies. Based
on these tests, landfill disposal appears
feasible  if the runoff is neutralized  and
the temperature rise is controlled. Lab-
oratory tests showed that LIMB ash could
oe mixed with water and pelletized to
make a  product  suitable for use as ag-
gregate  in asphaltic wearing  courses.
Pilot combustor studies showed that LIMB
ash could be recycled by hydration  and
remjection into the furnace These tests
yielded sulfur captures similar to those of
commercial hydrated lime  when com-
pared on the basis of equivalent available
calcium  Several ash/sorbent separation
techniques were evaluated including' use
of the existing particulate collector as a
separator,  steam  treatment for simul-
taneous hydration and separation, inertial
separation, magnetic separation, and froth
flotation The economics  of various  dis-
posal, utilization, and recycle  schemes
are being analyzed using modified ver-
sions of the EPA LIMB cost and recycle
models
                            Paper 7E
APPLICATION OF SPRAY DRY
PRODUCTS AT THE
SAND-LIME PRODUCTION
P. J. C. Bloem and
B. J. G. Sciarone,
N V KEMA
  The application of a fly-ash/spray-dry
product  mixture was examined. There-
fore,  sand-lime   samples  with various
quantities of different spray dry products
were prepared on laboratory scale Auto-
claving  was  performed for 8  hours at
187°C
  After  that  the  compressive  strength,
the splitting tensile strength, the absorp-
tion  coefficient at capillary action,  and
the porosity  were determined. Further
examination by means of SEM, EDX, and
X-ray diffraction were done.
  It appeared that applying fly ash and
spray-dry product as a partial substitution
for sand leads to encouraging results, if
the quantity of spray-dry product in the
fly-ash/spray-dry product mixture is 30%
or less. The best results  are obtained if
the ratio of sand and fly-ash/spray-dry
product mixture is about 1.0.
                            Paper 7F

TVA'S WASTE MANAGEMENT
PROGRAM FOR AFBC ASH
MATERIALS-UTILIZATION/
DISPOSAL

J. L. Harness,
M. W. Milligan, and
K. A. Cruikshank,
Tennessee Valley Authority
  The Tennessee Valley Authority (TVA)
is involved actively in the development of
the atmospheric fluidized-bed combustion
(AFBC) technology for use by the  utility
industry. As part of this effort, TVA has
instituted a program to develop methods
for the handling, disposal, and/or utiliza-
tion of the unique ash materials produced
by the AFBC process. Discussed will be
the operating and maintenance experi-
ence for the ash handling and disposal
systems of the TVA 20 MW AFBC pilot
plant, the design and development work
for those same systems for the 160 MW
AFBC demonstration plant, and the re-
search and development work underway
on  utilization options for the  ash
materials.

SESSION VIII:
COMMERCIAL SCALE
APPLICATIONS

Chairman, Michael  McElroy,
EPRI

                           Paper 8A

WALL-FIRED LIMB
DEMONSTRATION DESIGN
FOR OHIO EDISON'S
EDGEWATER UNIT 4*

R. V. Hendriks,
U. S. Environmental Protection
   Agency
P. S. Nolan,
Babcock & Wilcox
M. L Kunkle,
Ohio Edison
  The paper describes and discusses key
design features of the retrofit of EPA's
Limestone Injection  Multistage Burner
(LIMB) system to an operating, wall-fired
utility boiler at Ohio Edison's Edgewater
Station.  It further describes results of
pertinent projects in EPA's LIMB program
and shows how these results were used
as the basis for the design of the system.
The full-scale demonstration is expected
to prove the effectiveness and cost of the
LIMB concept for use on large-scale utility
boilers. Final engineering drawings are
now being prepared and the equipment
procured to install the LIMB system on
the  Edgewater boiler. System start-up is
scheduled for March 1987.
                           Paper 8B

RESULTS  OF DIRECT
DESULFURIZATION FROM
COMMERCIAL POWER PLANTS

M.  Y. Chughtai,
L & C. Steinmuller GmbH
  One possible method of sulfur removal
is direct desulfurization sulfur capture
within the boiler itself. The alkalis which
naturally occur in the fuel already capture
a certain proportion of the S02 produced
during combustion and this sulfur capture
can be increased by the injection of suit-
able additives, such as CaC03, Ca(OH)2,
CaC03.MgCO3, and Ca(OH)2.Mg(OH)2, into
the furnace. If the additive employed is
Ca(OH)2 or CaC03/ the initial reaction in
the furnace  is the dehydration or decar-
bonation of the additive according to the
following equations:
    Ca(OH)2

    CaC03
CaO + H20

CaO + C02
  Capture of S02 within the furnace itself
 and in the downstream parts of the boiler
 •s affected by the exothermic hetero-
 geneous gas/solids reaction
          S02+1/2O2
   •CaSO4
  The calcium utilization ratio, which cor-
 responds to the proportion of  injected
 calcium actually reacting with the S02, is
 the figure which indicates the success or
 otherwise of a process For the desul-
 furization process to take  place at  all,
 certain preconditions of reaction tempera-
 ture and residence time must  be met.
 Depending  on the firing  concept,  the
 additive  injection  location  has to  be
 selected to achieve a rapid mixing and a
 proper time/temperature history of  the
 additive in the flue gas.
  Desulfurization results from different
 power plants with different firing  systems
 will be presented along with the side

-------
effects of additive injection on the boiler
and its ancillary equipment.
  The results achieved so far show that a
desulfurization efficiency of around 60%
at a Ca/S of 2 is possible  with Ca(OH)2.
No side effects on power plants are known
so far which  could  cause operational
problems. Extensive investigations on ash
disposal  have shown that the ash from
direct desulfurization is definitely suitable
for land disposal.
  The economic  advantages of direct
desulfurization are considerable. Additive
consumption and disposal of the  ash
represent the  greatest operation costs.
Operating costs are a little bit higher than
for wet flue gas desulfurization  (FGD),
but the plant costs are only about 10% of
those for the latter. The  economic ad-
vantages with direct desulfurization  in-
crease as the sulfur content of the coal
and the size of the plant decrease. This
process would seem to be of particular
interest for older power stations where,
for reasons  of space, it is  impossible to
install FGD equipment but where, in most
cases, it  is relatively easy  to retrofit the
plants for direct desulfurization.
                          Paper 8C
FIELD EXPERIENCE  WITH
COMMERCIAL FURNACE
LIMESTONE INJECTION
G. Staudinger,
Technical University of Graz
H. Schrofelbauer,
Osterreichische Draukraftwerk
   AG
  In Austria, eight boilers, ranging from
20 to 330  MWe, having tangential or
wall burners are equipped with furnace
limestone injection. The fuel is local or
imported brown coal.
  Operating experience is  1 to 3 years.
  Ca/S molar ratios of 3 to 4 give average
sulfur removal rates  of 50 to 60%,
depending on the fuel as  well as on the
type and size of the furnace.
  Some minor problems of operation were
overcome. Surprisingly, the performance
of the electrostatic precipitators  was
significantly improved in  several instal-
lations. In one particular case the solids
emission was reduced from  180 to 80
mg/Nm3 flue  gas by furnace limestone
injection.
  Other operating data and experiences
are reported.
                          Paper 8D

THE TAMPELLA LIFAC SO2
REMOVAL PROCESS
T. Kenakkala,
Tampella Ltd.
M. Suokas and
J. Hautanen,
Tampere University of Technology
  Following successful laboratory and
pilot plant tests, a full-scale LIFAC desul-
furization plant was installed at the 250
MWe  coal-fired boiler of Inkoo  power
plant. Because of simplicity and suitability
of the LIFAC method, the plant was oper-
ating after 6 months.
  By means of the LIFAC process a con-
siderable improvement of the reduction
results can be achieved as compared to
limestone injection alone. The test results
show that by the LIFAC process as high a
degree of desulfurization can be reached
as by the wet scrubbing method  or the
semi-dry process.
  This paper includes results from pre-
vious pilot tests as well as full-scale test
results which have been achieved during
the last few months.

                           Paper 8E

DEMONSTRATION OF FRONT
END ALKALI INJECTION AND
COMBUSTION MODIFICATION
TECHNIQUES AT BOUNDARY
DAM G.S. UNIT 4- PROGRESS
REPORT

R. Feindel,
C-E Canada Power Systems
D. Gass,
Saskatchewan Power Corporation
  This report describes the field experi-
ments which evaluated the effectiveness
of a calcium  carbonate and calcium
hydroxide injection system in addition to
combustion system modifications for the
simultaneous  reduction of S02 and NOX
emissions. The combustion system modi-
fications consisted of the addition  of both
overfire  air and concentric firing to the
demonstration unit's  tangential  firing
system. The effects of this alkali injection
system  with  combustion  modification
techniques on the fouling and slagging
behavior of the furnace were also deter-
mined. Preliminary experimental  results
are reported  for the  injection of two
sorbents, calcium carbonate and calcium
hydroxide, while firing two Saskatchewan
lignites in a 476,190 kg/hr (1.05 x 106
Ib/hr) Combustion Engineering  steam
generator. The demonstration plant, con-
sisting of the alkali injection system and
boiler modifications, is described.
                           Paper 8F

DESIGN AND TESTING OF A
RETROFIT DRY SORBENT
INJECTION SYSTEM ON A 300
MW PULVERIZED FUEL BOILER

D. Cameron and
J. A. Arnott,
Ontario Hydro
  Ontario Hydro, as  part of its overall
commitment to reduce S02 emissions,
undertook full  scale development and
testing of a dry sorbent furnace injection
system. The injection system was retro-
fitted to a 300 MW tangentiafly fired
pulverized fuel  boiler with electrostatic
precipitators at the Lakeview Generating
Station in Toronto.
  Development  of the  overall design
parameters  and system design of  the
sorbent injection equipment, in coopera-
tion with Combustion Engineering anc
GATX-Fuller, occurred from  late  1984
until the December 1985 test period.
  This paper describes the Lakeview Uni
4 boiler, associated equipment, and the
retrofit injection system. Powdered lime
stone or calcium hydroxide can be injectec
into the furnace at a preselected elevatior
above the burners.
  The testing procedures, equipment, an<
test results are described.
  The results of the program  indicati
that 40% S02 reduction can be achieve!
at partial loads (up to 220  MW) witl
limestone injected at a 3 to  1 Ca/S ratic
  Results of calcium hydroxide injectioi
show greater S02 reductions with lowe
Ca/S ratios.
                          Paper 8(

FULL SCALE DEMONSTRATE
OF DRY SODIUM INJECTION
FLUE GAS DESULFURIZATION
AT CITY OF COLORADO
SPRINGS RAY D. NIXON
POWER PLANT
D. W. Albin and
J. J. Hammond,
FMC Corporation
D. B. Watts and
R. L. Ostop,
Colorado Springs Department o
  Utilities
R. G. Hooper,
Electric Power Research Institut
  The Department of Utilities of the Ci
of Colorado Springs, Colorado,  FM
Corporation, and  the  Electric Pow
                                72

-------
Research Institute are conducting a full-
scale demonstration and test program of
the injection of dry sodium sorbents up-
stream of a fabric filter for SO2 and NOX
removal The program was conducted on
the Unit  1  boiler at the Ray D. Nixon
Power Plant in Fountain, Colorado. This
boiler is  rated at 223 MW, fires  low-
sulfur western coal, and is equipped with
two  reverse-gas-cleaned  fabric  filters,
each designed to handle 50% of the total
flue  gas  flow. Sodium-based  sorbents
are injected upstream of one of the fabric
filters thus providing a 100+ MW demon-
stration, side-by-side with an  identical
fabric filter  operating without sorbent
injection
  The program consisted of three phases.
Phase I was the initial start-up and char-
acterization of the system. Phase 2 was
the demonstration of 70% S02 removal
on a 30-day rolling average basis using
sodium sesquicarbonate as the sorbent.
Phase 3 consisted of parametric testing
using  sodium  sesquicarbonate  and
sodium bicarbonate as sorbents, as well
as testing the process while burning a
different coal in the boiler
  The  full-scale demonstration proved
that dry injection of sodium sesquicar-
bonate can consistently and reliably meet
the current  federal  regulation of  70%
S02  removal.  During  the  55-day  con-
tinuous run to demonstrate 70% S02
removal efficiency on a 30-day  rolling
average, the system averaged 74% SO2
removal with better than expected utiliza-
tion of the  sodium  sesquicarbonate
sorbent. In addition, an unexpected bene-
fit in the operation of the fabric filter was
realized: the injection of sodium sesqui-
carbonate resulted in significantly re-
duced flue gas pressure drop Distribution
of sorbent in the flue gas stream was
shown to be important in achieving maxi-
mum sorbent utilization.
                           Paper 8H

EVALUATION Of SORBENT
INJECTION TEST AT
TIEFSTACK,  WEST GERMANY*

R. Payne and
J. Reese,
Energy and Environmental
   Research Corporation
Y. Chughtai,
L & C Steinmuller GmbH
  Sorbent injection for S02 control was
evaluated at the Tiefstack Power Station,
Unit 6, located  near  Hamburg,  West
Germany The boiler  is tangentially fired
with low sulfur bituminous coal and pro-
duces 160,000 kg/hr (352,000 Ib/hr) of
steam. A  temporary sorbent injection
system was installed and operated by the
utility owner. The evaluation involved field
measurements over a 3-week test period
combined  with  flow and  heat transfer
modeling and small scale testing. Results
of the field test showed S02 reductions of
22  and 43% at a Ca/S  of 2 for  high
calcium  limestone   and  atmospheric
hydrated lime, respectively.  Analysis of
the results showed the injection  con-
figuration  to have minimal  impact on
capture, with injection at  a furnace gas
temperature of 1520 K and adequate
dispersion of the sorbent.
UNPRESENTED PAPERS

PILOT SCALE STUDIES OF
LIMESTONE INJECTION
PROCESS
M.  S. Mozes,
R. Mangal, and
R. Thampi
Ontario Hydro Research Division

  The  limestone injection process for
SO2 control has been studied at Ontario
Hydro's 620 MJ/h combustion research
facility under conditions simulating the
Lakeview TGS quenching rate. Various
combinations of operating conditions af-
fecting S02 and particulate removal have
been evaluated using a non-porous local
limestone, while burning a 1.7% S eastern
U S. bituminous coal. The  program is
aimed at improving the overall efficiency
of sulfur capture and sorbent utilization,
and minimizing the impact of in-furnace
injection on particulate removal  by the
electrostatic precipitator (ESP).  Process
variables such  as limestone injection
location, addition rate, and particle size
have been  optimized. Under optimized
conditions,  SO2 removal efficiency was
about 43% and sorbent utilization up to
17%. The process produced finer and
about twice as  much ash as coal. The
carbon level of the waste  on a sorbent-
free basis was about 20%. Resistivity of
the waste  ranged between 105 and  106
ohm. cm. The ESP performed with a 10%
reduced efficiency, probably because of
the fineness of the waste and increased
dust loading.
FLUE GAS DESULFURIZATION
BY DRY ALKALINE SODIUM
SORBENT INJECTION USING
AN ELECTROSTATIC
PRECIPITATOR
T. W. Sonnichsen,
J. H. Jansen Company, Inc.
M. J. Pilat and
J. C. Wilder,
University of Washington

  Pilot scale tests of flue gas desulfuriza-
tion  using the dry injection of sodium
sorbents in combination with an electro-
static precipitator  have been conducted
at the University of Washington. The pilot
facility draws flue  gas from a pulverized-
coal-fired boiler using low sulfur western
coal. An extended  duct arrangement has
been used to provide a residence time of
up to 6 seconds prior to collection in the
ESP.
  Test results indicate SO2 removals  of
50 to 80% using  pulverized sodium bi-
carbonate at a normalized stoichiometric
ratio of 1.0,  pilot gas temperature  of
204°C, and 300 to 550 ppm  S02. Re-
movals were affected by sorbent injection
rate,  in-stream  residence  time,  and
sorbent particle size. Sodium sesquicar-
bonate appears to be less effective than
sodium  bicarbonate  in removing  S02
under test conditions.
  Essentially  all research and develop-
ment activities conducted prior to this
program involved  the application of dry
injection  in  combination with a fabric
filter baghouse. Accumulation of sorbent
material on the baghouse surface  was
believed to be necessary to provide an
extended contact  and reaction period.
The results of the  University's pilot tests
indicate the potential for application  of
the dry injection technology to both  new
and existing coal-fired boilers  equipped
with ESPs. S02 removal rates of 50% and
greater would appear to be possible. Ad-
ditional research will be required to more
completely evaluate the potential of this
technology  for higher sulfur coals and
varied flue gas temperatures. It is antici-
pated that more efficient SO2 removal
will be achieved with higher flue gas SO2
concentrations.
  The economics of retrofit installation of
the dry  injection process  to typical in-
dustrial and utility boilers has been briefly
examined. In  all cases, installed capital
costs for dry  injection are significantly
below those of available alternate tech-
nologies such as spray dryers and wet
scrubbers. Operating costs are strongly
dependent on the  price and quantity  of
sorbent used. Levelized busbar costs for
dry injection are significantly below those
of alternative technologies for low sulfur
coals. Levelized costs are comparable for
small to intermediate sized utility boilers
                                                                              13

-------
with relatively low capacity factors firing
medium (2%) sulfur coals.

RATE CONTROLLING
MECHANISMS OF SULFATION

Girard A. Simons,
Physical Sciences, Inc.
  A  sulfation model is  used to analyze
the S02 sorption data from a wide range
of CaO particle diameters (1 /^m - 1 mm),
SO2  partial pressures (60 Pa to  5 kPa),
and  temperatures (973 to 1478 K)  An
intrinsic SO2 reaction order of unity  has
been validated through fixed bed experi-
ments  in the 60 to 300 Pa  S02 partial
pressure range, while fluidized bed data
were used to validate the  reaction order
from 500 Pa to 5 kPa SO2 partial pressure.
The  absolute value of the intrinsic  rate
constant  and  the  activation  energy
(10,000 ± 1000 K) was derived from data
on small (1 nm) sorbent particles that are
in the limit of kinetic control. Rate  limiting
processes include' 1} the filling of  the
smallest pores and the  subsequent  loss
in the  internal surface  area,  2)  the
activated diffusion of the S02 through
the produce  deposits, and 3) the loss of
mtraparticle diffusion due to the complete
plugging of the largest pores. It is shown
that the reaction kinetics and the filling of
the smallest pores are generally rate con-
trolling  on time scales of 1  to 10s, but
product layer diffusion may become  rate
controlling after approximately 100s.
The EPA  author R.  Stern  (also the EPA Project Officer, see below) is with
  the Air and Energy Engineering Research Laboratory, Research Triangle Park,
  NC 27711; and G. Often  is with the Electric Power Research Institute, Palo
  Alto, CA 94303.
The complete report consists of two volumes, entitled "Proceedings: 1986 Joint
  Symposium on Dry S02 and Simultaneous SOz/NO* Control Technologies:"
  "Volume 1. Sorbents, Process Research, and Dispersion," (Order No. PB 87-
  120 465/AS; Cost: $30.95)
  "Volume 2. Economics, Power Plant Integration, and Commercial Applica-
  tions," (Order No. PB 87-120 457/AS; Cost: $48.95)
The above reports will be available only from: (costs subject to change)
       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

-------
                                     8? a,
                                     £?§
                                       8
                                       w
                                       8



o
   5

-------