United States
                  Environmentaf Protection
                  Agency	
Air and Energy Engineering
Research Laboratory
Research Triangle Park, NC 27711
                  Research and Development
EPA/600/S7-88/001  Apr. 1988
&ERA         Project  Summary
                  Fundamental Combustion
                  Research Applied to Pollution
                  Formation: Volume II.
                  Physics and Chemistry of
                  Two-Phase Systems
                  W.R. Seeker and M.P Heap
                    As a part of EPA's Fundamental
                  Combustion  Research (FOR)
                  Program, eight studies were
                  conducted by various investigators
                  to better understand the physics and
                  chemistry of two-phase combustion
                  with respect to pollution formation.
                  Reports from  these  studies  are
                  grouped into three sections, bound
                  separately, which comprise Volume II
                  of the FCR report series. Volume Ha
                  describes mechanisms of fuel
                  nitrogen processing in large-cale
                  utility flames burning pulverized coal
                  and heavy fuel oils, in three parts: (1)
                  The  High Temperature  De-
                  composition  and Combustion  of
                  Pulverized Coal  (EERC); (2) Detailed
                  Measurement of Long Pulverized
                  Coal Flames for the Characterization
                  of Pollutant Formation (International
                  Flame Research Foundation; and  (3)
                  Heavy  Oil Combustion  in  Shear
                  Layers (United Technologies
                  Research  Center). Volume lib gives
                  information on the influence  of
                  various combustion parameters (i.e.,
                  fuel type, stoichiometry,  residence
                  time, temperature and mixing) on the
                  fate of volatile fuel nitrogen, in four
                  parts: (1) Method for Characteriza-
                  tion of  Coal  During Thermal
                  Decomposition (United Technologies
                  Research Center); (2) The Volatility of
                  Fuel Nitrogen Species (Rockwell); (3)
                  Stirred Reactor Study of  Pollutant
                  Formation from Combusting
 Pulverized Coal Clouds (Acurex) and
 (4)  Pollutant Formation from Com-
 bustion of Fuel Oils in a Well-Mixed
 Combustor (Battelle).  Volume  lie
 gives information on the kinetic rates
 and mechanisms  of nitrogen oxide
 (NO) reduction on solid particles, in
 three parts: (1) Trials on NO and NH3,
 Alumina, Quartz, Graphite, and Soot;
 (2)  Comparative Study of Coal  and
 Char; and (3) Trials on Fly Ashes and
 Study  of  Reactions Involving
 Hydrogen Cyanide (Institut Francais
 du Petrole).
  This  Project Summary was
 developed by EPA's  Air and Energy
 Engineering Research Laboratory,
 Research  Triangle Park,  NC, to
 announce key findings of the research
 project  that is fully documented in
 three separate volumes of the same
 title (see Project Report  ordering
 Information at back).

 Introduction
  The EPA has conducted a wide range
 of  research projects under  its
 Fundamental Combustion Research
 Applied  to Pollution Formation (FCR)
 program. The principal goal of the  FCR
 program was to  generate  the
 understanding of combustion behavior
 necessary to  help develop control
 strategies to minimize nitrogen oxide
 (NOX) emissions from stationary sources.
 Because of the increased use of coal and
 residual  fuel oils, the dominant stationary

-------
source of NOX for the foreseeable future
is the conversion of nitrogen in the fuel.
This observation  implies a  need  to
understand  the various  phenomena
associated with burning fuels containing
high levels of nitrogen.
  The FCR  program involved program
planning, management, and synthesis of
the  overall program, and subcontracting
to various organizations separate projects
included  within program  elements.
Subcontracts accounted for 70% of the
program and were  used to ensure that
FCR had the benefit of the best scientific
talent available. Every effort was made to
use subcontractors  who  had the
necessary experience  and, in  many
cases, equipment  to produce  quality
results in a short time frame.
  A  wide  variety  of  combustion
phenomena can have a  direct impact on
the  NOX emission rate  from  pulverized
coal and residual oil fired boilers. The
FCR program  structure,  described  in
Figure 1,  was  established to  allow for
investigation into phenomena believed to
be critical to  meeting the program goals.
Three  major  program  areas  were
augmented by  two  program support
areas. The major program area.Transport
Process in Reacting Systems allowed  for
studies into the  ballistics  and  fluid
mechanics of bringing fuel and air into
contact  with each  other  and into the
transport of the reacting mixture  under
boiler  conditions.  The  major program
area,  Physics and  Chemistry of Two-
Phase  Systems,  is  devoted  to
understanding  the  decomposition and
processing of solid and  liquid fuels as a
function of the reactive environment and
local thermal conditions. The Gas  Phase
Chemistry major  program  area was
provided to  better  understand the gas
phase reactions leading  to NO formation,
particularly  reactions  involving
nitrogenous  species  from   fuel
devolatilization. The two program support
areas, Analytical Tool Development and
Measurement Systems were provided to
assist in performing experiments and
analyzing data. This program structure is
designed to  permit studies into  a wide
variety of  areas  influencing  NOX
formation  and  control.  It also permits
focusing individual studies leading to two
major program outputs:
 -  A description of the chemical  limits
    on  NO  production  in  order  to
    ascertain the lower bounds of both
    fuel  and thermal  NO  production
    under a series of process constraints
    which were not  limited in any way by
    fuel/ air contacting.
 -  A description of fuel NO formation in
    turbulent  diffusion flames for gas,
    liquid.and pulverized coal systems.
   Documentation of research performed
under the  FCR program  is contained in
four basic  volumes and a special report.
Each  volume  (or volume series  reports
on the work  accomplished  in a  major
program area or  program support area.
Volume I  (EPA-600/7-85-048) covers
work  in  the Gas-Phase  Chemistry
program  area.  Volume II  addresses
Physics and  Chemistry  of  Two-Phase
Systems, Volume III  is devoted to the
FCR effort in Measurement Systems.and
Volume IV is concerned with  Analytical
Tool  Development  and  Transport
Processes. In addition, a special  report
(EPA-600/7-85-048) has  been  issued
to document the  proceedings  of  a
workshop held by FCR contractors.
   Most of the FCR effort was devoted to
the Physics  and Chemistry  of  Two-
Phase Systems  program  area  which
consisted  of  eight individual  research
projects. Accordingly, Volume  II  has
been  subdivided into  three separately
bound components denoted  as Volumes
Ha, lib, and He.


Physics and Chemistry of
Two-Phase Systems
   The critical processes which must be
addressed to understand the  interplay
between  two-phase  combustion  and
pollution formation are listed in Figure 2.
The processes are displayed in  relative
position against a  photograph  of  an
actual  turbulent  pulverized-coal
diffusion flame.  The  ability  to address
these processes  represents  the
benchmark of the  accomplishments of
the FCR program in this major program
area.  The  processes, and, therefore, the
important areas of research, are:
Ignition. The ignition of flames influences
    not only  stability, which  is  an
    extremely  important operating
    parameter, but also premixing  which
    can strongly  impact fuel nitrogen
    conversion.
Physical  Processes  During Devola-
    tilization.  The mechanism  by  which
    volatiles are  released from the coal
    matrix can dramatically influence a
    wide variety of phenomena including
    temperature  response,  ignition, the
    environment  in  which  volatile
    nitrogen  will  react,  and  soot
    formation. Very  little is known  about
    how pulverized  coal  or fuel  oil
    actually decomposes in flames.
Nitrogen  Devolatilized.  Bench-scale
    tests indicate that the conversion 4
    volatile  nitrogen  is  the  largest
    contributor  to  total fuel  nitrogen
    conversion to NO. Also, the volatility
    of fuel nitrogen  is strongly tied to the
    fuel type and  the  time/temperature
    history of the  particular  application
    therefore, it is  imperative  that  the
    volatility  of  fuel nitrogen  be
    understood if  the conditions  fo
    minimum fuel  nitrogen  conversior
    are to be generalized.
Species Evolved and  Fraction  of  Mas;
    Devolatilized.  The  amount  o
    hydrocarbon  species  evolved  from
    the  fuel is expected to  impact  th<
    local stoichiometry, soot formation
    and the  local temperature;  also,  th<
    impact   of   fuel   type  am
    time/temperature  history  is not we
    understood. Of particular interest i
    the form the nitrogenous compound
    and the light and  heavy hydrocarbo
    yield.
Temperature Response. The temperatur
    response of the  fuel particles  c
    droplets influences  almost  a
    aspects of  the   pollution  forme
    sequence. Little is known about wh;
    factors  influence the  particl
    temperature and how that paramete
    in turn, affects other parameters.
Soot  Formation.  Soot formation  i
    important because it is a  primai
    pollutant itself, it can act as a smoli
    limit to how  rich the primary  stac
    can be  taken,  and it can  act as
    source of carbonaceous  material c
    which fuel nitrogen species can t
    reduced to  molecular  nitrogen. Tt
    mechanisms  of soot formation fro
    pulverized  coal and  fuel  c
    combustion have  as  yet not  be<
    explored. Little  is even known  abo
    the  amounts of soot formed in sui
    flames.
Char NO  Reduction.  NO can react wi
    heterogeneous  particulate  to I
    reduced to  molecular  nitrogen. T
    degree  to which  this occurs  f
    flame-formed particulate, and  unc
    what conditions, is not known.
Char Nitrogen Conversion  to NOX. T
    traditional staging Iow-N0x  conc<
    addresses only the volatile  nitrog
    species. The  conversion of  t
    nitrogen that remains in  the char
    NO thus represents the ultimate lii
    that can be  achieved  using stag
    combustion.  The  conversion of i
    remaining char nitrogen needs to
    explored if this  limit is to be lower
    Further, questions remain concern
    how  the  first-stage  processi

-------
  Major Program Areas
Transport Process in
 Reacting Systems
Physics and Chemistry
    of Two Phase
      Systems
                                                                        Gas Phase Chemistry
                                                                             Reactor Flame
                                                                              Experiments
Program Elements
Program Support Areas
                                                      Kinetic
                                                     Mechanism
                                                    Development
     Analytical Tool '•'•
     Development
     Measurement
       Systems
           Heterogeneous
           NO Reduction
                                             Description of Fuel
                                               NO Formation
                                                in Turbulent
                                              Diffusion Flames
                                                              Chemical Limits of
                                                                NO Production
                                                                    /VO» Control
                                                                     Strategy
 Figure 1.    FCR-I program structure.

-------
     Ignition
      Physical
     Processes
       During
    Devolatilization
                                                                   Char Burnout
                                                                       Time
                                                                                       Char Nitrogen
                                                                                        Conversion
                                                                                          to NO,
                                             Char NO
                                            Reduction
Fraction of Mass
  Devolatilized
  Soot
Formation
                                 • Fuel Effects
                                 • Influence of Time/Temperature/Stoichiometry History


 Figure 2.    Critical processes influencing the fate of fuel nitrogen in pulverized coal flames.
    influences the subsequent character
    and reactivity of the char. This can
    be an important aspect of whether
    the  char will burn  out within the
    combustor volume.

  A program plan was  developed to
both determine  what questions  were
necessary for  the  development of
multimedia  control  technologies for
stationary sources and to address  those
questions. It was  not possible to address
every issue,  but a focused  program
evolved  which was directed toward the
important issues of  relevance to the
Combustion Research  Branch  (CRB) of
EPA in order for the  results to  be
applied. This major  program  element
(i.e., the  physics and chemistry of two-
phase combustion) consisted  of  eight
well-defined priority  target  areas,and
      the  ensuing  research effort was guided
      toward the specific problem of the area.
        The target areas and  the  individual
      projects which made up these areas are:
      Flame  Combustion  Processes (Volume
      Ha)
      The High  Temperature   Decomposition
            and Combustion of Pulverized
            Coal - EER
      Detailed  Measurement  of  Long
            Pulverized-Coal  Flames  for the
            Characterization  of  Pollutant
            Formation -  International  Flame
            Research Foundation
      Heavy Oil Combustion  in Shear Layers
            - United Technologies Research
            Center

      Devolatilization and Volatile Reactions
      (Volume Mb)
             Method  for  Characterization  of Coal
                   During Thermal  Decomposition  -
                   United Technologies Research
                   Center
             The Volatility of  Fuel Nitrogen Species -
                   Rockwell
             Stirred  Reactor  Study  of  Pollutant
                   Formation  from  Combusting
                   Pulverized  Coal  Clouds  -  Acurex
             Pollutant Formation from Combustion of
                   Fuel  Oils  in  a Well-  Mixed
                   Combustor - Battelle
             Heterogeneous  NO Reduction (Volume
             lie)
             Mechanisms of Nitric Oxide Reduction on
                   Solid  Particles - Institut Francais
                   du Petrole
               The first  program  area (Volume  I la)
             consists  of programs  to define  the
             physical  and  chemical  processes

-------
occurring in large  scale  utility  flames
 lurning  pulverized  coal  or  heavy fuel
oils.  The  major   emphasis  is  on
understanding the  mechanisms  of fuel
nitrogen processing in flames  of this
type, developing  a  data base for model
development,and  exploring scale  effects.
The second area (Volume Mb) considers
only the  devolatilization  and early
combustion processes. The programs in
this  area  were designed to yield
information  on  the influence  of
combustion parameters such  as fuel
type,  stoichiometry,  residence time,
temperature, and degree  of mixing on
the fate of volatile fuel nitrogen. The final
area (Volume lie)  provides information on
the kinetic  rates and mechanisms  of
heterogeneous NO  reduction  for  typi-
cal combustion-generated  particulate,
which  are  necessary  to assess  the
relative importance of heterogeneous
reduction processes.

Volume Ha - Flame
Combustion Processes
   This  volume  consists  of final
documentation on three individual efforts
dealing  with  flame   combustion
processes.

High Temperature
Decomposition and Combustion
of Pulverized Coal (EER)
   Despite  the fact that  coal has been
burned for  decades and  has  been the
subject of study  for nearly as long, little
is actually  known  about the physical
processes occurring to the individual coal
particles  and their  close environment.
The purpose of this study was to fill this
data gap  by  directly observing  the
microscale  phenomena that occur around
individual coal particles as they  burn in
an environment  simulating  large-scale,
turbulent diffusion flames.
   The experimental facility employed
was a downfired  flat-flame burner with a
centerline port for injection of pulverized
coal particles.  The  methane/air fired
flat-flame  burner  produced   a
background of hot  reaction products. As
the pulverized coal  particles  entered this
environment, they were ignited and
produced  a  long  flame which was
remarkably similar to many  practical
boiler flames.
   A variety  of  optical  and  intrusive
measurement techniques  were em-
ployed to  examine  the microscale
phenomena associated with  the  burning
coal particles. Both holography and high
speed  shadowgraph photography  were
used  to visualize the  burning particles.
Two color optical pyrometry was used to
measure  the  particle temperature,and
sodium  D  line reversal  was  used  to
measure gas  temperature.  Extractive
sampling was employed to determine the
local gas chemical composition and to
examine  the  morphology   of the
particulate.
   Figure 3 presents  a  mosaic  of flame
observations  and  indicates   spatial
variation of surface morphology  and the
local environment of the burning coal
particles.  Photographs   from  re-
constructed holograms (on the righthand
side of Figure  3)  indicate that  the
devolatilizing coal particles can  produce
a cloud of soot particles which are seen
to collapse into  long  fiberous   shapes.
These results were the  first experiments
to indicate  the  formation of  either the
soot cloud  or  the long structures. The
experimental  program includes trial
series examining coal type, initial particle
size,  particle injection  velocity and
background  gas temperature and
stoichiometry.


Detailed Measurements of
Long-Pulverized Coal Flames
for the Characterization  of
Pollutant Formation (IFRF)
   This  study utilized state-of-the-art
measuring techniques to characterize in
detail a number of long pulverized coal
flames under well-defined  input and
boundary conditions. These experiments
were relatively large scale, having been
conducted in a 1.9 x 1.9 m cross section
furnace fired at the 2 MWt level.
   Measurements  in  these   flames
included the determination of local gas
and solids compositions and concentra-
tions, temperatures, and  radiative  and
aerodynamic  characteristics.  The mea-
surements were of  sufficient detail  for
use in validation of predictive  procedures
both for  global combustion  processes
and for specific  phenomena relating to
coal particle combustion.
   The flames  for  all  six  trials  were
produced  with a very simple double
concentric,  parallel-flow burner  system.
This burner  system  generates long
flames which are far-field  dominated
and where the heat release is protracted.
Such flame  shapes exhibit the real
characteristics  of certain  classes of
furnaces  flames  such as cement kilns
and  possibly  tangentially fired  boilers.
Further,  the  far-field  domination
significantly simplifies  the  problem of
mathematically  simulating the  process.
Thus, the experimental results   reported
represent  an  initial  data base  for
developing a verified model of pulverized
coal  combustion in  a practical  furnace
environment.
  The six long coal flames investigated
represent variations  due to  air  and fuel
input velocities and temperatures. These
variations have a direct impact on  the
flame mixing  characteristics and on  the
coal  particle heating rate. Complete flow
field  mapping data are presented in both
graphic and tabular  form. Despite  many
difficulties encountered in measuring in
such flames, particularly for sampling in
regions of  nigh temperature and  solids
concentrations, the  consistency between
the  various measurements  is generally
good. Indications are  that globally  the
combustion is  controlled by mixing and
that  the  influence  of  fuel kinetics  is
basically the  same  in  all cases.  In
particular, NHa concentrations are always
very  low, while HCN was found to form
very  rapidly and in large amounts  More
detailed evaluation  of  these data is  still
required  for a quantitative interpretation
of fuel nitrogen conversion to HCN and to
NO.


Heavy Oil Spray Combustion in
Shear Layers (UTRC)
  This program consisted of two phases:
(1)   design  and  fabrication of  an
apparatus,   and  (2)  preliminary
combustion experiments A unique  shear
layer mixing and composition apparatus
was  developed  which  simulates high
shear, diffusive combustion with heavy oil
droplet injection as found in  the  near-
field  combustion  zone  of  boilers and
furnaces. The apparatus, shown in Figure
4, provides:

  (l)Two-dimensional  mixing   and
    combustion of uniform streams of air
    and  rich  combustion products  fed
    from  either side of a splitter plate.
  (2) Uniform injection of  a monodisperse
    heavy oil spray into the shear layer
    by means of  a linear array of fuel
    injectors beneath the combustor wall.
  (3)Control of droplet path, residence
    time, and extent of vaporization.
  (4) Control of hot and cold stream inlet
    properties.
  (5)Three-dimensional  probing of  the
    flow field.

The  design  phase  of  the  program
consisted of two tasks: aerodynamic and
thermodynamic design,  and mechanical
design. A wind tunnel,  hot gas generator,
combustion chamber, droplet  injection
system,  and  phase-sensitive sampling
probe were designed initially. The design

-------
                        CH'4/'Air Main
                          Burner Flow I
    Coal and
  ''Transport Air
   CHt/Air
Dilution Flow

~a25J"^Water Cooled
              Sintered
             Copper Disk
                                                                        Holograms
                                                                           A/ote; / in. = 2.54 cm
Figure 3.    Coal flame visualization.

-------
                                                • Perforated Plates
                                                 60% Open

                                                          Honeycomb Flow
                                                          Straighteners
 Air Inlet
                                                                                                  Movable Injector Box
       Fuel Manifold
                                                                                              Splitter Plate
                                                            Vibrator
                                 Droplet Injector
Figure 4.   Heavy oil spray combustor.
of the apparatus  was  influenced most
strongly by the diverse requirements of
the fuel injection  system. The injector
was designed so that the angle of droplet
penetration into the shear layer could be
varied over  a wide  range.   Droplet
trajectories were computed for vaporizing
No. 6 oil droplets  injected from  the floor
of the  combustor into  a hot,  rich
environment  provided  by a  hot  gas
generator. Hot gas generator dimensions
and injector locations  were selected to
accommodate the required  droplet
trajectories, provide complete  vapori-
zation if desired,  and  provide  sufficient
hot soak time for substantial conversion
of volatile fuel nitrogen compounds to
molecular  nitrogen. The design  of a low
turbulence wind tunnel,  combustion
chamber,  and  phase  sensitive probe
completed this task.
   Due to program time constraints, only
initial system verification  tests  were
performed during  this effort.  The testing
which was  performed consisted of (1)
verification  of uniform properties of the
hot and cold streams at the splitter plate
trailing edge, (2)  demonstration of  high
injection velocities and uniform droplet
size and penetration with No. 2, 5, and 6
oils in vibrating  multicapillary  injectors,
(3) determination  of stable flow regimes
as a function of primary and secondary
stream velocities,  (4) flame stability and
flashback  studies, (5)  high speed
cinematography of the  shear layer with
injection of No.  2  and 6 fuel oil, and (6)
preliminary flame probing.
Volume 11 b - Devolatilization
and Volatile Reactions
   This  volume  consists  of  final
documentation of four efforts dealing with
devolatilization and  reaction  of  volatile
components.  As  such,  the  studies
concentrate on  the early flame  zone
processing of high nitrogen fuels.

Method for Characterization of
Coal During Thermal
Decomposition (UTRC)
   This  study  is a  continuation of an
earlier effort to develop a model of the
coal devolatilization  process. Basically,
the model assumes  that a coal  particle
can  be  represented  as  a series  of
functional chemical  groups  which  are

-------
split  into  tar-forming  and  non-tar-
forming fractions. The functional groups
include carboxyl,  hydroxyl, two  ether,
non-volatile,  aliphatic,  aromatic
hydrogen,  and carbon as  shown  in
Figure 5.
   The evolution of each component into
a gas phase (carboxyl into COg, aromatic
hydrogen into H2, etc.) is represented by
a first order function of the form Y(i)  =
Y°(i)exp(-Kjt).  There  are two  ether
components  since  CO  is  typically
evolved in two distinct steps. Values for
the initial coal composition representation
are derived from quantitative infrared
measurements, ultimate  analysis,and
pyrolysis data for the particular fuel.
   When  coal particles thermally
decompose, a substantial portion of the
initial  mass is evolved as a tar. Infrared
analysis of these  tars shows  that  their
spectra are very similar to those of the
parent coal,  except that  the tar  has
enhanced  aliphatic  content.   The
evolution  of the tar is represented by a
first-order  diminishing of the tar forming
fraction according  to X  =  X°exp(-Kxt).
   The kinetic rate constants KC and Kx
are   represented  by  Arrhenius
expressions derived  from  experimental
data.  The  thrust of the current program
was  to expand the  model to  include
nitrogen   evolution  and  to  examine,
theoretically and experimentally, a series
of coals being  used  in other  EPA  NOX
control development  programs. The ex-
perimental studies included vacuum and
1-atmosphere thermal decomposition
studies. The vacuum experiments were
performed in a heated grid facility which
allowed the coal sample to be heated at
a controlled  rate to a specific  final
temperature.  Fourier transform infrared
analysis was used to monitor the volatile
yield  and speciation as  a function  of
time. The 1-atmosphere  experiments
were  performed in an isothermal furnace
and allowed for particle heating rates one
to two orders of magnitude greater than
in the heated  grid  trials.  These  latter
tests included examinations  of  the
various coals as well as  studies of the
high temperature thermal decomposition
of the tars.
    The various results obtained in this
study provide insight into the thermal
decomposition process but also indicate
the need  for significant modification to
the  basic  hypotheses of the model.
Specifically, it  was  observed that the
volatile product distribution and evolution
rates are  sensitive functions of physical
parameters such  as thermal drive, final
temperature, and initial  particle  size.
Thus, variations  in product distributions
are not simply the result of variations in
the functional  group  mix of the parent
coal. An initial attempt to modify  the
model is presented as part of this report.
The basic  result, however,  is  that
additional  theoretical  and  experimental
studies are  required to  produce  a
generally  acceptable model  of  coal
decomposition.


The Volatility of Fuel Nitrogen
Species (Rocketdyne/Rockwell)
    This  FCR  program  task  is  a
continuation of an earlier EPA  sponsored
program  where  a   two-stage  inert
pyrolysis reactor was developed to study
the evolution of fuel  nitrogen species.
That earlier study showed  that reactive
volatile nitrogen evolved from  the fuel in
the  first-stage  pyrolysis  process is
converted  to  HCN in  the second stage
when the second stage is  held  at about
1400 K. This two-stage reactor tech-
nique,  therefore, permits  a simple
measurement  of  the  reactive  nitrogen
yield from fuel thermal decomposition.
    The thrust of this task  was to
examine samples of various fuels being
used in  other  CRB-sponsored  NOX
control development programs. A total of
19 fuels (including residual  oils, coal
derived liquids and coals) were subjected
to the  two stage pyrolysis process,and
the  HCN  yield  was  compared to  NO
production from controlled combustion
tests.For  the seven  coals tested,  an
encouraging  correlation  was  developed
relating the HCN yield from pyrolysis to
the  percent conversion of fuel bound
nitrogen to NOX  in the combustion tests.
This correlation,  shown in  Figure 6,
generates  several concerns   about its
generality and fundamental basis, but the
results do  indicate a  possible approach
for relating NOX  production potential to
the chemical characteristics of the fuel.


Pollutant Formation from
Pulverized Coal Clouds and Fuel
Oils
     The third and the fourth  tasks were
to  develop  coal- and oil-fired  well-
stirred reactors to simulate  the early
processing  of  fuels in   commercial
boilers.   The  coal-fired  reactor
development program report  is entitled,
Stirred Reactor  Study  of   Pollutant
Formation  from Combusting  Pulverized
Coal Clouds  (Acurex). The  oil-fired
reactor development  program report  is
entitled,   Pollutant  Formation  from
Combustion of  Fuel Oils in a  Well-
Mixed Combustor (Battelle).            |
    The major portion of both tasks was"
devoted to developing and characterizing
the facilities. In each  case,  cold  flow
models were first constructed to evaluate
the basic   design  concepts  and to
demonstrate the ability to produce  a high
degree  of   stirring  in the  two-phase
environment. Next,  reacting versions of
the facilities were  constructed, and ex-
tensive  characterization experimentation
performed.  These  tests  revealed
problems with each  facility which  casts
uncertainty on the  interpretation of
experimental data.  For  the  coal-fired
reactor,  the gas phase appeared  to be
well  stirred, but the solid phase  had
much more of  a plug flow  character.
Further, a  strong hysteresis  effect was
associated with the fuel injectors. For the
oil-fired  facility,  a significant  fraction of
the fuel spray impacted the furnace walls,
thereby  distorting the early fuel reaction
process. Limited combustion testing was
performed  in  both  programs,  but the
above noted experimental difficulties cast
doubt on data interpretation.

Volume lie •  Heterogeneous
NO Reduction
    This report documents a  study
performed  at  the  Institut Francais di
Petrol into the heterogeneous  reaction o
NO (and other nitrogenous species) or
solid-phase  material typically found ir
industrial   flames. This extensive
systematic  study was performed  in  <
packed-bed  reactor maintained a
controlled  temperatures.   Meterei
quantities  of gaseous  reactants  flowe<
through the   bed,  and the  outle
composition was monitored as a functioi
of time and reactor conditions.
    The program involved systematicall
varying  the complexity of  the  be
material, beginning  with  alumina, an
building to  soot and coal char. The ink
composition of  the reacting  gases wa
also systematically varied,  beginning wit
CO/argon  mixtures. Hydrogen and  the
hydrocarbons  were added. As regard
nitrogenous species, the  study include
NO, Na, NHa, and HCN as input reactai
gases.  The experimental  approac
included experiments with each mixtui
(gas-  and  solid-phase)  of   sufficiei
detail to extract global, intrinsic reactic
rates and overall reaction order.
    This experimental program,  thouc
massive in scope, should  be  considere
an initial screening  exercise  to  identi
what heterogeneous reaction paths are
potential importance to defining the N

-------
                                       Gas
                                                                        Char
                 a) Functional Group Composition of Coal
                         bj Initial Stage of Decomposition
                        Non-Tar Forming
                         Fraction (1-X°)
Tar Forming
Fraction X°
                    c) Later Stage of Decomposition
                                         W//////,
                                                    X
                                                           •H2O~

                                                           •CO-
                                                             Etc




                                                                  xiaeoeooeaeoeoQeofleoflooaaaeflt
                        d) Completion of Decomposition
                                                                                              ft
                                                                                              \
          Figure 5.    Basic coal devolatilization model.
emission  level from  pulverized- coal-
fired  experiments.  Of  particular
significance is the sequence of reactions
involving  HCN and NHa  on  soot. As
noted in the earlier  discussion the study
of coal thermal decomposition (Volume
lla),the early devolatilization  process is
often accompanied by the formation of a
cloud of soot particles. This environment
(see  Figure 3) closely  resembles  a
packed-bed reactor and  represents  a
field of solid material through which  the
coal volatiles must pass. The  intrinsic
rate  data  developed  indicate that
heterogeneous  reactions of HCN  and
NHa with soot may lead  to formation of
molecular nitrogen.

-------
        50
        40
    in
    3
    •Q


    O
o

o
        30
        20
        10
                                           BC19
                              % to NO= -0,013 + 1.99 (% to HCN)
                                                           I
                       5          10          15          20         25

                          Percent Fuel-N to HCN in Pyrolysis at 1100°C


Figure 6.    Percent NO in combustion vs. percent HCN in pyrolysis at 1100°C.
                                              *U.S.Government Printing Office: 1988 — 548-158/67113
                                         10

-------