U.S. Environmental Protection Agency Industrial Environmental Research     EPA~600/7~77"0733
    Office of Research and Development laboratory                . .  j *»-»-»
                    Research Triangle Park, North Carolina 27711 Jllly 1977
            PROCEEDINGS OF THE SECOND
            STATIONARY SOURCE
            COMBUSTION SYMPOSIUM
            Volume I. Small Industrial,
            Commercial, and Residential
            Systems
            Interagency
            Energy-Environment
            Research and Development
            Program Report
                 LIBRARY.
                 U.S. E:rv7;-; ",*.. ...iT-iL.
                 EMSGH, H.J.. i
EP 600/7
77-073a


-------
                        t ••!
•5* 9
                       RESEARCH REPORTING  SERIES
Research reports of,the Office of  Research  and Development, U.S.
Environmental Protection Agency, have been  grouped  into seven series.
These seven broad categories were  established  to  facilitate further
development and application of environmental technology.  Elimination
of traditional grouping was consciously  planned to  foster technology
transfer and a maximum interface in related fields.  The seven series
are:

     1.  Environmental Health Effects Research
     2.  Environmental Protection  Technology
     3.  Ecological Research
     4.  Environmental Monitoring
     5.  Socioeconomic Environmental  Studies
     6.  Scientific and Technical  Assessment Reports  (STAR)
     7.  Interagency Energy-Environment  Research  and Development

This report has been assigned to  the INTERAGENCY  ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series.   Reports in.this series result from
the effort funded under the 17-agency Federal  Energy/Environment
Research and Development Program.   These studies  relate to EPA's
mission to protect the public health and welfare  from adverse effects
of pollutants associated with energy systems.  The  goal of the Program
is to assure the rapid development of domestic energy supplies in an
environmentally—compatible manner by providing the necessary
environmental data and control technology.  Investigations include
analyses of the transport of energy-related pollutants and their health
and ecological effects; assessments of,  and development of, control
technologies for energy systems; and integrated assessments of a wide
range of energy-related environmental issues.

                            REVIEW NOTICE

This report has been reviewed by the participating Federal
Agencies, and approved for publication. Approval does riot
signify that the contents necessarily reflect the views and
policies of the Government, nor does mention of trade names
or commercial products constitute endorsement  or recommen-
dation for use.
This document is available to the public through  the  National Technical
Information Service, Springfield, Virginia  22161.

-------
                                        EPA-600/7-77-073a
                                                July 1977
          PROCEEDINGS OF THE SECOND
                STATIONARY SOURCE
             COMBUSTION SYMPOSIUM
              Volume I. Small  Industrial,
        Commercial, and Residential Systems
-o
                     Symposium Chairman Joshua S. Bowen
                        Vice-Chairman Robert E. Hall

                       Environmental Protection Agency
                      Office of Research and Development
                    Industrial Environmental Research Laboratory
                    Research Triangle Park, North Carolina 27711
                        Program Element No. EHE624
                      EDISOH, M.J*. pc;317, -
                           Prepared for
                    U.S. ENVIRONMENTAL PROTECTION AGENCY
                      Office of Research and Development
                         Washington, D.C. 20460

-------

-------
                                   PREFACE
     These proceedings document the more than 50 presentations and discus-
sions of the Second Symposium on Stationary Source Combustion held August
29 - September 1, 1977, at the Marriott Hotel in New Orleans, Louisiana.
Sponsored by the Combustion Research Branch of the EPA's Industrial
Environmental Research Laboratory-Research Triangle Park, the symposium
presented the results of recent research in the areas of combustion
processes, fuel properties, burner and furnace design, combustion
modification, and emission control technology.

     Dr. Joshua S. Bowen, Chief, Combustion Research Branch, was Symposium
Chairman; Robert E. Hall, Combustion Research Branch, was Symposium Vice-
Chairman and Project Officer.  The Welcoming Address was delivered by Dr.
John K. Burchard, Director of IERL-RTP; the Opening Address was delivered
by Robert P. Hangebrauck, Director, Energy Assessment and Control Division,
IERL-RTP; and Dr. Howard B. Mason, Program Manager NOX Environmental Assessment
Program, Acurex Corporation, delivered the Keynote Paper.

     The symposium consisted of six sessions:
     Session I:


     Session II:


     Session III:


     Session IV:



     Session V:


     Session VI:
Small Industrial, Commercial and Residential Systems
Robert E. Hall, Session Chairman

Utility and Large Industrial Boilers
David G. Lachapelle, Session Chairman

Special Topics
David G. Lachapelle, Session Chairman

Stationary Engine and Industrial Process Combustion
Systems
John H. Wasser, Session Chairman

Advanced Processes
G. Blair Martin, Session Chairman

Fundamental Combustion Research
W. Steven Lanier, Session Chairman
                                       m

-------

-------
                                   VOLUME I
                              TABLE OF CONTENTS
Opening Address:  "Stationary Source Combustion Control --
Environmental Engineering R&D at IERL-RTP," R. P.  Hangebrauck
Page


   1
Keynote Paper:  "Environmental Assessment of Stationary Source
NOX Combustion Modification Technologies," H. B. Mason, L. R.
Waterland   .	      37


     - SESSION I:  SMALL INDUSTRIAL, COMMERCIAL AND RESIDENTIAL SYSTEMS -

"Design Optimization and Field Verification of an Integrated Resi-
dential Furnace," L. P. Combs, A. S. Okuda  . .	      85

"Performance of a Thermal Aerosol Oil Burner," 0. E. Janssen, J .0.
Glatzel, U. Bonne	     123

"Effects of Fuel and Atomization on NOX Control for Heavy Liquid
Fuel-Fired Package Boilers," M. P. Heap	     163

"NOX Control Techniques for Package Boilers:  Comparison of Burner
Design Fuel Modification and Combustion Modification," 0. E.
Cichanowicz	     165

"Evaluation of Emissions and Control Technology for Industrial Stoker
Boilers," R. D. Giammar	     167

"Field Testing:  Application of Combustion Modifications to Control
Pollutant Emissions from Industrial Boilers -- Phase II," D. R.
Bartz, S. C. Hunter	     207

"Field Tests -- Stoker Fired Industrial Boilers as  Specified by
ERDA," B. C. Severs	     247

-------

-------
                   OPENING ADDRESS
STATIONARY SOURCE COMBUSTION CONTROL — ENVIRONMENTAL
             ENGINEERING R&D AT IERL-RTP
                         By:

                  R. P. Hangebrauck
   Director, Energy Assessment and Control Division
   Industrial Environmental Research Laboratory-RTP
       Office of Energy, Minerals and Industry
          Office of Research and Development
           Environmental Protection Agency
          Research Triangle Park, NC  27711

-------

-------
                                   ABSTRACT
     The purpose and outline of lERL-RTP's programs in the area of combustion
assessment and control are presented, including environmental assessment and
control technology development aspects.  The energy (combustion) technologies
covered are pointed out and the fuel use picture for these indicated.  The
applicable control approaches, including existing clean fuels, fuel cleaning,
combustion modifications, and flue gas cleaning, are defined and examples of
the emissions affected by these control .approaches are given with emphasis on
the air emission aspects.  Each of the combustion control areas is discussed
in terms of program activities and needs.  Trends affecting the program, as
well as some of the challenges, are discussed.

-------
                              ACKNOWLEDGEMENTS
     The author acknowledges the direct input and/or the availability of
information developed by IERL-RTP personnel relating to program activities
in the Utilities and Industrial Power Division (Bill Plyler, Director), and
the Energy Assessment and Control Division (Branch Chiefs Josh Bowen,
Kelly Janes, and Pic Turner).  Also individual comments as input were pro-
vided by Josh Bowen, Bob Hall, Bill Rhodes, Wade Ponder, Bruce Henschel, and
Stan Cuffe/John Copeland.

-------
                    WHERE ARE THE CHALLENGES?—CONCLUSIONS

     There are a lot of R§D challenges to the environmental engineer looking
ahead at the energy/environmental area.  Figure 1 lists a few key ones which
relate to the status/trends of existing energy technologies and environmental
engineering methodology development/evaluation.  These are complex areas;
however, I do believe we are on target in working toward the best environmental
tools and focusing them on reality, in terms of the combustion/energy technolo-
gies which are here now and here to stay, and those that will shortly be in
greater use.

     Something soon becomes obvious to an environmental engineer working in
this area.  There are only a limited number of generic types of control systems
available for application to a variety of stationary combustion sources, and
while a particular control system may be aimed primarily at one type of pollutant,
it invariably affects others.  It is also soon obvious that combustion efflu-
ents are complex, multi-pollutant streams.  It is further obvious that methods
for evaluating the true impact of complex effluents are in the developmental
stage.  The need for such technologies (environmental assessment methodology)
is of tremendous importance in doing the most economical job of applying the
limited number of control approaches that we have, in combination, to produce
truly the most environmentally effective results.
     For example, consider the use of bioassays for effluent testing as an
advanced environmental assessment tool.  If coupling application of such
techniques to the development of a combustion modification control for an
oil-fired furnace results in eliminating the mutagenicity of particulates
emitted and reduces the overall toxicity of the effluent by a factor of 10,
or even to an undetectable level, we will have taken a considerable step
forward in achieving true control effectiveness.

-------
                        PURPOSE AND OUTLINE OF PROGRAM

     lERL-RTP's program in combustion control is aimed at early definition
of environmental problems and development of control technology.  It is also
a program aimed at identifying and/or providing for environmental alternatives
as they relate to control approaches.  There are two key activities in this
program:

     0 Environmental assessment
     0 Control technology development

These two activities go hand in hand and are inseparable.  Environmental assess-
ment can be thought of not so much as the determination of the environmental
acceptability of a particular energy process, as compared to some other energy
process, but as the determination of the effective reduction in environmental
emissions for one control option or one set of control options compared with
other options.  Thus what is really being evaluated is the environmental effec-
tiveness of control or disposal options.

     A more complete definition of environmental assessment is shown in
Figures 2 and 3.

     Control technology development on the other hand includes specific develop-
mental activities which may be undertaken for one or more control approaches
as depicted in Figure 4.  These activities include basic and applied research
and development (R&D), engineering analysis, and specific control process
development/evaluation.
     The central purpose behind the whole program is R£D support of the develop-
ment of needed multi-media, multi-pollutant environmental regulations by identi-
fying processes or sources for which regulations could be set, the techno-
logical options available for waste stream prevention or control, and potential
new regulatory approaches for complex effluents.

-------
     Most combustion technology sources covered in the combustion-related
part of lERL-RTP's program are indicated in Figure 5.
     This paper, because of the orientation of our Symposium, is focused
more on discussion of the air emission-control aspects of the control program
areas shown in Figure 6.

-------
                           PRESENT FUEL USE PICTURE
     Figure 7 provides a summarized picture of U. S. stationary source fossil
fuel consumption for 1974.  ^  A look at the totals reveals that 39 percent
of the consumption was gas, 29 percent oil, and 32 percent coal.  It also can
be discerned that over 57 percent of all fuel was combusted in other than
utility boilers.  The emphasis on increased coal use in the future will show
up substantially in the utility boiler usage category as a practical means
for supplying energy now and in the future to residential, commercial, and
industrial sectors.  Electrical energy has the environmental advantage of not
creating additional pollutants during its utilization as an alternate to fuel
burning in area type sources.  Results of an EPA study published in March 1976,
"EPA-600/2-76-049 a/b, Electrical Energy as an Alternate to Clean Fuels for
Area Sources," are of interest here.
                                    (2)
                                        8

-------
                         APPLICABLE CONTROL APPROACHES
     What are the control approaches applicable to combustion sources?  Most
of us have some degree of familiarity with the options, and Figure 8 is
a listing of some of the main ones.   Figure 9 gives an overview of the appli-
cability of the general control approaches as they relate to categories of
conventional combustion systems and advanced combustion systems.  As can be
seen, a considerable variability exists in the applicability of the general
control approaches.

-------
                   EMISSIONS AFFECTED BY CONTROL APPROACHES

     For the purposes of this discussion, most of the air pollutants from
combustion processes can be included in the two broad classes as indicated
in Figure 10; that is, Inorganics and Organics.  Each of these classes can be
further broken into two groups:  Gaseous and Particulate.  Figure 10 also
includes a brief list of some pollutants of concern under each of the groups
indicated.
     Perhaps the next question to ask is, "What classes of pollutants are
controlled by the generally applicable control approaches?"  It is desirable
that control approaches be effective for a wide range of pollutants, but
this is only partially possible in practice.  It is suggested, however, in
our further discussion that the advantage of "broad spectrum" control tech-
nologies (those that control a wide range of pollutants) be kept in mind as
a means for maximizing cost effectiveness.  Figure 11 builds on the pollutant
groups defined in Figure 10 and attempts in a very simplified fashion to indi-
cate the relationship of various control approaches to the types of pollutants
of concern.  Ideally, one would like to use the minimum number of broad
spectrum control techniques to achieve effective control at a minimum of
cost; this would truly be the most effective or "best control system"
for the particular air pollution combustion source at hand.  In the context
of this paper a base case involves the use of a "dirty" fuel; therefore,
a combination of control approaches is almost a necessity to achieve a "best
control system" configuration--because most control processes are only
partially effective for at least some of the general pollutant classes.

     It should be further pointed out that the use of various control approaches
can produce some additional pollutants.  In Figure 11 the minuses (-) indicate
a potential reduction in the pollutant; the pluses (+) indicate possible
undesirable production of other pollutants.  The illustration that Figure 11
provides is not an attempt at great accuracy or notation of all the subtleties
of effects but to give an overview of control approach selectivity or lack thereof.
Indeed, the only way this can be thoroughly addressed is through the application
of comprehensive environmental assessment methodology.  This is one reason
why no entries were included for "Overall Effluent Acute Toxicity" in Figure 11.

                                       10

-------
Thus far we have few direct measurements of effluent tbxicity.  Approaches
for making determinations in this area include as part of the environmental
assessment:  comprehensive chemical analyses, bioassays and environmental
alternatives analysis.

     One can conclude from study of Figure 11 that certain combinations look
attractive for effecting some control of all classes of pollutants.  Other
reasons exist for using combinations, including the fact that cost can be
reduced in some cases.
                                       11

-------
                      DISCUSSION OF CONTROL PROGRAM AREAS
     The following discussion of control program areas is aimed at just cover-
ing the highlights.  Many important facets are not included because of time
limitations.
Existing Clean Fuels
     lERL-RTP's program as it relates to existing clean fuels is primarily
focused on coal and mainly low sulfur coal.  Activities here as indicated in
Figure 12 are environmental assessment, characterization of contaminant
levels in various coals, R§D leading to enhancement in the useability of
low-sulfur Western coals in conventional combustion systems, and use of
Western coals in advanced combustion systems like fluid bed boilers.

     Fuel contaminant studies are underway as part of projects with the
Illinois State Geological Survey, U. S. Bureau of Mines, U. S. Geological Survey,
and Battelle.  Information developed and forthcoming should provide basic
information of value in evaluating fuel resource applications and related
environmental control needs.  Such studies point to the problems of substan-
tial variability in sulfur and other contaminant levels within various coal
beds.

     Efforts to enhance the useability of Western low-sulfur coals and lignite
have included an already published study by Monsanto, "EPA-650/2-75-046, Evalua-
tion of Low-Sulfur Western Coal Characteristics, Utilization, and Combustion
Experience,"     and another contract project with KVB which has focused on a
series of mini-demonstrations of Western coal applications to the smaller area-
source stoker-fired boilers where control alternatives were limited and less
evaluation of low-sulfur Western coal use had been done.  Other than useability
and environmental difficulties, it is of interest here to determine the predicta-
bility with which the alkaline type ash in Western coal types combines with
some portion of the fuel sulfur.

     Use of the Western low-sulfur coals has also been examined for use in
fluid bed combustors which have been shown to be capable of handling a wide
variety of fuel types and in which conditions for utilization of alkaline
ash in coal should be optimized.  Studies in this area are being conducted by
EPA and ERDA with encouraging results.
                                        12

-------
Fuel Cleaning

     The major emphasis in lERL-RTP's program related to fuel cleaning is on
coal with a lesser emphasis on residual oil.   Figure 13 shows that the
activities relating to coal cleaning include:  physical cleaning, chemical
cleaning, synthetic fuels, environmental assessment, and control technology
development for fuel processes.

     The physical coal cleaning program has an element concerned with the
characterization of U, S. coals with respect to their cleanability.  EPA-
sponsored work in this area is primarily at the U. S. Bureau of Mines.  The
latest publication in this area is "EPA-600/2-76-091, Sulfur Reduction Poten-
tial of U. S. Coals:  A Revised Report of Investigations." ^ '  Other work
relating to this activity involves the U. S.  Geological Survey.  Another element
of this program concerns the development/demonstration of physical coal clean-
ing technology.  This work has led to a demonstration of coal cleaning as a
control approach for SO  at the Homer City complex, owned by the GPU Service Corp.
and the New York State Electric and Gas Corp.  This 50-million dollar cleaning
facility utilizes a multiple-stream coal cleaning strategy (MCCS) to control
approximately 1200 MW of existing capacity plus an additional new capacity of
600 MW which must meet New Source Performance Standards (NSPS).  EPA's engineer-
ing support activity in the coal cleaning area has resulted in a new publication
which should be of interest to those evaluating the use of coal cleaning,
"EPA-600/2-76-138, Coal Preparation Environmental Engineering Manual."^ '

     Chemical cleaning program activities have centered primarily on evaluation
of alternative processes and pilot plant or bench scale support of some related
processes.  The TRW/Meyers ferric oxidation  leaching process is now being
evaluated at the pilot plant stage.  The report, "EPA-6SO/2-74-025a, Applica-
bility of the Meyers Process for Chemical Desulfurization of Coal:  Survey of
Thirty-Five Coals,"    includes data on removal of other contaminants from coal.
Application and process improvement work is  also underway on the Battelle
Hydrothermal Cleaning Process.  Other processes under evaluation include flash
desulfurization, microwave processing, and microbial leaching.
                                        13

-------
     Under EPA's environmental assessment program, environmental aspects
associated with the processing/utilization of synthetic fuels as clean fuels
are being evaluated for this type of more intensive coal refining.  Several
reports have been issued including, "EPA-600/2-76-101, Evaluation of Pollu-
tion Control in Fossil Fuel Conversion Processes:  Final Report."^    Because
of the longer term development involved here, further discussion is not
included.
     Other environmental assessment and control technology development work
aimed at problem definition and evaluation is underway in the coal cleaning
area.  Major contractors include BMI (environmental assessment) and Versar/
Joy Denver (control technology development).

     The residual-oil related program is aimed at environmentally sound use
of residual oil through contaminant removal.   Related to this is the problem
of conversion of existing gas-fired power plant capacity to residual oil or
coal fuels.  EPA has sponsored the development/application of the Chemically
Active Fluid Bed (CAFB) as a front-end/on-site desulfurization/demetallization
process for residual oil or lignite.  A demonstration plant is currently under
construction on a small gas-fired utility boiler at the La Palma Station of
Central Power and Light (CPL) in San Benito, Texas.  Other techniques for
cleanup of residual oil feedstocks have primarily focused on improved cata-
lyst systems for HDS/HDN/demetallization and have involved projects at Hydro-
carbon Research, Inc.  (HRI) and MIT.  Another important component of the overall
program is environmental assessment of residual oil processing/utilization
alternatives.  Catalytic, Inc., is the major contractor here with additional
work being done by another contractor on characterization of hazardous substances
in residual oil.
                                        14

-------
Combustion Modifications Control

     Figure 14 outlines the major elements of the combustion modifications
control program.  I am limiting the extent of coverage here since the material
will be covered intensively as the subject of this Symposium.  A major con-
tractor, Aerotherm, is being used to provide environmental assessment and
systems study support.  Applications testings is being done by two contractors:
Exxon for electric utilities and KVB for industrial combustion sources.  This
part of the program, in addition to providing various operation guideline
manuals and design guideline manuals, will provide documentation on two or more
detailed long-term corrosion tests on utility boilers under baseline and "low-
NO " operation.
  A

     That part of the program aimed at utility and large industrial boilers
has its primary emphasis on the development/application of low-emission coal
burners for both industrial and utility boilers with a near-term goal of
0.3 Ib NO /10  BTU  (130 nanograms/joule).  Important components of the program,
         A
aimed at small industrial, commercial and residential systems, are:  applica-
tion of optimum control technology for commercial package boilers (Ultrasysterns),
demonstration of the performance of a low-emission oil-fired residential  furnace
(Rocketdyne), and development of technology for control of stoker-fired coal
boilers (Battelle).  The part of the program aimed at afterburners and
industrial combustors is limited to an environmental assessment of after-
burner technology applications.  Development of combustion modifications  for
other industrial combustors will be initiated when work on higher priority
sources is further  along.  In the stationary engine area a major contractor,
Pratt and Whitney,  is focusing  on CM approaches to achieve R£D goals of 50 ppm
(at 15 percent 0_)  for  clean fuels and 100 ppm  for fuels containing up to 0.5
percent bound nitrogen.  The goals represent a  75 percent reduction  from  uncon-
trolled NO  emissions.  A major contractor is also just initiating work to
develop control technology for  large reciprocating engines.  Fundamental  com-
bustion research studies are an important part  of the program and are  coordinated
under a large contract  aimed at defining optimum combustion  control  for all
systems.
                                        15

-------
     Combustion modifications based on advanced processes are an important
part of the program.  Attention is being focused on development of design
criteria for an advanced, optimized, conventional utility combustion system
(Ultrasysterns), an initial application of catalytic combustion (Aerotherm),
and burner design criteria for low-BTU gas.

     EPA's program on fluidized bed combustion  (FBC) over the past several
years led to the breakthrough technology development of the use of a limestone/
dolomite sorbent (in-process reactant for SO-)  as the primary fluidization
material in FBC.  Current work is focusing on environmental assessment and
control technology development of the emerging  FBC systems.  EPA is working
closely with ERDA in this area to assure that needed environmental testing
and evaluation support is applied and independent environmental analysis is
conducted leading to recommendations for standards.  A good example of the
cooperative effort is the work planned for the  30 MW Rivesville atmospheric
fluidized-bed combustor built by ERDA.

     The approach being taken in the FBC program, as in other energy technology
areas, is to establish environmental goals for  the FBC process based ultimately
on health and ecological effects.  Simultaneously, emissions data will be
collected from comprehensive analysis on operating FBC units.  By comparison
of the goals with these measured emissions, the degree of control required will
be identified, and a program will be designed to develop control technology to
meet the established goals.  Results will be used by IERL-RTP to assess the
environmental impact of the FBC process to provide the data base for appropriate
environmental standards for the process, to prepare manuals of best available
control technology, and to design and conduct a program to develop any necessary
control technology.

     Aside from attention being given to particulate control, solid waste
treatment/disposal, and other key areas, work is underway to define combus-
tion modifications for the FBC process to minimize NO  emissions.  Advances
                                                     X
in control of NO  from conventional combustion  systems, as a product of
R§D in this area, can be projected to result in levels of NO  which are lower
than the moderately low levels from FBC units.
                                        16

-------
    ^ Gas Cleaning

     Figure 15 outlines the major elements of work in the area of flue gas
cleaning, including particulate control, flue gas desulfurization, and flue
gas treatment for NO .

     The particulate control research program is directed toward assessing
and extending the capabilities of conventional particulate control systems
(electrostatic precipitators (ESP's), scrubbers, and fabric filters) to
abate aerosol emissions.  In addition, the program will explore new and
improved control methods with potential for cost or performance advantages
and will develop technology specific to existing and potential major particu-
late emission problems, such as low sulfur coal combustion, new fuel and
power production, fugitive emissions, and other selected priority sources.
Major elements of this program include effective application of conventional
particulate control technology, fine particulate control for combustion
processes utilizing low-sulfur fuels, development of new fine particulate
control technology, development of high-temperature/high-pressure (HT/HP)
particulate control, and technology transfer.

     The flue gas desulfurization program area includes activities relating
to non-regenerable processes, regenerable processes, technology transfer/
supporting studies, and control of waste and water pollution from combustion
systems.  The non-regenerable program is focusing on lime/limestone tech-
nology development including the Shawnee alkali scrubbing test program,
Louisville Gas and Electric1s Paddy's Run lime scrubbing test program, appli-
cation of FGD systems to industrial boilers, and a stack gas reheat assessment
study.  A second key portion of the non-regenerable program is a full scale
demonstration of the double alkali process and development/demonstration of
double alkali limestone regeneration techniques.  In the area of regenerable
processes primary attention is being given to demonstration of the WeiIman-Lord/
Allied process, a magnesium oxide test program, aqueous carbonate process demon-
stration, citrate process demonstration, and comparative economic studies at TVA.
In the area of FGD technology transfer and supporting studies work is empha-
sizing engineering application/information transfer, flue gas cleaning decision
models and data books, FGD process environmental assessment studies, and non-
utility source studies.  In the area of control of waste and water pollution
                                        17

-------
from combustion sources, activities include assessment of flue gas cleaning
disposal technology, by-product utilization, and development of water utiliza-
tion/treatment technology for power plants.

     Flue gas treatment for NO  is the smallest part of the flue gas cleaning
                              A
program area.  It is being given attention, however, because high NO  removal
efficiencies for stationary combustion sources may be necessary, based on
studies indicating additional adverse health effects of NO , atmospheric
                                                          A
chemistry studies identifying NO  as a precursor for other adverse pollutants,
                                X
the relaxation of NO  control requirements on mobile sources, and a potential
short-term standard which may require greater NO  control.  The aim is to
demonstrate flue gas treatment technology for control of NO  and SO /NO  from
                                                           A       A   y*
stationary combustion sources.  The flue gas treatment technology is being con-
sidered as add-on technology supplementing combustion modification techniques
when high removal efficiencies are required or when the source under considera-
tion is not amenable to extensive control by combustion modification.  The
overall approach is to conduct strategy and technology assessment activities
in parallel with an experimental program progressing from pilot/prototype
to full-scale demonstration.  Although strategy/assessment studies are incom-
plete relative to the requirements for flue gas treatment technology, it
is planned to proceed with small scale experimental work.  Due  to its advanced
state, Japanese technology is expected to play a major role in planned hardware
proj ects.
                                       18

-------
                                    SUMMARY

     In summary, the Industrial Environmental Research Laboratory at Research
Triangle Park is pursuing a very comprehensive environmental engineering
program on stationary source combustion control as part of the Interagency
Energy/Environmental Program.  This program should continue to provide
much needed technical data and assessment information for use by concerned
organizations in the environmental regulation area, in the industrial
user areas, and many others.  Figure 16 is a partial listing of primary users
of program activities/results.
                                        19

-------
                                 REFERENCES
1.    Preliminary Environmental Assessment of the Application of Combustion
     Modification Technology to Control Pollutant Emissions from Major
     Stationary Sources," (Draft, to be published), EPA Contract 68-02-2160,
     Acurex/Aerotherm.

2.    Electrical Energy as an Alternate to Clean Fuels for Stationary
     Sources, EPA-600/2-76-049a/b (NTIS No. PB 251-829/AS and 251-830/AS),
     March 1976.

3.    Evaluation of Low-Sulfur Western Coal Characteristics, Utilization and
     Combustion Experience, EPA-650/2-75-046 (NTIS No. PB 243-911/AS),
     May 1975.

4.    Sulfur Reduction Potential of U.S. Coals:  A Revised Report of Investi-
     gations, EPA-600/2-76-091 (NTIS No. PB 252-965/AS), April 1976.

5.    Coal Preparation Environmental Engineering Manual, EPA-600/2-76-138
     (NTIS No. PB 262-716/AS), May 1976.

6.    Applicability of the Meyers Process for Chemical Desulfurization of
     Coal:  Survey of Thirty-Five Coals, EPA-650/2-74-025a  (NTIS No. PB 254-
     461/AS), September 1975.

7.    Evaluation of Pollution Control in Fossil Fuel Conversion Processes:
     Final Report, EPA-600/2-76-101 (NTIS No. PB 255-842/AS), April 1976.
                                         20

-------
Environmental Engineering R§D Aimed at Technological Trends
   - More coal combustion
   - Conversion to oil and coal
   - Increased use of electrical energy from coal as an
     alternate to clean fuels
   - Increased use of existing capacity
   - Increased use of coal in intermediate sized sources
     (stokers)
   - New combustion systems
   - New fuels
   - Conservation
Environmental Engineering Methodology Development/Evaluation
   - Multi-media environmental assessment  (more effective techniques
     for rapid evaluation using available knowledge)
   - Overall environmental evaluation of complex effluents,
     including comprehensive chemical and biological analysis
   - More systematic techniques for control approach evaluation
   - Increased emphasis on most effective combined controls for
     broad spectrum control  (control systems)
   - Reduced emission levels at the same or reduced cost
   - Finding practical candidate options for new comprehensive
     emission regulatory approaches based on environmental
     assessment methodology  (e.g., minimum acute toxicity
     effluents)

   Figure 1.  Environmental engineering R§D challenges  in  the  stationary
              source combustion area.
                                      21

-------
An environmental assessment is a continuing iterative study aimed at:

     (1)  determining the comprehensive multi-media environmental .
          loadings and environmental control costs, from the
          application of the existing and best future definable
          sets of control/disposal options to a particular
          set of sources, processes, or industries; and

     (2)  comparing the nature of these loadings with existing
          standards, estimated multi-media environmental goals,
          and bioassay specifications as a basis for prioritiza-
          tion of problems/control needs and for judgement of
          environmental effectiveness.
         Figure 2.  Environmental assessment definition.
                                                                                  I
                                   22

-------
 i^  *  ai =»._

tl  1  ill   i  I  i  =
M  j  Hit  i  Hi  i
2S« S  SB »«  r.s5 I  ₯
   Ilil  Ill
   < « i=S  > Si
   5 5«s  210

   llfi  i!!
   £ cd XX  o I*-
   > f ig« £ £3
   I 5 E^S 6 I"   «
   I S 122 S S5li? !
   5 • •   • •   •
i  2|i
§H «S°
i| ii§
«o It^S
       -
      Sc

|i
si
«3
                      s  i.
                      IIM
                      I! If
                      1  !'
          2|gi

                                 I
                      S B S
                      « S f=
                      12 ib *
                      S S S
                         «
                      | is
                      -
                                                       &

                                                       c>
                    2
                    S
                                      •H
                                      0)
                                      0)
                                      •o
                                      (1)
                                      
-------
C3 OC
a


f




•










CO
_l Ul

*>
2 £ to 22
E£<3g

ZK»O —
< U Z OC
S»ui o ul
O "» O 55;
Ul **| J


>
f^»
a
BASIC AND APPLIED R&
NCH AND PILOT EXPERIMI
UDIES TO ASSESS GENERK
R EFFECTIVENESS &SECG
VIRONMENTAL PROBLEM!
NDAMENTAL STUDIES
CQ CO

*
X ^. — o
M
£ z
O ul
0 %?
cc 3 <
1 g
0 .

i-
CA ^~
1=
O ^
-g

0
o
f*.
1
\
L*
^

O Z 3
^* UJ ^U

*









S* S . Cfl CO
QZ -» S _j z z
ujujoa< o o
^ ^ b* Ul ' ^ ^"
Pozz2; 5 >
_ioco:=S 5 ui
a — o 2 a < oc
s| S • •
Ul
t
CO
> UJ
^™ • S5 ^"
KM o a
^ * C u **
*•> uj ••• S ,_
>• 1-0 -"-~ CO
<  CO ^ ca § ul CO
Ul 
DC OQ Q.  J-
g2 fe |

















i
i en
> UJ
» 5
1 ^"
"1 s * i
-»5 Ji o °
o > "o ?• Z
gi" z fe 5
'[=•-" 2 Ss |
gg a «3 s
0E Q S^ Z
= 2 <1|1
s> i~is
^ v> o u S ui h-
* z P08 o
0 £ C3 ^
o e
9 <

3H CW
) ^

"— — — • """ """ "~

t;
M UJ
o<
^J ^j
H


-J
ii

"-1

CO 64
«g2 0° W Ul«-l J
§2 IS gl Ep 1
2 E Sir £2 t»c«z S
£§ |s i £Ss a
s "s
""" *^ ^..^
"~ — — __
^•^
igSs
ij z LU iZ
fegz S
*
^

r
z z
Ul Ul
Ul
ii M
£ 2 OC UJ
uiHo- ui
ec ** < eo
Q.



24

UJ*5
2^
^ £
UJ •
^«









OC

Ul
I
u.
IELD TESTING OF STATE C
ND RELATED SYSTEMS
u. <
0
















                                                                  £
                                                                  2
                                                                  en
                                                                 .2
                                                                 •O

                                                                  O)
                                                                 ^
                                                                  O

                                                                 "§
                                                                 I
                                                                  en

-------
                        Conventional  Combustion  Systems
  Utility and Industrial Boilers
     - Natural-gas-fired
     - Distillate-fired
     - Resid-fired
     - Coal-fired
  Small Industrial and Commercial  Boilers
     - Natural-gas-fired
     - Distillate-fired
     - Resid-fired
     - Coal-fired
  Commercial/Residential Furnaces
     - Natural-gas-fired
     - Distillate-fired
     - Resid-fired
     - Coal-fired
  Industrial  Process Furnaces and  Afterburners
     - Natural-gas-fired
     - Distillate-fired
     - Resid-fired
     - Coal-fired
  Internal Combustion Engines
     - Natural-gas-fired
     - Distillate-fired
  Gas Turbines
     - Natural-gas-fired
     - Distillate-fired        „
     - Resid-fired
                          Advanced Combustion Systems
0 Fluid Bed Combustion
     - Resid-fired
     - Coal-fired
     - Coal-reject-fired
0 Catalytic Combustion
     - Natural-gas-fired
     - Distillate-fired
                Figure 5.  Major stationary combustion sources.
                                       25

-------
  0  Existing  Clean  Fuels
  0  Fuel  Cleaning
  0  Combustion Modifications
  0  Flue  Gas  Cleaning
Figure 6.   Control program areas.
               26

-------
                         Coal
                      Oil
                 Gas
Total
1015 kJ/yr
Utility Boilers 10.833
Packaged Boilers 3.449
Warm Air Furnaces
and Miscellaneous
Combustion
Gas Turbines
Reciprocating
io15
3.
5.
2.


0.
0.
kJ/yr .
483
801
132


844
327d
IO15
4.
6.
4.


0.
0.
kJ/yr
906
323C
542


681
9136
io15
19.
15.
6.


1.
1.
kJ/yra
222
573
674


525
240
(43
(35
(15


(3.
(2.
.5%)
.2%)
.1%)


4%)
8%)
1C Engines


Total
14.282 (32%)
12.587 (29%)    17.365 (39%)   44.234 (100%)
a 1 BTU (British Thermal Unit) = 1.056 kJ (kilojoules)


  this sector includes steam and hot water units



  includes process gas


  includes gasoline and oil portion of dual fuel

g
  includes natural gas portion of dual fuel
                 Figure 7.  1974 stationary source fuel consumption.
                                                                     (1)
                                          27

-------
0 Existing Clean Fuels

     - Low-sulfur fuel
     - Low-nitrogen fuel
     - Low-ash fuel
     - Low-chlorine fuel
     - Low-trace metal(s) fuel

0 Fuel Cleaning

     - Physical cleaning
     - Chemical cleaning
     - Hydrodesulfurization (HDS)
     - Demetallization
     - Hydrodenitrogenation (HDN)
     - Chemically active fluid bed (CAFB)
     - Synthetic fuels

0 Combustion Modifications (CM)

     - Low excess air (LEA)
     - Staged combustion (SC)
     - Flue gas recirculation  (FGR)
     - Water injection
     - Good operating practice
     - Modified burner
     - Optimum operation and burner/furnace design
     - Fuel additives
     - Furnace reactants (for S, N, Cl)
     - Afterburners

0 Gas Treatment (Flue Gas Cleaning)

     - Electrostatic precipitator (ESP)
     - Fabric filter
     - Particulate scrubber
     - Mechanical collectors
     - Granular bed filters
     - Nonregenerable flue gas desulfurization (FGD)
     - Regenerable FGD
     - Flue gas treatment for NO
     - Catalytic devices

  Combinations (a few examples for a coal-fired utility boiler)

     - Low-S fuel/CM/baghouse
     - CM/ESP/FGD
     - CM/coal cleaning/FGD
     - CM/coal cleaning/ESP
     - CM/alkali reactant/baghouse
     Figure 8.  Partial list of control approaches for combustion sources
                                       28

-------
     Stationary
Combustion Source Type
Existing
 Clean
 Fuels
  Fuel
Cleaning
 Combustion
Modifications
Flue Gas
Cleaning
Conventional Combustion
        Systems

0 Utility and Large Indus-
  trial Boilers

   Natural-Gas-Fired
   Distillate-Fired
   Residual-Fired
   Coal-Fired

0 Small Industrial and
  Commercial Boilers

   Natural-Gas-Fired
   Distillate-Fired
   Residual-Fired
   Coal-Fired

0 Residential Furnaces

   Natural-Gas-Fired
   Distillate-Fired
   Residual-Fired
   Coal-Fired

0 Industrial Process Furnaces
  and Afterburners

   Natural-Gas-Fired
   Distillate-Fired
   Residual-Fired
   Coal-Fired

0 I.C. Engines

   Natural-Gas-Fired
   Distillate-Fired

0 Gas Turbines

   Natural-Gas-Fired
   Distillate-Fired
   Residual-Fired

Advanced Combustion Systems

0 Fluidized Bed Combustion
   Residual-Fired
   Coal-Fired
   Coal-Cleaning-Rejects Fired

0 Catalytic Combustion

   Natural-Gas-Fired
   Distillate-Fired
  N.A.
  N.A.
  N.A.
  Yes
  N.A.
  N.A.
  N.A.
  Yes
  N.A.
  N.A.
  N.A.
  Yes
  N.A.
  N.A.
  N.A.
  Yes
  N.A.
  N.A.
  N.A.
  N.A.
  N.A.
   No
   No
  Yes
  Yes
   No
   No
  Yes
  Yes
   No
   No
  Yes
  Yes
   No
   No
  Yes
  Yes
   No
   No
   No
   No
  Yes
     Yes
     Yes
     Yes
     Yes
     Yes
     Yes
     Yes
     Yes
     Yes
     Yes
     Yes
     Yes
     Yes
     Yes
     Yes
     Yes
     Yes
     Yes
     Yes
     Yes
     Yes
  N.A.
  Yes
  N.A.
  N.A.
  N.A.
  Yes
  Yes
 Maybe
   No
   No
    Maybe
    Maybe
    Maybe
    Maybe
    Maybe
  Maybe
  Maybe
   Yes
   Yes
  Maybe
  Maybe
  Maybe
  Maybe
   No
   No
   No
   No
  Maybe
  Maybe
  Maybe
  Maybe
  Maybe
  Maybe
  Maybe
  Maybe
  Maybe
    Yes
    Yes
    Yes
    No
    No
N.A. = Not applicable
                 Figure 9.   Applicability of general control approaches.
                                             29

-------
Inorganics
       Gaseous Inorganics

          - NO
          - SO*
          - HCT
          - Overall Effluent Acute Toxicity

       Particulate Inorganics

          - Total Inorganic Particulate
          - Primary Sulfates
          - Fine Particulate
          - As
          - V
          - Pb
          - Fe
          - Cr
          - Overall Effluent Acute Toxicity
Organics
       Gaseous Organics and CO

          - Nonmethane HC
          - CO
          - Overall Effluent Acute Toxicity
       Particulate Organics

          - Total Organic Particulate (including carbon)
          - Polycyclic Organic Matter
          - Overall Effluent Acute Toxicity
  Figure 10.  Overview of some example classes of air pollutants
              of concern from combustion processes.
                                  30

-------
Types "f Pollutants
of Concern
( Examples)
Insraaaics:
* Gaseous Inorganics
NOX
S0x
HC1
Overall Effluent Acute
Toxic ity
* •
Total Inorganic Particu-
late
Fine Particulate
Primary Sulfates
As
V
Pb
l:e
Cr
Overall Effluent Acute
Toxicity
Organic*:
0 fi.'isoous Organ ics ft CO
Nonmeth;inc IIC
CO
flvcrai 1 Affluent Acute
Toxicity
0 Participate Qrpanies
Total Organic Particu-
3 ate
POM
Overall Effluent Acute
Taxicity
Mxisting
Clean Fuels
V V w
SLL tt. «
v*
U. C — » V
V V C Z
f, 00 P -rl
a o u. »* t>
-7?_
•*• * * *

+••*••*•



flue G*s Cleaning
(G«* TreatlMnt)
Electrostatic Precipitator (ESP) ,
Fabric Filter |
Particulate Scrubber
Mechanical Collector
Granular Bed Filter
Non-Segenerable FGD
Regenerable FGD
J=GT for NOX
J 000000-
0 0000--0
0 0 - 0 0 - - 0


•f
. - - 0 0 	
+ +




------?
	 ?
^




0 000077?
0 0000000



7 ?_?"___

Figure 11.  Relationship of control approaches to the types of
            pollutants of concern controlled.
                               31

-------
           Existing Clean Fuels

              - Environmental Assessment
              - Fuel Contaminant Studies
              - Enhance  Useability of Western Low-Sulfur
                Coals and Lignite
              - Use in Advanced Combustion Systems like
                Fluidized Bed Combustors
Figure 12.  Program activities related to existing clean fuels.
                                 32

-------
Fuel Cleaning
   - Coal Cleaning
        00 Physical Cleaning
        00 Chemical Cleaning
        00 Synthetic Fuels
        00 Environmental Assessment
        00 Control Technology Development for Fuel
           Processes

   - Residual Oil
        00 Development/Application of Chemically Active
           Fluid Bed (CAFB) as Environmentally Acceptable
           Approach for Conversion of Gas-Fired Boilers
           to Residual Oil
        00 HDS/HDN/Demetallization
        00 Environmental Assessment of Residual Oil Processing/
           Utilization Alternatives
   Figure 13.  Program activities relating to fuel cleaning.
                                33

-------
Combustion Modifications Control

   - Environmental Assessment of Combustion Modifications (CM)
                                            4
   - Applications Testing

   - Development of CM for Utility and Large Industrial
     Boilers

   - Development of CM for Small Industrial, Commercial and
     Residential Systems

   - Development of CM for Industrial Process Combustion
     and Afterburners

   - Development of CM for Stationary Engines

   - Development of CM Based on Advanced Processes

   - Fundamental Combustion Research Studies
      Figure 14.  Program activities relating to combustion
                  modifications control.
                                34

-------
         Flue  Gas Cleaning
            -  Paniculate  Control
            -  Flue Gas  Desulfurization (FGD)
            -  Flue Gas  Treatment  (FGT)  for NO
Figure 15.  Program activities relating to flue gas cleaning.
                               35

-------
to On- going Program
mental- Review of ERDA
4->
3
ft

rH

rH
rt
4->
C
CD

g
2
• H
>
C
PJ
,



c
o
rl
•H
>
C
U)

4->
C
(D
'O
C
0)
ft
0)
•o
c
hH
1



4-1
c
€
ft
o
rH
a)
>
a>
a

X
00
o
rH
O
r*
0
0)
H




sment Methodology
in

<

rH
«
4->
C
a>
£
C
O
rl
•H
>
C
UJ
1



Recommendat ions

X
OO
o
rH
0
c
f.
o
o
H

rH
2
4->
C
O
U
1









to
9
0>
•H
>
O
OS

OO
•H
tn
a>
Q
I







tn

a>
•H
>
d>
CA

rH
rt
tn
o
ft
0
r4
O.
I



cts of Environmental

erf

&
rl
(D
C
UJ
1
Si
tu
o


I/)
4>
X
0
rt
o
r<
ft
&<
<

rH
O
rl
4->
d
o
U




00
c
• H
X rl
i^ a>
0) V
•o c
M -H
U 00
 B
rt 0)
Z T3
rt
o o
+J  rt
rlS
ff-H
rH 4->
rt
rH 2
rt ^,
4-1 in
c o
a; o
i c
B a) «
S-H (U
0 .rt
•H CO T3
> 3
C 
U O CO

c
o
•H
4->
&

rH
O
r<
4J
§
U
US
in
o
>
•H
+->
rt


0)
4->
rH
CO <
a,
3 rH
O rt
ai 4-> c
u c o
0 .H
J B 4->
< C rt
g 2g
OJ -rt O
S > HH
% £ G
O W rH
C£



rl
O
•H
rH W
ft 4J
s*s
CO 4->
"*- d
in 3 in tn
r( W rl rl
Q) C 0> 4)
ft O V) O
O^U 3 -H
-H -^-s^ J.
ID in tn M
> rl r4 4>
0)  O 4-> -H CO
 -H rH .H
>- O C > rt 3
DC ri o c o cr
H 0, U UJ U W
CO



in
rH
rt
3
TJ
•H
>
•H
•o
C rH
M2
<4-l 4->
o c
o
fl> U
w *^
3 in
4-» 0
0 rH
o x>
rl ^
•H N
Q a.
rl C
0 O
<4H
c
w o
U  -H
rH (3 4J
J -H rt
QQ rH e
=> V M
CU Tl O
•H t+H
•J 3 C
< U rH
Cz!
                                                            0>
                                                            y
                                                            rt
                                                            2
                                                            ft
enta
ro
4-1
in

T)
      o
      rt

      U
      u
      4->
      C tn
      rt -H
     •H 4->
     *O rt
      1> rl
     rH  C
      3  O
     s  o
                                                            M
                                                            0)
                                                            tn



                                                            fr
                                                            rt
                                                            B
                                                            •H
                                                            ri
                                                            a,
                                                            H
                                                            3
                                                            bo
                                                            • H
                                                            U.
to

o
          i
          rH
          z
          O
  36

-------
                KEYNOTE PAPER
ENVIRONMENTAL ASSESSMENT OF STATIONARY SOURCE
   NOX COMBUSTION MODIFICATION TECHNOLOGIES
                     By:

       H. B, Mason and L. R. Waterland
    Acurex Corporation/Aerotherm Division
           Mountain View, CA  94042
                      37

-------

-------
                                  ABSTRACT
       The EPA has determined that expanded technology for combustion modifi-
cation NOX control is needed to maintain air quality for N02 in the 1980's and
1990' s.  In response to the expansion, there is currently a need to (1) ensure
that the current and emerging control techniques are environmentally sound,
and  (2) ensure that the scope and timing of control development is adequate to
achieve compliance with current and anticipated air quality standards.  The
environmental assessment program supports these needs through two main objec-
tives:  (1) identify the multimedia environmental impact of stationary com-
bustion sources and NOX combustion modification controls; and, (2) identify the
most cost-effective, environmentally sound NOX combustion modification controls
for attainment and maintenance of current and projected N02 air quality stan-
dards to the year 2000.  The initial effort was a preliminary assessment in-
volving compilation and evaluation of process and impact data and determination
of the program approach and priorities.  Air quality projections for Los Angeles
and Chicago showed that development of advanced combustion modification controls
will be needed for all  major stationary sources of NOV.  Source priorities for
                                                     X
testing and process evaluation were set, in the following order, according to
current and projected emissions and prospects for control implementation:
utility boilers, watertube industrial boilers, gas turbines, firetube industrial
boilers, warm air furnaces, 1C engines and industrial process combustion.  Data
for the economic, operational and incremental environmental impact of combustion
modification controls are generally sparse.   Incremental emissions of CO, UHC
and parti cul ate can be minimized through control development engineering.  Other
effluents — ROM's, segregating trace metals and sul fates -- show potential for
increased emissions due to some combustion modifications.  The current effort
is focused on field tests and process evaluation to quantify this potential.

-------

-------
                                 SECTION  1
                               INTRODUCTION

       The 1970 Clean Air Act Amendments designated oxides of nitrogen (NO )
as one of the criteria pollutants requiring regulatory controls to prevent
potential widespread adverse health and welfare effects.  Accordingly, in
1971, the EPA set a primary and secondary National Ambient Air Quality Stan-
dard (NAAQS) for N02 of 100 ug/m3 (annual average).  To attain and maintain
the standard, the Clean Air Act mandated control  of both mobil and station-
ary NOX sources each of which emits approximately half of the nationwide
manmade NOX<  Stationary sources are regulated through EPA Standards of Per-
formance for New Stationary Sources which are set as technology becomes
available based on the best system of emissions reductions.  Additional sta-
tionary source standards required to attain air quality in the Air Quality
Control Regions can be set for new or existing sources through the State
Implementation Plans.  Mobile source light duty vehicle emissions were to be
reduced by 90 percent to a level of 0.25 g N02/km (0.4 g/mile) by 1976.
       Since the Clean Air Act, a moderate level of stationary source NO  con-
                                                                        A
trol (30- to 50-percent reduction) has been developed and implemented for a
variety of source/fuel combinations.  EPA standards of performance for new
stationary sources were set in 1971 for gas-, oil- and coal- (except lignite)
fired large steam generators.  A more stringent standard for bituminous coal-
fired large steam generators is being considered based on technology devel-
opments since 1971 (1).  Standards are in preparation for lignite-fired
large steam generators, gas turbines, reciprocating internal combustion
engines and intermediate size steam generators.  Additional local standards
have seen set, primarily for new and existing large steam generators and gas
turbines, in several NOX critical areas as part of State Implementation
Plans.   The net result of this regulatory activity is that there are
                                     41

-------
currently over 200 stationary sources controlled to a level of 30- to 50-
percent emission reductions.   The number of controlled sources is increasing
as new units are installed with factory equipped NO  controls.
                                                   A
       A comparable level of emission reduction has been implemented for
mobile source light duty vehicles.  Although the stationary goal of 90-
percent reduction (0.25 g NOg/km) by 1976 has not been achieved, emission re-
ductions of about 25 percent (1.9 g/km) were in effect for the 1974 to 1976
model years and a 50 percent reduction to 1.25 g/km is currently in effect.
Achievement of the 0.25 g/km goal has been deferred indefinitely due to
technical difficulties and fuel penalties associated with this level of con-
trol (2).  Initially, the Energy Supply and Environmental Coordination Act
of 1974 deferred compliance to 1978.  More recently, the House and Senate
versions of the Clean Air Act Amendments of 1977 abolished the 0.25 g/km
goal and replaced it with an emission level of 0.62 g/km (1 g/mile) for 1980
and beyond (Senate Bill 252) or 1982 and beyond (House Bill 6161).  The EPA
Administrator has requested the option of reviving the 0.25 g/km standard in
1983 if shown necessary by studies of the effect of N02 on human health (3).
       The relaxation of the mobile source emission regulations has placed a
greater burden on stationary source NO  control for maintenance of air qual-
                                      /^
ity for NOp.  A number of air quality planning studies have evaluated the
stationary source NO  control requirements in the 1980's and  1990's in view
                    A
of recent developments (4-7).  These studies all conclude that the relaxation
of the mobile standards coupled with the continued growth rate of stationary
sources will require stationary source controls more stringent than provided
for by the current and impending standards of performance for new stationary
sources.  This conclusion has been reinforced by the recent switch to in-
creased coal use in stationary sources.  The studies also conclude that the
most cost-effective way to achieve the added stringency  is through the use of
combustion modification NO  controls in new sources.
                          A
       While the above discussion deals with the requirements for stationary
source controls to achieve and maintain the current NAAQS (100 ug/m3, annual
average), there is also the possibility that separate NO  control requirements
                                                        *»
will be needed to attain and/or maintain additional N02-re1ated standards.
Recent data on the health effects of N02 suggest that the current NAAQS should
                                     42

-------
be supplemented with a limitation on short-term exposure (7-10),  and the
House Clean Air Act Amendments (HR 6161) require that the EPA consider
setting a short-term NOg standard.  EPA plans to consider the need for a
short-term standard in 1978 when the NOg air quality criteria document (11) is
updated (12,13).
       For the longer term, EPA is continuing to evaluate the need for addi-
tional NOV regulation as part of the oxidant control  strategy or  for control
         A
of pollutants for which NO  is a precursor, e.g., nitrates and nitrosamines
                          A
(7,8,12-15).  These regulations could be in the form of source emission con-
trols or additional ambient air quality standards.  In either case, addi-
tional stationary source control technology could be required to  comply with
the standards.
       In summary, the developments in NO  regulatory control since the Clean
                                         A
Air Act are leading, in the near term, to fairly extensive implementation of
moderate levels of control through hardware modification to existing units or
new units of conventional design.  For the far term, air quality  projections
show the need for levels of control more advanced than originally anticipated.
Control of new sources through low-NO  redesigns is the preferred approach to
                                     A
achieving these advanced levels of control.  This expanding use of combustion
modifications highlights the need for an environmental assessment program
which is the subject of this paper.
                                    43

-------
                                SECTION 2
                         NOX E/A PROGRAM OVERVIEW

       With the increasing use of conventional combustion modification tech-
niques and the prospect of development and use of advanced techniques, control
developers and users need to know the environmental, economic, energy and en-
gineering implications of combustion modification technology.   This informa-
tion is needed to aid in selection of control systems, to support the setting
of standards and to guide control R&D programs.   The "Environmental Assessment
of Stationary Source NO  Combustion Modification Technologies" (NO  E/A)  was
                       A                                          A
initiated in June 1975 to provide this information.   Specifically, the program
addresses two questions:
       t   What are the impacts, and potential corrective measures, associated
           with the use of specific existing and advanced combustion modifica-
           tion techniques, e.g.
           —   The change in gaseous, liquid and solid emissions to the air,
               water and land due to NOX controls
           —   The capital and operating cost of NO   controls  per unit reduc-
                                                   A
               tion in NO
                         A
           —   The change in energy consumption efficiency
           —   The change in equipment operating performance
       •   What are the priorities and schedule for NO  control technology
                                                      X
           development considering:
           -   The above impacts for each source/control combination
           -   Control requirements to attain and maintain the current annual
               average N02 ambient air quality standard
                                     44

-------
           -   Control requirements to attain and maintain a potential short-
               term N0? standard, or, other NO  related standard (oxidants)
                      Cm              (         ^
           -   Alternate mobile source standards
           -   Alternate energy and equipment use scenarios to the year 2000
               in the Air Quality Control Regions with potential NOV problems
                                                                   A
       The NOX E/A program has a major task to address each of these ques-
tions.  The "Process Engineering and Environmental Assessment" task quantifies
impacts from specific combinations of stationary combustion sources and NO
                                                                          A
controls ranging from the use of existing control technology on conventional
equipment to the use of developing control techniques on advanced equipment
design.  The process engineering component of this task is evaluating the
hardware modifications required for NO  control and quantifying the.accompany-
                                      A
ing cost and impact on fuel consumption and system operation.  This effort is
drawing on the results of past and ongoing field demonstration tests of NOX
controls.  The environmental assessment component of the task quantifies the
near-source impacts on human health as well as terrestrial and aquatic ecology
from the change in gaseous, liquid and solid emissions resulting from the use
of NOX controls.  The results of the task are rankings of source/control com-
binations, now and in the future, based on the potential impacts associated
with the use of NOX combustion modification techniques.
       The second major task, "Systems Analysis," evaluates the priorities
and required schedule for NO  control development based on current and pro-
                            A
jected control requirements on a regional and national basis.  This task
uses air quality projection models for several NOV critical regions.   These
                                                 A
models together with the source-specific impact data from the process engi-
neering task, suggest the most cost-effective and environmentally sound mix
of stationary control techniques for given constraints on source growth,
ambient air quality goals and mobile source standards.  Several scenarios for
energy use and mobile source standards between now and the year 2000 are
being used so that the impact of these contingencies on control needs can be
evaluated.
gram priorities.
planning.
The results of this task are used within the NOY E/A to set pro-
                                               A
      They are also intended to assist the EPA in control R&D
                                    45

-------
       The two major tasks cited above are supported by three other tasks,
"Emission Characterization," "Impacts and Standards," and "Experimental
Testing."  The relationship and scope of these tasks are shown on Figure 1.
Here, the rectangular boxes denote subtasks and the ovals show program output.
The initial effort, shown at the top of the figure, developed the approaches,
supporting data and priorities on sources, controls pollutants and impacts for
the assessments and system analyses currently in progress.  These initial re-
sults were documented in detail in a preliminary environmental assessment re-
port (16), the final of which will be available in September.  The remainder
of this paper summarizes the results of the preliminary assessment with empha-
sis on those areas most relevant to this symposium.
                                       46

-------
                                 SECTION 3
                         SOURCES OF NOY EMISSIONS
                                      A
       The major sources of oxides of nitrogen emitted to the atmosphere are
shown on Figure 2.  The stationary fuel combustion sources being assessed in
the NOV E/A are bracketed at the top of the figure.  On a global basis, it is
      A
estimated that up to 90 percent of NO  mass emissions are from the natural
processes of biological decay and lightning (17).  In urban industrial areas,
however, the reverse is true.  There, manmade sources account for up to 90
percent of ambient N02 concentrations.
       Manmade NO  emissions in the United States in 1974 were 21.1 x 106
                 A
metric tonnes {23 x 106 tons) on an N0« basis.  Of this amount, 9.6 x 106 MT
(10.6 x 106 tons), or 46 percent were due to mobile sources consisting of
light and heavy duty highway vehicles, aircraft, trains, ships and off-road
vehicles such as construction equipment (18).   The remaining 54 percent of
manmade NO  emissions are due to stationary sources distributed by sector,
          A
as shown in Figure 3.  The four largest sectors are utility and industrial
boilers, 1C engines, and residential and commercial warm air furnaces.  The
"other" category is fugitive emissions, largely open burning.  Combustion
sources account for over 98 percent of stationary emissions with conventional
fossil fuel combustion responsible for 94 percent.
       The emission inventory in Figure 3 includes estimates of the effect of
utility boiler N0y controls on nationwide emissions.  These estimates show
                 A
that the controlled nationwide utility boiler emissions are only 3 percent
below uncontrolled levels for 1974.  Although there are a number of gas tur-
bines with NO  controls, the effect of these controls on a. nationwide emission
             A -
                                  Thus, the current impact of NOX regulations
estimate was negligible in 1974.
on nationwide emissions is small.  The extent of control implementation is
increasing, however, particularly as new sources covered by emission standards
are installed.                       "
                                    47

-------
       The distribution of stationary emissions by fuel is shown on Table I.
Coal-fired equipment generated nearly 37 percent of stationary NO  but con-
                                                                 A
sumed only 32 percent of fuel energy used in stationary sources.  By contrast,
natural gas-fired sources consumed 40 percent of stationary fuel energy.but
emitted only 32 percent of stationary source NO .  This contrast is largely
                                               A
due to the higher NO -forming potential of coal due to high concentration of
                    A
bound nitrogen in the fuel.
This effect of fuel type on NO  emissions is of
                              /\.
particular relevance with the increase in use of potentially high NO  fuels -
                                                                    A
coal and heavy oil - as a result of the clean fuels supply deficit.
       An emission inventory of stationary combustion NO  sources was also
                                                        A
generated for other pollutants for which emission data were available.  The
results for the criteria pollutants, sulfates and polcyclic organic matter
(POM) are summarized on Table II.  The inventory for trace metals is given
in Reference 16.  The emission factor data for the criteria pollutants was
generally of good quality.  As expected, the emissions of hydrocarbons and CO
from the major stationary NO  sources (except 1C engines) are small compared
                            A
to mobile sources.  Also, the only significant source, on a mass emission
basis, of combustion generated SO  and particulate is boilers firing coal or
                                 A
heavy oil.  One exception is the process heating sector (cement kilns, etc.)
where the emissions are largely process generated rather than combustion gen-
erated as in the other sectors.  The emission factor data for noncriteria
pollutants was generally sparse and of low quality.  For example, little data
is available for individual species of ROMs, and the data for the total PQMs
as shown in Table II exhibits a scatter of over two orders of magnitude.
Similarly for trace metals, data are available only for elemental loading
rather than for metallic species.  The limited elemental data available show
a high dependence on fuel composition.
       The source sector groupings shown in Figure 2 and Table I consist of a
variety of equipment design/fuel combinations, many of which may have signif-
icantly different potential for NO  emissions and/or NO  control.  The emis-
                                  *»                    X
sion inventory cited herein considered nearly 100 such combinations.  However,
the ranking of the most significant of the combinations, given in Table III,
shows that the top 30 emitters account for over 90 percent of stationary NO
                                                                           A
emissions.  NOX control technology can thus be leveraged considerably by
                                     48

-------
treating the major emitters in depth and using this basic technology for
spinoffs to the minor sources.
       The NOX E/A emission inventory summarized in Figure 3 and Tables I to
III is being updated and revised as new field test data, particularly for
noncriteria trace pollutants, become available.   The inventory is used in the
NO  E/A in two ways.  First, the total mass emissions for a given source and
  "                                                                         j
pollutant are used to set preliminary relative priorities among the source/
fuel combinations.  Second, the emission factor data in the inventory are used
as input to the impact assessment calculation.  Maximum ground level concen-
trations of pollutants are estimated by approximating the dispersion and, if
relevant, the transformation of the flue gas pollutant concentration.  These
ground level concentrations are then compared to impact threshold levels where
the concentration of pollutants may be potentially harmful to human health or
the ecology.  This procedure highlights potential source/pollutant problems
requiring attention in the assessments and field testing.
                                     49

-------
                                  SECTION 4
              EVALUATION OF COMBUSTION MODIFICATION TECHNOLOGY

       The regulatory program to control emissions from the sources discussed
above emphasizes application of best available control technology in new units.
New source controls are emphasized since experience has shown them to be more
effective, less costly, and less disruptive than retrofit control of existing
equipment.  The current and impending Standards of Performance for New Sta-
tionary Sources applicable to NO  control for combustion sources are summarized
in Table IV.  EPA's Office of Air Quality Planning and Standards anticipates ad-
ditions to the standards shown on Table IV (6,19).  These additions may involve
both the inclusion of sources not presently regulated and the setting of more
stringent standards for sources under current or impending controls.  This
trend is true also for state and local standards on new and existing equipment.
       The stationary source NOV emission standards set to date and those en-
                               A
visioned for the near future are based on the use of combustion modification
techniques.  This is because the rate of formation of NO  is dominated by com-
                                                        A
bustion conditions and is accordingly cost-effectively controlled by modifica-
tions to the combustion process.
Both thermal NO  — formed from the oxidation
               A
of nitrogen in the combustion air — and fuel NO  - formed from oxidation of
                                               A
bound nitrogen in the fuel —are strongly affected by the local availability
of oxygen in the primary fuel/air mixing zone.  Additionally, thermal NO  is
                                                                        A
extremely sensitive to the local flame temperature.  The modified combustion
conditions which have been shown to suppress NO  emissions are:
                                               A
       •   Decreased primary flame zone 0« level

       t   Decreased time of exposure at high temperatures

       •   Chemical reduction of NO  in the post-flame region
                                      50

-------
       Table V indicates several concepts for achieving these modified combus-
tion conditions and relates these concepts to actual  hardware modifications for
the major equipment categories.  The controls listed  for decreased 0~ are also
generally effective  for peak flame temperature reduction but are not repeated.
The applicable controls are categorized according to  their relevance to current
and near-term application of operational adjustments  or hardware modifications
on existing units or new units of conventional design.   The table also lists
techniques for far-term application to major redesign of new sources.  Generally,
there is more latitude in control selection for thermal NO  than for fuel NO .
                                                          A                 A
All control concepts listed are effective for thermal NO  reduction, but those
                                                        J\
that act mainly to decrease peak flame temperatures are relatively ineffective
for fuel NO .  With the firing of coal and heavy oil, simultaneous suppression
           A
of thermal and fuel NO  is desired.  Here, the preferred control techniques
                      XV
are those that reduce Op in the primary flame zone:  low excess air, burner
modifications and off-stoichiometric combustion (biased burner firing, burners
out of service, overfire air).
       In the NO  E/A, the  applicable equipment/control combinations indicated on
                A
Figure 5 were preliminarily evaluated on the basis of data and operational ex-
perience from field demonstrations.  This evaluation  was done both to obtain an
initial data base to conduct the process engineering  studies and to set program
priorities for the assessment of source/control combinations.  The results of
the evaluation are summarized in Table VI.  These results, particularly the
achievable emission level, the estimated cost and the operational impact, are
currently being evaluated in detail in the process engineering task.
       In Table VI, the current technology column refers to control techniques
which are at least in the field demonstration stage.   In some cases, the current
technology cited is continuing to be refined by field tests.  Emerging technol-
ogy refers to techniques which are still under development as well as to those
which are not currently used due to lack of regulatory requirement.  The most
extensive application of combustion modification technology has been to the
control of new and existing utility boilers.  For gas- and oil-fired units, the
vast majority of controls are retrofits to existing units since few new units
of this type are being sold.  For coal-fired units, most control applications
                                      51

-------
          . .   LEA is also effective for increasing unit
Its use is limited by the increase in smoke or CO emis-
are for new units to achieve compliance with the Standards of Performance for
New Stationary Sources.
       Low excess air (LEA) firing is the most widely used technique for con-
trol of both thermal and fuel NO.,
thermal efficiency.
sions which occur at low levels of excess air.  Low excess air firing is
typically the first technique implemented as part of a control program and
is normally included when other techniques are used.  Its use requires mini-
mal operational modification involving closer supervision by plant personnel,
or in some cases, installation of more sophisticated automatic control systems.
       Off-stoichiometric combustion through use of biased firing, or burners
out of service (on existing units) and overfire air ports (on new units) is
a'very effective technique for control of both thermal NO  and fuel NO .  NO
                                                         X            XX
reduction in the range of 30 to 50 percent has been demonstrated for both oil-
fired boilers (20) and coal-fired boilers (21-23).  Three of the major boiler
manufacturers are incorporating staged combustion in new unit designs as the
means of complying with the 0.7 Ib N02/106 Btu standard of performance for
new stationary sources.  Emission reductions achievable with staging are
limited by operational problems such as CO or soot emissions and convective
section fouling, which may arise at high levels of staging.  Additionally,
there is concern that operation of the primary flame zone at fuel-rich con-
ditions can accelerate watertube corrosion rates (21).
       Advanced burner design is an alternate method for thermal and fuel NO
                                                                            ^
reduction through controlled mixing of fuel and air.  Recent tests with a new
dual register coal burner (25) showed a 35-percent NO  reduction relative to
                                                     J\
conventional burners.  A 50-percent reduction resulted when the dual register
burners were fired with staged combustion.
       Flue gas recirculation (FGR) has been implemented to a limited extent
for control of thermal NO  with the firing of natural gas and oil.  FGR does
                         X
not appear to be as effective for control of fuel NOV emissions, however.
                                                    X
Recent tests with a coal-fired boiler showed only a 13- to 17-percent NO
                                                                        3\
reduction at 15-percent FGR (23).  Thermal NO  reductions achievable by FGR
                                             X
are limited, on some units, by the occurrence of flame instability and boiler
rumble at high levels of recirculated flue gas.
                                      52

-------
       Current developmental activity in the utility boiler area involves
(1) studies of boiler performance and potential  operational problems under
conditions of extensive control (24, 26); (2) development and application of
new burner designs (27) and development of advanced staged combustion tech-
niques (28); and (3) evaluation of NOV controls  with alternate fuels and with
                                     A
the firing of mixed fuels (29).
       NO  control application is currently very limited in industrial boilers.
Their use is expected to increase, however, in view of impending local stan-
dards for existing units in some areas, and the  planned NSPS for new units.
Control technology is in the demonstration stage.  Overfire air and flue gas
recirculation have been applied to gas- and oil-fired firetube and watertube
commercial and industrial boilers (30, 31).  NO   reductions of 30 to 40 percent
were typically achieved.  The results were mixed, however, showing that the
effectiveness of OFA and F6R in industrial boilers is very dependent on equip-
ment design and fuel.  Meanwhile, advanced low-NO  burner development is under-
way by the EPA (32) and in the private sector.  In the long run, this may prove
a more attractive approach than external combustion modifications due to the
wide diversity of commercial and industrial boiler designs.  Stoker-fired
industrial boilers are being tested in the field to determine the feasibility
for western coal firing (33) and their NO  emission characteristics (34).
       NO  controls for stationary gas turbines  have been implemented as part
of state and local pollution control regulations.  Additionally, an NSPS of
75 ppm at 15-percent 0,, is impending.  Current practice is to use water injec-
tion into the primary flame zone to reduce the flame temperature and suppress
NO  (35).  This approach can reduce operational  efficiency and may not be as
attractive in the long term as dry controls centering on advanced combustor
can design.  The incremental annualized operating cost for water injection
systems on industrial utility gas turbines is 0.4 to 1.5 mils per kWh.  This
cost is equivalent to about a 2-percent increase in operating costs.  Emerging
technology for gas turbines is focused on combustor can redesign.  Candidate
concepts include premixing, prevaporization and can aerodynamic optimization
(36).
       Control development for residential warm air furnaces has centered on
burner modifications matched to the firebox aerodynamics and heat transfer
                                      53

-------
characteristics.  A prototype residential  system (37) using an optimized burner/
firebox combination was able to achieve a 65-percent reduction in NO  emissions
                                                                    /\
relative to conventional systems while increasing the fuel consumption effi-
ciency by up to 10 percent.  Long-term field validation tests are planned for
the prototype unit (38).
       There is very limited field implementation of NO  controls for recipro-
cating 1C engines.  Lab demonstrations have shown that 20- to 40-percent emis-
sion reduction is achievable with a combination of spark or injector retard,
leanout of air-fuel ratio, reduced manifold air temperature and derating (39).
A consumption penalty of 2 to 8 percent is associated with these controls,
however.   Use of catalytic converters and exhaust gas recirculation based on
mobile source technology is still under development.  In the far term, NO
                                                                         J\
control by combustion chamber redesign appears to be the most attractive
approach.  The EPA is planning a developmental program to pursue this.
       A field test program is underway to identify emission characteristics
and control options for industrial process furnaces and heaters (40).  Initial
results of tests with minor operational modifications show that the NO  control
                                                                      A
potential for industrial process equipment may be limited by the constraints on
the combustion conditions imposed by the process requirements.  Future tests
will evaluate major equipment modifications on approximately five industrial
process equipment types.
       The results of the control development efforts summarized above are
being used in the NOl  E/A to develop and standardize a control process data
base and to update projections of availablility of advanced technology.  The
remainder of this paper discusses priorities being used in this effort.
                                      54

-------
                                   SECTION 5

              PRELIMINARY BASELINE AND CONTROLLED IMPACT EVALUATION

       Assessment of potential  adverse pollutant impacts from baseline,  un-
controlled operation and operation with NO  controls is  a key objective  of
                                          A
the NOx E/A.  During the initial  stage of the  program, a preliminary  screening
of potential impacts was made to  assist in setting priorities in  the  experi-
mental testing and analysis and in guiding the subsequent, more detailed,
impact assessment.
       The baseline impact screening was done  in three steps:
       t   Compilation of emission data for criteria pollutants,  sulfate,
           trace metals and classes of vapor phase hydrocarbons and  polycyclic
           organic matter
       •   Estimation of maximum pollutant ground level  concentration by a
           dispersion approximation considering average  stack height and
           unit capacity (point sources) and average source density  (area
           sources)
       •   Comparison of pollutant ground level concentration to  estimates
           of threshold pollutant levels for potential  hazard to  human
           health via inhalation
       For each major source/fuel combination, three groups of pollutants
were  identified:  those below 10 percent of the threshold level;  those between
10 percent and 100 percent of the threshold level; and those that exceed the
threshold level.  The results for the last two groups are summarized in
Table VII.  It is stressed that the analysis used in Table VII was for screen-
ing purposes only and that a listing of a pollutant in the table does not imply
that  a quantifiable adverse impact exists.  More data of better quality for
emissions and impact criteria are required together with a more detailed
                                       55

-------
dispersion calculation to quantify Impacts.   It is also noted,  however,  that
there are numerous potential  pollutants for  which the emission  data were too
sparse to include in the screening analysis.  Thus, the availability of  more
emissions data in the future  could add new pollutants to the list as well  as
remove some of those presently listed.
       For operation with NO   controls, the  three-step screening approach
                           A
described above was practical only for the criteria pollutants  due to a  lack
of data for other trace pollutants.  Here, the screening analysis consisted
of a qualitative estimate of  the potential of a given combustion modification
technique for increasing the  emissions of a  class of pollutants.  This was
done by relating the change in combustion conditions due to control, e.g.,
residence time, local temperature, stoichiometry, and the formative mechanism
of the pollutant class.  The  potential for increased emissions  was divided into
three levels:
       •   High potential, where the emissions data unambiguously show that
           applying the NOV control results  in significantly increased
                          A
           emissions of a specific pollutant
       •   Medium potential,  where preliminary screening of formative mecha-
           nisms indicates that NOY control  could conceivably cause increased
                                  A
           pollutant emissions, but confirming data are lacking, contradictory
           or inconclusive
       •   Low potential, where the emissions data clearly show that specific
           pollutant emission levels decrease when the NOX control is applied,
           or where the preliminary screening definitely indicates a similar
           conclusion, even though data are  lacking
       The results of the screening evaluation for boilers firing coal or
heavy oil are summarized in Table VIII.  Similar results were generated for
reciprocating 1C engines and gas turbines which are documented in Reference
16.
       It is emphasized that a designation of "high potential" does not
necessarily  indicate that the pollutant impact is significant, but only that
the emission rate is thought to increase due to the use of NOV controls.
                                                             A
                                      56

-------
Better data are needed to determine if the incremental  impact due to N0x con-
trols is significant compared to the baseline impact.
       As Table VII illustrates, applying preferred NO   combustion controls to
                                                     ^
boilers should have few adverse effects on incremental emissions of CO, vapor
phase hydrocarbons or particulates.  It is true that indiscriminantly lowering
excess air can  have drastic effects in boiler CO emissions, and that partic-
ulate emissions can increase with staged combustion and flue gas recirculation.
However, with suitable engineering during development and implementation of
these modifications, adverse incremental emissions problems can be minimized.
In contrast, residual emissions of sulfate, organics, and trace metals show
medium to high potential for increases in applying almost every combustion con-
trol.  For trace metal and organic emissions, substantiating data, are largely
lacking, but fundamental formation mechanisms give cause for justifiable concern.
In the case of sulfate emissions, fundamental formation mechanisms suggest that
these emissions should remain unchanged or decrease with all controls except
ammonia injection.  However, complex interactive effects are difficult to elu-
cidate, and this pollutant class is sufficiently hazardous to justify expressing
some concern in the present absence of conclusive data.
       The preliminary impact evaluation suggests that with baseline operation,
the emissions of some pollutants may approach a:threshold level of potential
hazard.  Also, combustion modifications show potential  for increasing the emis-
sions of some pollutants.  More data are needed to determine if these effects
have a significant environmental impact and to suggest corrective action if
needed.  The value of this preliminary evaluation is in guiding the gathering
of more data so that the environmental impacts can be better quantified.
                                      57

-------
                                  SECTION 6
                   EVALUATION OF NOY CONTROL REQUIREMENTS
                                   A.

       The extent to which combustion modification controls are needed in the
future is crucial in setting priorities within the NO  E/A as well  as in
                                                     A
guiding control R&D.  To estimate this extent, a preliminary analysis was made
to identify the type and level of stationary source NO  controls to attain the
                                                      A
National Ambient Air Quality Standard for NOp.  Since this preliminary analysis
was intended primarily for screening program priorities, a simple air quality
model (rollback) was used.  Although this model has been used in most prior NOX
strategy studies, it is limited in accuracy and flexibility and will be supple-
mented with more complex models later in the program when more definitive
estimates of control requirements are generated.
       Control requirements were examined for the years 1985 and 2000 for two
Air Quality Control Regions - Los Angeles and Chicago - and for a variety of
growth scenarios and mobile source control strategies.  Los Angeles and Chicago
were selected because they have the highest ambient NOX levels in the U.S.
Furthermore, these two areas have distinctly different sources of NO .  Mobile
                                                                    A
sources account for 68 percent of the NO  emissions in Los Angeles but only
                                        A
38 percent in Chicago.  Coal-fired power plants and industrial sources account
for 45 percent of the NOV emissions in Chicago, whereas no coal is burned in
                        A
Los Angeles.  Therefore, analysis of these two areas gives a good composite
of the controls which should be developed to meet current air quality goals
anywhere in the nation.
       Two growth scenarios for stationary sources were considered - one - a
high growth case - representing a continuation of historical trends and the
other - a low growth case - to consider moderate energy conservation and
rising energy costs.  Two mobile growth scenarios were also considered:  a

                                      58

-------
historical growth case (3.5 percent/year) with N0x emission regulations of
0.62 g/Km (1 g/mile); and a low growth case (1 percent/year) with NOX regu-
lations of 0.25 g/Km (0.4 g/mile) beyond 1981.
       The modified rollback model was used to relate changes in NO  emissions
                                                                   A.
from mobile or stationary sources to changes in the maximum annual average N0«
ambient concentration.   The form of the rollback model used allowed the emis-
sions from each category (mobile, power plants, residences, etc.) to be
separately weighted to account for dispersion effects such as stack height.
       For each AQCR and each mobile/stationary scenario, a succession of
stationary control was added to the modified rollback model until the air
quality standard was achieved.  The order of selection of stationary controls
was based on control cost per unit reduction in ambient NCL concentration.
The data for this selection were generated in the preliminary evaluation of
NO  controls discussed in Section 4.  Control costs were estimated both for
  J\
existing technology and for technology projected to be available in 1985 and
2000.
       The" results show that, to 1985, nearly all available combustion modi-
fication techniques will be needed to attain and maintain air quality for the
two regions.  Beyond 1985, all advanced combustion modification, including
ammonia injection, will be needed for all cases considered except the low
mobile growth case in Los Angeles.  The implications of these results are that
(1) all current and advanced NO  controls will need to be considered in the
                               /\
NO  E/A; and (2) expanded development of advanced combustion modification
  A
controls will be needed to assure that air quality goals are met.
       The above analysis is currently being refined through use of a photo-
chemical reactive air quality model to consider:
       •   Additional AQCR's
       •   Secondary pollutants  (oxidant-hydrocarbon-NO  reactions)
                                                       /\
       •   Alternate N0~ related standards (short term NO^, oxidants)
       •   Source stack height and source density effects.
                                      59

-------
       These refinements, together with revised data on control effectiveness
                                           t

and cost will be used throughout the remainder of the program to update the


estimated control requirements.
                                      60

-------
                                   SECTION 7
                     N0y ENVIRONMENTAL ASSESSMENT PRIORITIES
                       A

     The final step in the initial NO  E/A effort was the setting of priorities
                                     A
among the numerous combinations of combustion sources and combustion modifica-
tion techniques.  These priorities are used within the programs to set the
schedule and level of effort for the detailed process engineering studies and
the field testing.  The prioritization was done to focus the early effort on
those sources and controls which are most important in near term regulatory
control programs.  Minor sources and far term controls will be considered in
the third year.
     The prioritization was done in two steps.  First the major source/fuel
combinations were ranked according to:
     •   Extent of current or impending NO  regulations for the source
                                          J\
         category
     •   Ranking of source NO  emissions on a national basis
                             A
     •   Relative potential for adverse environmental impacts
     •   Current and projected effectiveness of the source in urban NO
                                                                      A^
         abatement
     The ranking for each of these criteria and the overall ranking is shown
on Table IX.  The equipment design types within a source category are divided
into "major" and "minor" categories.  Major refers to conventional designs
likely to be controlled for NO .  These design types will be given primary
                              A
emphasis in the process studies and are candidates for field tests.  Minor
refers to those design types which are obsolete or otherwise unlikely to see
extensive NO  control implementation in the near term.  These design types
            A
                                     61

-------
will be given secondary emphasis in the process studies and will generally
not be candidates for field tests.
     The source prioritization discussed above is extended on Table X to
include consideration of specific source/control combinations.  The table shows
which source/control combinations are to receive major or minor emphasis in the
process studies of near-term applications.  The table also shows preliminary
selection of which advanced source/control combinations will be evaluated in
the later study of far-term applications.  The prioritization of current tech-
nology considered the extent of current applications of specific source/control
combinations and the cost effectiveness of a given control compared to competi-
tive techniques.  Major emphasis will be given to the vast majority of source/
control combinations likely to see significant control in the next 5 years.
The selection of advanced techniques for study in the far-term effort was based
on the developmental status and schedule as well as the potential availability
of competitive techniques.  The selection of advanced techniques also considered
the results of the preliminary evaluation of control requirements which showed
the need for advanced combustion modifications and, possibly, ammonia injection
in the 1980's and 1990's.  Advanced techniques which are being covered by other
assessments efforts (e.g., fluidized beds, advanced cycles) will be given minor
emphasis in the far-term effort.
     The NO  E/A priorities are currently being used to guide three major
           *\                          ,
efforts shown in Figure 1:
     •   Conduct field tests for baseline and controlled multimedia emissions
     •   Generate control technology process studies (control effectiveness,
         cost and operational impact) for the major sources
     t   Select, adapt and apply photochemical air quality model to the esti-
         mation of control requirements
As these results become available, they are being combined with the results of
the supporting tasks on emission characterization and impacts and standards to
generate the environmental assessments of control techniques.
                                     62

-------
                                 REFERENCES

 1.   Copeland,  J.  0.  "Standards  Support  and  Environmental  Impact Statement -
     An Investigation of  the  Best System of  Emission  Reduction  for Nitrogen
     Oxides  from Large Coal-Fired Steam  Generators,"  EPA Office of Air Qual-
     ity Planning and Standards, October 1976.

 2.   Clean Air  Act Amendments of 1977, Hearings  before  the Committee on
     Environmental  Public Works, United  States Senate,  Serial 95-H7,
     February 1977.

 3.   Costle, D.  G., Press Statement on April  18, 1977,  Office of the White
     House Press Secretary.

 4.   Crenshaw,  J.  and A.  Basala, "Analysis of Control Strategies to Attain
     the National  Ambient Air Quality Standard for Nitrogen Dioxide,"
     presented  at the Washington Operation Research Council's Third Cost
     Effectiveness Seminar, Gaithersburg, MD, March 18-19, 1974.

 5.   "Air Quality, Noise  and  Health - Report of  a Panel of the  Interagency
     Task Force on Motor  Vehicle Goals Beyond 1980,"  Department of Trans-
     portation, March 1976.

 6.   McCutchen, G.  D., "NOX Emission trends  and  Federal Regulation,"  presented
     at AIChE 69th Annual Meeting, Chicago,  November 28 -December  2,  1976.

 7.   "Air Program Strategy for Attainment and Maintenance  of Ambient  Air
     Quality Standards and Control of Other  Pollutants," Draft  Report,
     U.S. EPA,  Washington, October 18, 1976.

 8.   French, J. G., "Health Effects From Exposure to Oxides of  Nitrogen,"
     presented  at the 69th Annual Meeting, AIChE, Chicago, Illinois,
     November 1976.

 9.   "Scientific and Technical Data Base for Criteria and  Hazardous Pollutants
     - 1975  EPA/RTP Review,"  EPA-600/1-76-023, Health Effects Research Labora-
     tory, U.S. EPA, January  1976.

10.   Shy, C. M. "The Health Implications of  a Non-Attainment Policy,  Mandated
     Auto Emission Standards, and a Non-Significant Deterioration  Policy,"
     presented  to Committee on Environment  and  Public Works, February 10,
     1977, Serial 95-H7.

11.   "Report on Air Quality Criteria for Nitrogen Oxides," AP-84,  Science
     Advisory Board, U.S. EPA, June 1976.

12.   "Control Strategy for Nitrogen Oxides," Memo from  B.  J. Steigerwald,
     Office of  Air Quality Planning and  Standards, September 1976.
                                     63

-------
13.   "Report on Air Quality Criteria:   General  Comments and Recommendations,"
     Report to the U.S.  EPA by the National  Air Quality Advisory Committee
     of the Science Advisory Board, June 1976.

14.   Personal Communication with M. Jones, Strategies and Air Standards
     Division, Pollutant Strategies Branch,  September 15, 1976.

15.   "Control of Photochemical Oxidants -Technical  Basis and Implications
     of Recent Findings," EPA-450/2-75-005,  Office of Air and Waste Manage-
     ment, OAQPS, July 1975.

16.   "Preliminary Environmental Assessment of the Application of Combustion
     Modification Technology and Control Pollutant Emissions for Major
 :    Stationary Combustion Sources, Volume II Technical Results", Draft
     Report prepared by Aerotherm Division, Acurex Corporation, February 1977.

17.   "Control Techniques for Nitrogen Oxide Emissions from Stationary Sources,"
     Draft Second Edition of AP-67, prepared by Aerotherm Division, Acurex
     Corporation, for Office of Air Quality Planning and Standards, January
     1977.

18.   Cavender, J. H., et al., "Nationwide Air Pollutant Emission Trends, 1940-
     1970," Publication No. AP-115, EPA, January 1973.

19.   Habegger, L. J., et al., "Priorities and Procedures for Development of
     Standards of Performance for New Stationary Sources of Atmospheric
     Emissions," Argonne National Laboratory, EPA-450/3-76-020, May 1976.

20.   Norton, D. M., et al., "States of Oil-Fired NO Control Technology",
     Reported in Proceedings of the NO  Control Seminar, EPRI Special
     Report SR-39, February 1976.     x

21.   Hollinden, G. A., et al., "Evaluation of the Effects of Combustion
     Modifications in Controlling NO  Emissions at TVA's Widows Creek Steam
     Plant," reported in Proceedings of the NO  Control Technology Seminar,
     EPRI Special Report SR-39, February  1976.*

22.   Selker, A. P., "Overfire Air as a NO  Control Technique for Tangential
     Coal-Fired Boilers," presented at EPA Symposium on Stationary Source
     Combustion, Atlanta, Georgia, September 24-26, 1976.          .       .

23.  Thompson, R. E., et al., "Effectiveness of Gas Recirculation and Staged
     Combustion in Reducing NO  on a 560  MW Coal-Fired Boiler," reported  in
     Proceedings of the NO  Control Technology Seminar, EPRI Special Report
     SR-39,  February 1976.x

24.  Crawford, A. R., "Field Testing:  Application of  Combustion Modification
     to  Power Generating Combustion Sources," presented at the  EPA Second
     Symposium on Stationary Source Combustion, New Orleans, Louisiana,
     August  29 - September  1, 1977
                                       64

-------
25.  Crawford, A.  R.,  et al.,  "The Effect of Combustion Modification  on
     Pollutants and Equipment  Performance of Power. Generation Equipment,"
     presented at the  EPA Symposium on Stationary Source Combustion,
     Atlanta, Georgia, September 24-26, 1976.

26.  Selker, A. P., "Overfire  Air Technology for Tangentially Fired Utility
     Boilers Burning Western U.S. Coal Types," presented at the EPA Second
     Symposium on Stationary Source Combustion, New Orleans, Louisiana,
     August 29 - September 1,  1977

27.  Gershman, R., "Design and Scale-up of Low Emission Burners for Industrial
     and Utility Boilers," presented at the EPA Second Symposium on Stationary
     Source Combustion, New Orleans, Louisiana, August 29 - September 1,  1977.

28.  Brown, R. A., "Investigation of Staging Parameters for NO  Control  in
     Both Wall and Tangentially Coal Fired Boilers," presentedxat the EPA
     Second Symposium on Stationary Source Combustion, New Orleans', Louisiana,
     August 29 - September 1,  1977.

29.  Beer, J. M., and G. B. Martin, "Application of Advanced Technology for
     NO  Control:  Alternate Fuels and Fluidized Bed Coal Combustion,"
     presented at AIChE 69th Annual Meeting, Chicago, November 30, 1976.

30.  Bartz, D. R., "Field Testing:  Application of Combustion Modifications
     to Control Pollutant Emissions from Industrial Boilers -Phase II,"
     presented at the EPA Second Symposium on Stationary Source Combustion,
     New Orleans, Louisiana, August 29 - September 1, 1977

31.  Cichanowicz, 0. E., "NO  Control Techniques for Package Boilers:
     Comparison of Burner Design, Fuel Modification and Combustion Modifica-
     tion," presented at the EPA Second Symposium on Stationary .Source Com-
     bustion, New Orleans, Louisiana, August 29 - September 1, 1977

32.  Heap, M. P., "Effects of Fuel and Atomization on NO  Control for Heavy
     Liquid Fuel-Fired Package Boilers," presented at th£ EPA Second Symposium
     on Stationary Source Combustion, New Orleans, Louisiana, August 29 —
     September 1, 1977.

33.  Maloney, K. L., "Western Coal Use in Industrial Boilers," presented at
     the EPA Second Symposium on Stationary Source Combustion, New Orleans,
     Louisiana, August 29 - September 1, 1977

34.  Severs, B. C., "Field Tests of Industrial Stoker Boilers," presented
     at the EPA Second Symposium on Stationary Source Combustion, New Orleans,
     Louisiana, August 29 - September 1, 1977

35.  Durkee, K., et al., "An Investigation of the Best Systems of Emission
     Reduction for Stationary Gas Turbines — Standards Support and Environ-
     mental Impact Statement,"  (Draft) EPA, Research Triangle Park, North
     Carolina, July 1976.
                                      65

-------
36.   Mosier, S.  A., "Advanced Combustion Systems for Stationary Gas Turbine
     Engines," presented at the EPA Second Symposium on Stationary Source
     Combustion, New Orleans, Louisiana, August 29.- September 1, 1977.

37.   Combs, L. P.,  et al., "Residential  Oil Furnace System Optimization Phase
     I," EPA-600/2-76-038, February 1976.

38.   Combs, L. P.,  "Design Optimization and Field Verification of an Inte-
     grated Residential Furnace," presented at the EPA Second Symposium on
     Stationary Source Combustion, New Orleans, Louisiana, August 29 —
     September 1, 1977.

39.   "Standards Support and Environmental Impact Statement - Stationary
     Reciprocating Internal Combustion Engines - Draft," Draft Final Report,
     OAQPS, November 1975.

40.   Hunter, S.  C., "Application of Combustion Modifications to Industrial
     Combustion Equipment," presented at the EPA Second Symposium on
     Stationary Source Combustion, New Orleans, Louisiana, August 29 -
     September 1, 1977.
                                      66

-------
TABLE I.  SUMMARY OF 1974 STATIONARY SOURCE NOX EMISSIONS
          BY FUEL - 1,000 Mg (PERCENT OF TOTAL)
Sector
Utility Boilers
Packaged Boilers3
Warm Air Furnaces
Gas Turbines
Reciprocating 1C
Engines
Industrial Process
Heating
Noncombustion
Incineration
Fugitive
Total
Coal
3,564
(31.0)
679.7
(5.9)


—
—
—
—
4,243.7
(37.0)
Oil
848
(7.4)
886
(7.7)
129
(1.1)
216
(1.9)
(3.9)
—
—
_
—
2,535
(22.1)
Gas
1156
(10.1)
779
(6.8)
190
(1.6)
120
(1.0)
1400
(12.2)
—
—
—
3,645
(31.7)
Total
5568
(48.5)
2344.7
(20.4)
320
(2.8)
336
(2.9)
1856
(16.2)
333
(2.9)
193
(1.7)
40
(0.34)
498
(4.3)
11,478
alncludes steam and hot water commercial and residential heating units
 Includes gasoline
                              67

-------



_J
Ld
U.
>-
Q£
^^
Z
O
1—*
^_
^^
1
1^
^
o:
U_

2^
o
i— i
co
00
I-H
^
LiJ

t-
<:
^
_j
i
0
o.

a
i—i ^~»
_j a>
8*
0
0 O
z o
< -
fV ^^^
t— 1
— 1 — 1
C£. •=>
< O1
"•e" iii

=3 0
c/> z
»— 4
•Si- 2;
I-, DC
O1 ID
i— CO

»— i
»_-i

U-l
,_;
CO
1—



r—
•o >
u ^
+t-» ^
^|
S
E
L* (y

_c
1/1
"*


o
CL






 CM
CM r-»

CO O
VO CM
r^. •*
<£> •£>

n in
in f>
in CM



V)

t- QJ
01 I—
•^ o
o co
03
•o
? 2
••- o
4J «J
=> a.

i iii
i iii



i iii





J3 -Q _O
g
O






J2 J3 J3
^J-
U3



fO CT> LO r—"
o^ r** r™1 co
ro r— oj *
*3h


y-J ^ ^ ^
OJ f-.
CM CO 00 O
ro r*-. *« *•
r— •— O>



r*** f*** cO vo
• • !*•*• v£>
o^ ro ir» f—
CM «—

CM in \o co
ro CM
CNJ o Ch VD


i — • GO P"*- <"*">
csj m i^> co
co m oo n



vi in
cu 
*o •»- c
i- J3 l/> C 4->
3 E Q) LU IO
U_ O C CLt
(_) "^ C 3 *T"
S- J3 •-•
ifl i— a. ai
«^ E S E
>v IO Ql ^
~f oO i ft oj f>_

I
1



1
1





cn
<£>








CM
CO



in
CM
»
IO
'

S
*
^~



CTv
OO
OO

CM
CM

CM

r-.
C3
r~~







^
o
1—









»—
>
1
•o
"o
{/)

£>
•a
4->
c
CU

j_
QJ
O.
O
CM
cn
•§
4->
OJ
£

cn
c
u
•f*
ZJ

>>

•r- a.
> .c
10
+J
5- C
O 0)
-*-* u
u s-
10 x ai
t- O Q.
z
c O
o -o co
tn .— c
f> i— O
§i- -a
*-> £
O O Jo
z: c-J en
IO .Q (j
68

-------

Q.
t—I


o-
LU


2:

O CD
HH 2:

I— HH
 ^
3 Z
OQ - O
a: ct:
O LU
CO LU
u_ u.
o
^ I
z:
 h-
cn i— i
< en
E: o

  XQ
O Z
LU
—I
CO
4-1 J*
fc. C
03 It)
CL Ct
.!£
O S=
^5
_^
O (=
°s
xJSt
s|
.— .X
111 C
3 10
u- Of
 w
z <=
10 0
*Z t.
2 0)
1 o-
0
01
>
4-> • 	
1 §
0
ix-
^- o
sȣ
IS

C
3
CT
LU


i_
O
4^
u

(O U 2tn ^J- CT> — O CJ a* _c s- u. 5 S2 OJ o ca s- 4-> ^> f*> f«> o> CO ro •J3 m ^ o u> CO «* «r s <»> to CO — u lone Furnace S M i- A ** ^j- 00 0 DO r*. ts o» c I_ tJ- '« 2 tn k. Of O CO £ V => in 00 rx. CVJ r-%. o\ 00 00 CM o s o o h~ in g o i o O) c c u_ rti ~3t t tu o CQ £ 4-> Z> 10 o (V) A O tn A •Cf CM S A *r ro ^o o in o> r** o 10 g r»- co p*. e*> is izontally Opposed i- o a: V* L. HI O QO 5 +J z> r*» u> CM PO ro s A o P"» A CVJ CO 0 MJ/hr - 270 MJ/hr/cyl o r-. CM o D> •M fD U £U> V Q. C o: LU CO o .0 3 4-> ^ Of +J fl 3E £ CT> 0) .O t 01 •M IO ^ tf] t- a> o CO T3 n CM •t r*. 0 01 r^- CM r** O O oo o r-. CM 1^- 0 Izontally Opposed o 3C = CM CM u> CMj S U3 U> CX l« o> o ^» g r-^ g ^t- OJ rr> CO 0 »- S O g A 1 •fe> t. 0> +J S 2 ^ o cc XI s. |Q U ID CL CM Ol s A r-H CM o CM LO r- o «*- *J c 0> cn c K5 Lrt L. 0) O CO .£• +J n ro U3 O ro A O n - - <*• m o ^r to 51 t*- g er* ro O CM O _c CJ +J O d tn o> .0 ±3 4-> o CO A ir> 0 IT) O ^- s o CO g o o 00 o t- _c ^»* 3 in o A Ol .0 3 4-> 1- « m 2 C 0) o oa •a OJ a> R) 0 s in r-. CM o CO A O n r*- o ro A r-» \£> O CO ^* h*. CM CO O O Cft r-. r». 0 izontally Opposed o 3= V) i- OJ o oo £ 4-1 Z) <0 cr» o CO A o U Ol S c V o 00 T5 g, CVJ o ro A V> Ol O cr» tr> 03 tn 00 o ir> en t«- *»• 5 4J «t- « s- 1=1 f— in n t- i. (U 3 4-) •)-> ID fO CU ZZ *£ "SS CJ -> vt VI o o en fo r^. CO 0 o o


-------

U.CC
? *.
2 P
1 ,2
u

4J ^^
— £
3 «—
<_J
iX =
°-— S
M5
5^
1
. OJ
Q.


c
1
Q.

CT
UJ


O
-*J
u
X

CO

cn
CM
a
J*s

o
CO
A

•a-
CO
CO

0
CM
m
u>
cn
Cn
CO
o
LT>
CO
in
CM

'«
0
o


(-
u>
o
in

Uatertube
£
cu
c§
Packaged
CM
S

*r
in
s
A
O
O

LO
<*
CO

O
s
ir>
cn
O
s
CO

o



&.

3

^f
in
f
in
CO

0>
c
.£)
u
3
J—
Vt

m
00

s
n'
CM
cn
S
«*
(O

o
L.

??

O
v

1 Uatertube
£
o
o
CO
Packaged
m
CM
in
CM

00
O
o
A
CM

in
to
CO

S
fx
r1-.
CO
CO
01
g
U3
O

in
<§







Central
*/>
0>
u
s
S-
3
U-
L.
<
1
•d-
CM
U3

o
o
n
A
10

0>
CM

If!
r*.
CO

o
cn
r*.
cn
CO
*j-
01
CD
*r
CD
CM
O

1
0
s_
£
3
m
o

V

CM
O
A
O
m
o
co
A
o
A
O
CO
A

ro
S

8
i/)
tn
10
o>
g
rv
eft

0







i.
s:
3
S
A



§
en
CM
r*v
O
o
o
o
«»
en

iS





o'

-C
i
0
r^
CM
A
U
«— t
C
*J
Reciproca
Engines
CO
CM
O

r*.
o
c*>
A
0
A
ul

O
oi

§
CM
CO
CT*
§
*
cst
o>

(/)
10
o
4->
If.

o
^ S2
£ ID
*J 10

*>,
~S£
felt
£S

J2
^ R
Industria
Process C
a
O
A
8
A
S
A
CM

O
f>
A

0>
en

S
r-
CM
cn
en
§
O
Qi

nf
3

I-

-------
        £12-1
00
UU
o
or
o
o

*f.

az
O
u_
01
LLJ
O.
CO
O
O


<
o;
o
z:
LU
cn
O
CO
<:

0>
u
c
2
0)
it-
di
OL









•o
s.
It)

c

+J
IO













(/I
3

IO

CO
* »> y
i/l
S- O
o in
*J CM
IO — '
s-
0) S-
c .c
^s
E
*c ro

3
O
OJ
C3




^
fx*
^
f*v 1^
- CM
en
CO CO CO r^1*
^ ^- «a- in
CM
/v
u.

VO
CO

^"
+J
CO

u>
0

-*^
rt
(—

CO
f
o

•-5
"«v^
c"

o
ro
r~
p-
1
CO
CM
1
CM
P—

TJ

^"* r^
§3
E
e s
O_ CL,


1 ?
'l"~ 'l"~
If- tf*-

1 1
r— r^
OJ *— "* OJ
3 0» 3
<4- ••-> >*-
•r—
p— C r—
•r— CT> *r~
tn -p- in
vi i — in
0 0
t~ 4-> 4-
CL
"O QJ "O
•r- O QJ
i— X X
to '*— * 2£Z





3
3 •»->
.. CO
r>

o

^^_
_o
r^

M 10

+ 0^
^—
p~3
+ -^.
CT!
X C

o
IO
CM

«0
r*««.
i
CM
CM
1
CM
f™
x>
OJ
t/t
o
Gt
O
CL





T3
£
•^
if-

1

i— •
IO
0
o

OJ

•p—
en
i




cn
C
1 -p-
l. S_
QJ 'p—
Q.H-

O r^
•M QJ
^5
OJ lf-
3
-o-o
•r- Q)
U1 X
21

-o 
^ E i/)

IO

1
CM
CM
1
p™*
> —
•o
OJ
T3
d
OJ

"*





T3

o
u

I/I
3
O
c:
•p—
^
3
CO







*— ^
CM
O

4.)
C
CD
U
s_
01
CL
in
r«
«»— ^

1.
O-

in
r~-

1

^
o>


JL.
01
c
3

CO
( — 1
LU
IO
to





4_> ^— ^
3 S-
C — .
•i- 3

4^ CO
*O ^"
o
-C U">

• rxr^.
CO *-^
OJ
c s_

f 3
4_>
CT*
U1 *
ro r^
O A
























g
•I—
2>
0)
SM.

S- >5
OJ "O
T3 3
£T i %
3 in

00 S-
i — i OJ
111 "U
CO C
CO ^3





E
O>
in •<->
D i/>
c

cn 0)
C N
0) -t-
1/1

>> 
i i ^
CO (-1

























































I/I
1-
0
id
s_
OJ
f^
0)
CO
                                                                                                           in
                                                                                                           u>
                                                                                                           O
                                                                                                           O
                                                                                                           1/1
                                                                                                           VI


                                                                                                           i
                                                                                                           I/I
                                                                                                           10
                                                                                                           C7>
                                                                                                          T3

                                                                                                           >
                                                                                                          •i—


                                                                                                          *O  -I"3



                                                                                                          ^  i
                                                                                                           CL  OJ
                                                                                                           C  4->


                                                                                                          -fJ  CO

                                                                                                           0)  4->
                                                                                                          iT  U
                                                                                                               tO
                                                                                                          r-  CL
                                                                                                           O   E

                                                                                                           ^   §
                                                                                                           c   s-
                                                                                                           OJ  -p-
                                                                                                           (U  LU
                                                                                                           a.
                                                                                                               TD
                                                                                                           ID   C
                                                                                                           .c   n?
                                                                                                           ai   i-

                                                                                                           15   &
                                                                                                                CL
                                                                                                           M   3
                                                                                                               l/>
                                                                                                           •o
                                                                                                           c  TP
                                                                                                           R)   S-
                                                                                                                10
                                                                                                            -  13
                                                                                                           >-   c

                                                                                                            -  5
                                                                                                           x  to
                                                                                                          ia   JZ
                                                      71

-------
    6E81-1
a.
UJ
o
o
r—*



g


HH

U-
oo
00
UJ
o
o
a:
a.
o
o
O



>-





V)
'p
I-
c
o
o
 ,_
-* S
I—


dj .,
g|
-.°-
~ 5

c




— 4->
3 0.
J U
C C
S3

c: v»
o c
P- O
l_» .,_
Ift +•>
§1
ss







i
3
>1-
(J.~.
U.
•u
o» c
o
O) •»-
=1 +J

J-
«o

1/1

o> c

o> c

3 C71 trt
IS> O O1
0 S_ ••-

x £ a
Ot C -r~
^

Q> 0» g
Of*- C 0
DC O 'i- -*-1
- irt
"5 1x»
1_ (_> flj
i o 6
CNJ Q.  '
at a>









s- c
0) en
II

s x
oC

V)
a*
c

X

_^

1


w

£Z
i
t- 4->
QJ 1/1
C 3
^'5
CO IB
C

r- CM O
(U Z
3 Of
**- o u
+<> c
OI Ot
1 — I/I l/>
-i- oi .a c:
4-» 0 » 01
2ffSS
•o
*° c* i E^
i'l^
o c  &-
zr CL











'i
3


> c

•-• N






 C
U 4-> OJ
O 0» C •— C
«A •«- in Oi •*->
0> -C C ±3 wt
3 -P- T- .Q
^ O O»r— £
a> -M a* i™ o
OS tfl t- 
- fO
- •— OJ
1H-0
1=
1 O
X Q *r»
9L S- XI
•r- 0 E
**. V 0
-*s. O
t- t=
a> a> o>
£'tf»  o -C
o a. o

O trt CT
4-> ^2.S
3 .0 S-

Lf d)
$- U t.
a> •>- ai
|£|
CO *rt JD
C

M-
r— CSJ O
a> z

^4- O  tfl
!M *•« c
r— -Q O> >,
O OI .C X
> S. 4J 0
C O. 1
*~ § t e*
C7*+J Ot
•— O O -«
Si— N re
0)
u - >, Q.
CM J-
U_ r— Q.-O


0
* to
vi a>
1- 0
or re
< — c
£2


i
s »
*" 1
i««
r^s
O. L. N














„
O
44
U
_a*
c

t.

3- u.


IL.


at ro
U QJ
3 .e
-0 01
0) t




at

•£

**-
o<


1 r—
S i E
J- S- (rt
ex at -r—
Q--C c
3 4-> «J
W) £
*- U
Q yi 2C


O in
C
* 01 -O
£ S :>

« — c
O 3 10
CD **- ff*

i
re
C 01
I-
0 3
Of M- +J
«3 m i-
t- nj ex







X S-
O 01
is
-o
TJ O*
£_ m >r-.
•— O re
ceo.
UJ IT" I/I











O

0

1

1
-U 0)
o c
OJ'r-
J- X
•M -r- -i—
o-a e

••- o

i- U U
ge*
'f- ti- «*-
E  a>

E
C c ID
T3 «— O Q.
a o t— e
VI O , a.
a>
u
I
1
^
S,

^
I



c
0) °-&
it> Wl VI
Of 3 C
i* ^ -*
S8 =
a>

3
0) 4->
(/> rt>

111 I


sii-
01 13 «* O
JB Ul t- -P-
c s-C n
cna> c
.r-V . >.
Ul W V) "O
|SS8













^J


4-9

C
3
OQ






+J

*»-
at


I
Cg|
"S^lt
S8..S
?! a.^^
4 **•». Oi
C c 2
T3 i— Ot
S 0 *"
1J tj OI
L. 01 3 Of
o c-o I
c o at-i-
•-. M S- -W







i L-
uo
cuu:
> c
gc o
o^-
•*-* o
c c* ai
••— w c
T3 > CO
£?s

c:
^u
** o
cu
'co
(Or—

E*^ V
JW1 +•>
Sc
0.3










c
o


VI

1
o
o



c
o
4->
v>
o
e
o
u
s
s

1
^
c
0*

o
ca

01
c
•«—
•a

||
(Z
o
>, X '«-

r— Z I ai

O Ot i/i
jc at c i—
O fc. — *»-
                                           72

-------
       S9l-i
CD
o
O
o:
o
o
u.
o
yj

CD



v>
4-1
C
|
O





£
o
o
c
-C
0
OJ
1—
c
e


1 1 1











o"

o
c

u
OJ
c

t

o













Eg
•2!°

*i




OJ en
t— —
i
4- r-.

•»~ O
 4->

C +* 4->
— C C
OJ O}
*o •*-> S
•r- O OJ T3
c a.<— at
O Q-'O
Jl- 0. OJ
O 3 (U
U. 1/1 c



C -fc>
O VI C
10 •!- 10 OJ
'= t! "B

IHcl;

Bc
OJ
"O C
oj s-
U 3
ID
> X
•o o

•r°° <«
c c
OJ OJ O Ol'i-
r— VI ••- C
-Q IO VI -i- C JC
•r- «J 0 0>0 I/I
vi i- s. cnja ID
VI O S- (0 S. >>
O C O i— "O —
a. -i- o vi u -a u
3 -r- OJ O>
J3 en J3 ' •* -r-J
OJ C C
TJ TI T3 O ••-
N X N 4->  t- 3 E E 0
D.-r- i— O E >^-

1 I/I
1 OHJ
^O fO Q-

trt O ' ^
XVI C
O t- T3 O
z OJ OJ o
e u
x L. c en
O 3 10 C
_,_
-M S
•p* r— O
£ >O C
i. •!-
••-» OJ C C C Ol OJ
lor-ojcnt-cncc
S XI OT<- 0> 1- *r- "i-
o c in f wi e <—
O i, (O OJ +*J O) O C"
Z (X*>T> OT3 0 6

2

^
O
CM
1
O




_C3
O

(M

IT) •
•— O
CM • 	
£
O irt
t.
+ C)

LU  U
4-1 OJ 3

3 0*0.2
S -t~ o
OJ at 4J c
C VI ••- J^
•i- E u


U 10
OJ C7>
C OJ
-^ 3 4->
ai- ti
c •• 5

1 o ai

i
Z.

!.>.|- C
XV) C O
O i- OJ i-
Z OJT3 4J .
a£^S
O 3 ••- "-
_ua o t-
• f.
h1 i

OJ r- >
r- ..- c
-Q J3 1. O
•i- OJ <0 OJ i-
M E 4J^ *J
vi 10 vi *r- ra
Oi— COL.
Q. n_ -r- JD J3



3

evj
i



m

o
O 1

i — in
1 OJ

^~ ^D
.— 	
LO 1
O OJ C
.- I. 0

C*J "O -M
>
' ..II
VI i— JC

'E £S
3  c
a) i— oj
C OJ S-
> t.
^^3

i i»
CJ fU
OJ Ol
C OJ
••- 3 -«J
IO <*^ QJ
C •' 4->
O C 10





X
0 4-
s £
32
.,
$ i

OJ r~ >
r— •.- C
JD -O S- O
VI E *•* r— *J
l/l , VI
+3i+t -r- ai
VI * « — <—
X IO +-> O

1
*o
c

u en

*J J3 CL
C O
OJ^-^—
t r— CO

3S«

X
o
1
^3 "^ C
QJ ^1— O
M -v. .. IO -r-
E OJ Dl C O
•r- C -^ O OJ
a. 3 OJ S c?
o -o ~o S -i-


OJ 1 1
vi c is
CU U •» v> 'F- 4-*
V. C C > -r-
Ol— 0 -r- .« C«t-
C OJ — >> «- O O
•r- 3 4-* OJ 4J OJ •»- 1-
frS E *O r— -r* IO OJ
^ C 3 i— •— O i- i-
i -r- Utt- J3J3 JS^—


JC
Q.
a.

*%>•

i




o

O 1

•— CM
1
in o
00 ~-~
g
+ * *

j. i JQ o ao
_i*— ca oa



T3 ra OJ

5-^3^
*4- «5> t- a>
sill
                                                    73

-------
       ssi-i
•a
 O)
 3
O
O
















£
o
"o
c
x:
u
aj
t-
J1
01
£
Ul










1"

o

.c
u
H™
C
t
3
o











•-^
i
a.

3
CT
UJ




VI
c
3J
g
o
0







g o
^" 1


O> CTl

s. I**-
CO f"-
01 o>
SC r—

CO
C
.° **
4->  *•».
<§^c.
01 Ol

-a O o-

•c- C JC
ID O O
> O 01
«£ 1—


,_ O
^J D>


o
"o
c
u en
01 C
4-1 •*•* "O
4J J3 u.
C O
£:= u
3 4J 01
0 1/1-0

•o

I"S . A

•u o c
Ol •*— O
N 4J 
•O 3 C U
•i- .n o 01
3 £ E-r-J
•— O 5 C
U_ U 1C T-

<*-
O C
3 Ol
§01 *f^
c in
•«- 01
l/l C "O
3 i-
c u. c
>-" O 3
 L. cooi —
•— O r— O -r- C t-
-Q C 01 f- in T- O
m t- a. s- en 4-i 4J
O r- C 3 O— S- S-
a. i i- i/i  S-
d> w i_ a)
J* 3 Oli—
O -O -M -f-
•M C *D O
0^ *•- ^ ^3


o%
o en
i— C
c o
5?+,
01 ai c
+j -o u
4-> 3 Q.

1- ^ >
3 4-> O>
O in T3
x-2


V E

TJ-r- <0 C
OIK- O
.,- L. C 4J
E 01 01  t- UI •<-»
a. 3 ai c
O J3 T3 -t-

S- T-
XO-^
o c c
X •" 3 Ol
in i-
3^-^1/1
O Ol 01 01
_J C C T3
01 1
in c X
ro O 4-1 1 —
Or- O ^ * C C
c a* •!- .a s- o o
U- Q^ E 4J r- *J 4-*
?2clr^c'o£t
} .,- t/> *«~ -r- JD -O ^-^


Q.
a.

-«-

i
M

^^
O

ro '
r- C\J
1
ss
«8
to o
a m

*c ^^^*
«t u. u_
LU O CO
— 1' — 'CO


•0 10 01
a)--- -o
C «- 3 in
•r- 4-> 4J i.
M- ui i- at
1 3 01 f—
IO C ro 5
C3 •*- 3 J3
.= &
*3 O
•c- O
4-1 C
C J=
O U
u a>
t4->
C
t.
O L.
•»- 3
O) W
ai c
Q 0
X
o
.a
0|

^7 >r~
01 H—
•"lir c
S 01 O>
•i— E *i—
4-> S- in
CL 3 01

1
C
S- i- C
3 O S Ol
-a $••-
< c in
xu_ ai
OO C T3
X in te-r-
O *- UJ C
—1 Ol U- 3
Ol 1 1
in E 1C
10 O 4-> — «
Ol U ••> 4->
i. C C ••-
, O

S* E <0 r- 0>
S- c S — i- t
1 -r- in If- Xi — '
a.
a.

o

i
m
^
in

a

o

i «
in o
cc. o
IS V)
Ll- O


ss
_l _1


S- X> in
in 4J o>
3 OJi—
•0 t-r-
E -r- O
« <(- Ji


ot
O
1
 c in
>, ox> c
r— -r- 11 Ol
,
in c r—
ID O C
01 U •- O
S- E
Or— O in f—
C O» t- 4J «
•r~ 3 4«* O E
1- Q. Ol C
r^2 E It J=
( -r- « "8 4->


3

CM
1
r~-
V*
in
CO
O
O 1
• — in
1 CM.
o •
i— O
c
o
« 4J
S- E U
01 ID 01
4-1 OI"->
*4-> C
3 Wf-
in
Oi
|

in
3
                                            74

-------
      SSL-i

-a
(U
o

o
o
LU

aa

•3


*- S
 *sT
•g^xaj-Q
«z c.
ai ai
1—1—3
J3 O CT
10 i- •!-
i— ••-> C
*o o o
> U CO
«t h-

[ *
c ^
S: o
0. "oi S1
*^" 1*

UJ °



>>T>
O +*
r— (A
O CU
^

O) C

s
c
S- r^
t- -p-
3 ••-•
O VI



C
o o


(O ^
-*-* E
>o o
o u

1
S- C i—
•a  S- X «
^3 3 O Ol
(U 1
*o o
Ql O
b-S
*Sli
t- 0.

in c 3
1 1- Ut
03 >*>
O 0
o """"
1 X *—
Cl iC" CO Q.
0^4?
<» £• — CD


^^.
*~i
O

1

O VD
«* 0
1
m o
CNJ —



O S- O C
C 01 •i-^'
O 3 '^ CU O
-j^i- -8---
"io

4J V)
C CU
 tn
SOI
TJ

1
TJ ( 3 C
E c c —
O - c
NM CU 4-> 3

. •
i K I ••• cn"O
ci- c c c ro ai
•r- o o i- o in
a> u •*- &. , — c

r™ OJ  X
«••» Z — CO



So"
CM •


1 1
O
T-- in
o •
"Od
r— •«•

01
•" C

C Ol
ggju.

VI
01
c
'01
tz
CU
u

in
O)
*> 0)
c ra
tin

>>
01  c
ai 1-
•a €

•o-J
<-> a.

o c

1
F^
C X C
i- 0) 01

4) in
!n C •§

w c *J
in o i-
OT- C
ex in 3

c *io ^—-^
tc o •*-
L. in l/> O
 •— o s-
o s- o *J o -P
£01 oj 4-> of s- ••-
> C U in
O3O»i-OU.!=R>
	 i ja -o o u**-- o -o



c
1

Jk£
c




o
c
=>


in

o

o >

CM C\J
1 •
in o
00-




et
LLJ

ro
•*- V)
I. VI 01
+J w U
vi 01 «a
3 U C
•0 O 4-
C S- 3
—. 0.14-
                                     75

-------
TABLE VII.  SUMMARY OF POTENTIAL POLLUTANT/COMBUSTION SOURCE HAZARDS
Pollutant Class/Group
Vapor Phase Hydrocarbons
Total
Al dehydes

Carboxylic Acids
One-Ring Aromatics
Parti dilates

Sul fates

Organics
Anthracene


Pyrene

Benzo(a)pyrene

Benzo(e)pyrene

Perylene
Trace Metals
Be
Cd
Co
Cr

Ni
Pb

V

Zn
Combustion Source

1C Engines
Utility Boilers, all Fuels
Oil -Fired Industrial Boilers
Coal -Fired Utility Boilers
Utility Boilers, all Fuels
Coal -Fired Boilers
Oil-Fired Industrial Boilers
Coal- and Oil-Fired Utility
Boilers

Oil-Fired Boilers
Coal -Fired Residential Units
Coal -Fired Utility Boilers
Coal-Fired Residential Units
Boilers, all Fuels
Coal-Fired Residential Units
Coal -Fired Industrial Boilers
Coal-Fired Residential Units
Coal -Fired Boilers
Coal-Fired Residential Units

Coal -Fired Utility Boilers
Coal-Fired Utility Boilers
Oil-Fired Utility Boilers
Coal- and Oil-Fired Utility
Boilers
Oil-Fired Utility Boilers
Coal- and Oil-Fired Utility
Boilers
Oil-Fired Utility Boilers
Coal-Fired Utility Boilers
Coal -Fired Utility Boilers
Emission Exceeds
Potential Hazard
Threshold



X
X
X
X


X

X
X

X

X

X

X

X



X
X


X


Emission Exceeds
10% of Potential
Hazard Threshold

X
X




X





X

X

X

X



X
X




X

X
X
                                 76

-------

to
a:
o
CO
o


o
Q-

Q-
CO
o
on
o
o
o

to

o
*—<
to
oc
CJ
z
o


o
»~4
I—
— <





















cn
c v>
-t-> 
CD  fl3
C 4.
0 1—
4J 4->
0) =
Dl °
U
C
no
cn
0
4->
03
^
U
•f^
4~*
i-
Q-
OJ
•P
re
OJ
Ul
03
D-
CJ
s- ac
o
CL
o
o
S-
+J
c
o
CJ

X
0
z






o o + + o o o





-t- + + o o o +


t t + + + +




o + + o o + +






+ + + + +4- +


o o o o o o o


£ o o o o o o
«5 O T3
•r- J- to
W C 4-> -r- O
1/1 O  -t- l/» i — O O
O | * fljj ~^ ^^ [ * ^3 tfnt (^ ,^
X ~Ol/> C3U 
UJ IU3 S_ OOI <-> 4-O CO
O) .O OJ-r- 3.C 3 OIo i— 
 
 Q. E O
    3 CL
.C -r-
 C7»T3 5
••- <1J O
                                                                                           (/I I/I
                                                                                           a>
                                                                                         T3 T3 T)

                                                                                         +  +0
                                                           77

-------
a:
o
ac
Q.
o
ex


o
oo
CO
m g>
|i
«/i P
o c
0 '£"•
ill
"'ii
-a
ls£
.2 = S
ft) o
c
0
M- '_£
O J3
£ c S




"s
u. *"•
lj>

0
i/i
a.
'™ c
z.?

^
§,
S
V.
s

•-
ar

3;



X
_
V>
it
•o ±J
O o

i/i C
U •*->
3 .JH
o 'x
 0
C *4-

"^
§U i-
•*- at

*-3 > U)
Tangential,
single and
opposed wall-
fired, turbo

$
4->
13
TJ
O>
, t.

8

 a, "M
—1 _I 3=

x z a:



s: _i 3:
^- CO 1/1


S ?8
W (U 4^ J-
»£ 0 •*- £
^J X
Oi n) Ol Cn
> c
•i- VI in t- •*-
w> 0> nj QT3
ot i- at c>
V 3 E 3O.
X O 10 O E
LU IA t/l _J -r-

"S ^
c ^ a*
o t-s-
•— QJ t.

t? =0
J, s L
O O QJ
« « 0
ifl i/t t/l Ol
1 1 II
4>
3
-f— JT— di
±J 3
"0 -0 01
QJ flj i.
i- .JL -r-

1 1 • —
^ * S
O CS> tJ

«> ~ ? ^ £2
= i -1 x i

a: = z z i



Z _l Z -1
o . S « *



Q> O) 4) 41 O»
o o o o o


i/i (n in (rt i/l
Q) flu Q) JU ^
K- l~
fy at
*• **"
x x
J- O O
i « .«
O t_ t-

i/> U_ Lu
L- i-
•O Of "O O>
c c c c
«u «,^
01 ^ OJ J3 -^C .C
"OJ^ 'pl'ZJ +J 4J
C •— C •— O O
.^ 3 ._ 3 U 0
(/)££ l/> tO

A A Ji O> O)
13 3 3 J3 JS
4-> *«* *J 3 3
L» L. QJ *J -M
O> Ol 1- QJ W
W -^ H! .^ .^
i i I i.i
£ C "7 C IT
i t •— i t
»- in fl i— in
•t~ (O O f~ *Q
O 13 O O O

"*
£
1
X

,



"
-
t.
cn o
C *4-
l/l C
S5"i
Q*
O E t/t
U,-^ 01
u
Moderate
sources,
new sour
r- O*
>»•<"
U L.
i!
o o>
O L.
Industrial,
utility,
simple cycle

*
0 c
If

1
t/l (A

t"-
ac
i

,




CJ
J- G
o •*-
*t- +J
a; >

i/i

c o
Increasi
energy c


t.
41 

(/I .C
Res., Conn.
furnace

Ixi
**- fl*
•o ^
C •'-
ra 
^- O
Negligib
sources ;
new sour

a?
t?
Ol OJ
2 >
O Q
•— y
CO Wt
Turbocha rged
c
Q
4-> OJ
•^-- X
,§ ^"^
'Irt C T?
trt f— C
OJ CT*
t3"^
 i-t 4-

cn
z:

:r



£
CvJ



>


rO
as
to





Turbocharged
naturally
aspirated


c
o
••- crt
CD C.
.* C*
IL. 0)
40
Q.O

2
i
z:

.



*
m





ou

Negligib





Process heat-
ers, furnaces,
kilns

M
at
|
«— c
m o
tft 3
T> e
c o
                                                                              "§
                                                                              -I

                                                                              OJ
                                                                              -O

                                                                              o
                                                                              2

                                                                              S
                                                                              I i
                                                                                 ..

                                                                              II
                                                                                 "
                                        78

-------
     HZ-L

o:
Q.
o
o:
O
to
o

>-
CO
QD





°z
II
8I





5
.
it;

i «j
11
^— s


..
m c U 43
3i I .2 *° v»

S3 *Su
i— i. i— v >>
U- «J «- A U
ffi
s2**5
C
O C
*s
sf








X
II
£««
all





s
i
i
=?
U s
r.
"*- L.
1 tt
31
	 1



- 'S


l$5
le -fe'S'S v? §^-
U .3 C U 3 «J ? J-^K i— t- Z t-
C.O— C£C->- O -0 41
« a) «} *j *J - . c , .
I8i 3?^i £.^fe Sll^
X
O O
•^ ti tj
C ^ 3
£"= S S
-8 T, „
1? 3 3

•o
T
i
z 1



o o ><
oC f
9 Q du i
O o c O
uT uT"° < « 1*^01 "*I -o
iC^ --^ 4» > «— at -^ *v
i— t — fli.a— IQ^O — i-
••- 'i- O Q 3 3 O 3 *J -*- -r-
^-^ 	 — "- -.







1 U
53 Lk «e
-0 O u

o
i
s
2 1

3
g
K X
if i



£

- «J
SE
< ><






S
O

i ii
T3 4J - flj
il ?C
^S SI
53 5C

~
ift





ss'^.s
3S^B
i
«->
£
i
c
•o
o


*
5«
Ifc
S O




g
I
3



_


ii

I
•«- i/i

-3








fi O ^ >i
.•8 .^
Sr-^s







I-
1
S




C
\
°
i


•6
s
i

« u
X" «0
II

•a Of
= 1
*O L
•0 <0
si
ts x




•^
*» Bj

hi
E -
£•;


 -o


"





L.
3 »*
t°£
5fc5






C
I
jQ
K
j



g
*J
01 (VI
"Sao
S|f




VI K»
F.
S S
irt
1
51
L. «>
»* 3
3 4
Jl


"
                                                                                           s
                                                                                           o a>




                                                                                           fe g
                                                                                             u.

                                                                                           S C


                                                                                           I:
                                                                                           o i
                                                                                           *J W-


                                                                                           | I





                                                                                           "E u
                                                                                             u


                                                                                           is
                                                                                           e °

                                                                                           I i
                                                                                           o c

                                                                                           e I
                                                                                           s s
                                              79

-------
                                                                              3S
5

a
       E  M  l/>
       n-i  >•

       ^J  ^  UJ




       %  ^  i


       ^  Q  ^

       M  1  a!

8
of
i
o
z
UJ
a-
VI

,
(S- £•;
« *
j_ ,„_
SB
S |

§ £
CC ft.






LJ
|

1—











t
h- <
a. D
S °
«c oc
o 
tJ 1— _J
UJ O UJ
UJ UJ O

UJ  UJ
o  —

§  £
O  «£


f^      o:
o  x  -x
LU  i—  a
*"»  3C  Z

S  S  ^
oc  m  ca


g  g  *  ui

O  3£  i—  *-«
    LU  -a;  to
t/l  Q£  Z  LU
1/1  —  DC  (—


l/i  o-  i^  S


«S  QE  <  i/l

rt
n

2 §

£ 1
1 I








a
1-
^
9
GENERAL


1
t^
3C

EMISSION










S5
UJ
l/l
i/> «e

*** 0 Z
UJ — O
as a. •>•
                                                                                                                                                          O
                                                                                                                                                          
                                                                             80

-------
                  I   I
•*-   o    ^
o   ja   t,.
 till
o   a*

"^   £
C   O-
•;;   ^
                                                        4_>  —   a.   i.   4->
                                                   J	1
                            -
                            3
                                                                                                     £   £

                                                                                                                  L.
                                                                                                                  O
                                                                                                                  c
                                                                                                            O

                                                                                                            00
                                                                                                                              (U

                                                                                                                              

                                                                                                                              2
                                                                                                             
-------
Industrial  Process Combustion 2.9%
            Noncombustion 1.7%
      Warm Air Furnaces 2.
       Gas Turbines 2.9%
       Fugitive 4.3%
Incineration  0.3%
                      Reciprocal! ng
                       1C Engines
                         16.2%
                      1974 Stationary Combustion Source NOX Emissions
Utility Boilers
Packaged Boilers
Warm Air Furnaces
Gas Turbines
Reciprocating 1C Engines
Industrial Process Combustion
Noncombustion
Incineration
Fugitive
TOTAL
1 ,000 Mq
5,553
2,345
321
338
1,857
333
193
40
498
11,478
1,000 Tons
6,116
2,583
353
372
2,045
367
212
44
548
12,640
Percent
Total
48.4
720.4
2.8
2.9
16.2
2.9
1.7
0.3
4.3
100
            Figure 3.   Distribution of stationary  anthropogenic NO
                         emissions  for the year 1974 (stationary fueT
                         combustion:   controlled NO   levels).
                                     82

-------
           SESSION I:
SMALL INDUSTRIAL, COMMERCIAL AND
      RESIDENTIAL SYSTEMS
         ROBERT E. HALL
            CHAIRMAN
                  83

-------

-------
DESIGN OPTIMIZATION AND FIELD VERIFICATION OF
      AN INTEGRATED RESIDENTIAL FURNACE
                     By:

         L. P. Combs and A. S. Okuda
  Rockwell International/Rocketdyne Division
            Canoga Park, CA  91304
                         85

-------

-------
                                   ABSTRACT

The purposes of this project are:  (1) to further optimize the design of an
integrated, low-emission high-performance residential warm-air oil furnace, a
prototype version of which has been used earlier for proof-of-concept demon-
stration of burner and firebox design principles and (2) to verify its pol-
lutant emissions and thermal efficiency performance in actual residential
installations over an entire winter-heating season.

The project is being conducted in two sequential phases.  The first phase has
involved analysis and laboratory testing to further optimize the furnace
design and to document its pollutant emissions and thermal efficiency per-
formance.  The goal of this phase was to achieve a commercially producible
design that is safe for use in residences.  The optimized furnace's capabil-
ities to operate with alternate fuels also are being evaluated.  In prepara-
tion for the next phase, the logistics of residential field testing have been
delineated and appropriate arrangements made.
The second phase of the project is just getting underway and will be concerned
with:  construction of a number of integrated furnace units to be field tested,
their installation in selected residences, and their operation during an entire
annual heating season.  The emission and performance characteristics of each
test unit will be determined initially and remeasured monthly.  Any tendency
for drift in emissions or performance will be allowed to continue from month
to month without corrective adjustments unless it is necessary to correct an
unsafe condition or one producing excessive pollutant emissions.

The paper summarizes the results of the first-phase laboratory investigations
and describes the plans for field testing.  It is based upon work performed
                                     87

-------
for the U.S. Environmental Protection Agency under Contract 68r02r217&.  The
EPA Contract Officer was Mr. G. Blair Martin, Combustion Research Branch,
Industrial Environmental Research Laboratory, Research Triangle B$r|t? Hprth
Carolina.
                                      88

-------
                                INTRODUCTION

Under a previous EPA contract, design criteria were determined whereby gun-
type pressure-atomizing oil burners, such as are commonly used in residential
and commercial space-heating systems, may be modified so that they produce
substantially lower emissions of oxides of nitrogen (NO ) and burn smoke free
                                                       JH
at more efficient operating conditions.  Those design criteria were used to
modify existing burners of two sizes—a 1.05 ml/s (1.0 gph) residential burner
and a 9.47 ml/s (9.0 gph) commercial burner—and were shown to be valid in
laboratory testing (Ref. 1).

Further laboratory research with the smaller of those two.low-emission burners
provided additional design criteria for fireboxes matched to the burner to
achieve even lower NO  emissions (Ref. 2).  Thereafter, proof-of-concept exper-
iments were carried out in the laboratory using a prototype residential warm-
air furnace embodying the several design criteria (Ref. 3 and 4).  The result-
ant NO  emissions were reduced to about 35% of the estimated average from
      X
comparable, existing, installed units.  Further, the laboratory performance
data showed that the prototype furnace's cyclical efficiency should be 10 or
more percentage points higher than the estimated average of the existing resi-
dential oil furnace population.

The present investigation is a logical continuation from those encouraging
laboratory results.  Potential benefits of commercializing the derived tech-
nology are to be demonstrated by conducting field tests of several low-emission
furnaces in actual residences.  First, however, it was appropriate to effect
further refinements of the burner and furnace designs, partially to further
optimize emissions and efficiency performance and partially to improve commer-
cial producibility.  Thus the two objectives were:  (1) beginning with the
                                      89

-------
prototype furnace tested earlier, to further optimize the design of an inte-
grated, low-emission high-performance oil-fueled residential warm-air furnace,
and (2) to verify its pollutant emissions and thermal efficiency performance
by operating units over an entire winter heating season in actual residential
installations.

Further design optimization was approached by making modifications to the pro-
totype low-emission furnace described in Ref. 3, and retesting the unit in a
Rocketdyne laboratory.  Eventually, this led to the construction and verifica-
tion testing of a second, all-new integrated furnace unit prior to assembly of
six units of the final design to be installed in the field.  Simultaneously
pursued was the definition of requirements for selecting test locales, arrang-
ing for local support, selecting host residences, and the logistics of shipping,
installing, activating, servicing, measuring performance and, finally, removing
the experimental furnaces and restoring the host homes' heating systems to
their former conditions.

At the time this paper is presented, all the preparatory work will have been
completed and the test units will be ready for installation and initiation of
the field test period.  Thus, this paper constitutes a progress report on the
design finalization and the arrangements made for the field verification.

                 SUMMARY OF THE PROTOTYPE FURNACE INVESTIGATION

The experimental furnace unit tested previously (Ref. 3 and 4) has been called
a "prototype optimum furnace."  For clarity in this paper, that earlier test
unit will be designated the "prototype furnace" and the further optimized unit
derived from it will be referred to as the "integrated furnace."  In this sec-
tion are given brief descriptions of the prototype furnace and the laboratory
facility in which it was tested, followed by a summarization of its emissions
and efficiency performance.
                                      90

-------
PROTOTYPE FURNACE DESCRIPTION

The prototype test unit was based on modifying or replacing several specific
components in a commercially available warm-air oil furnace of contemporary
design*.  The prototype furnace and the advanced technology modifications made
to achieve it are illustrated in Fig. 1.  The central line drawing is an inte-
rior side view of the furnace; the front is on the left-hand side.  Each of
the photographs surrounding the drawing illustrates a modification made and an
arrow denotes its location in the furnace assembly.

The optimum burner head is illustrated alone although, in actuality, the fur-
nace's entire burner was replaced with the optimum burner that had been tested
extensively before, rather than replacing only the head.  The combustion air
fan in the replacement burner was also fitted with a quiet stator plate to
prevent coupling of combustion air flow pulsations to combustion in the
chamber.

The unit's original firebox, a rather typical 0.249 meter (9.75 inch) inside-
diameter refractory-fiber-lined design, was replaced by a larger 0.303 meter
(12 inch) inside—diameter uninsulated air-cooled firebox.  Its outside surface
was heavily finned to ensure adequate heat extraction from the flame zone, a
key contributor to the reduction of NO  emissions.  The machined and welded
                                      X
mild steel prototype firebox was more massive than its predecessor by about
50 kg (110 Ibm) and so constituted a substantial heat sink.  Built as an
experimental tool, the firebox was intentionally overdesigned in the interests
of maintaining wall temperatures as nearly uniform as possible and retaining
high wall temperatures during standby.

Two related modifications were concerned with the combustion air supply. First,
combustion air was admitted to the furnace through a "sealed air" supply sys-
tem, adopted from Ref. 5.  Simply stated, this means that combustion air,
together with air drawn into the flue through a barometric pressure control
*A Lennox Model 011-140 furnace, supplied by Lennox Industries, Inc., was
 utilized.
                                     91

-------
damper, is brought in from outdoors rather than consuming heated (and perhaps
humidified) air from within the residence.  Pollutant emissions usually are
not influenced by this change, but fuel consumption may be reduced by 5 to
10% or more (Ref. 5).  Second, a separate filter was provided for the combus-
tion air so that the burner could be tuned for operation with close to minimum
(~15%) excess air without an accumulation of lint, hair, etc., on the burner
air passages, fan blades, or head forcing it into a smoky and/or high CO
condition.

The combustion air inlet to the burner was fitted with a weighted damper which
closed automatically when the burner was turned off.  It was designed to elim-
inate the draft air loss of heat up the flue during furnace standby.  Fuel
savings effected by this device will probably average between 2 and 3%.  This
device also included a separate butterfly damper whose position controlled the
flow of combustion air.

Minor modifications also were made to a few other furnace components.  Louvers
in the burner vestibule closure panels were covered to make the ves.ti.bule part
of the sealed air system.  Inside the cabinet, some minor structural reinforce-
ment was added to support the heavier firebox, and flat, vertical-panel baffles
were added in the warm-air passages to force the warm air to  flow over the
finned firebox.  Otherwise, the warm air  blower and filter, the compact heat
exchanger, the furnace cabinet, and all electrical circuits and controls were
left unchanged from the original stock furnace.

LABORATORY FACILITY

Performance of the prototype  furnace was  evaluated in an outdoor  laboratory
facility having provisions for measurement of pollutant  emissions,  operational
characteristics, and  thermal  efficiency.  Figure  2 is a  schematic of  the  fur-
nace evaluation  system; it shows  the installation of gas and  air  flow ducting
and a variety of instrumentation.  Basic  thermal performance  measurement  tech-
niques conformed with requirements of ANSI Z91.1-1972  (Ref. 6).   Other instru-
mentation were added  to provide  (1) enlarged understanding  of furnace behavior
and (2) data for calculating  cycle-averaged thermal efficiency.
                                      92

-------
Constituents in the flue gases were measured by continuously withdrawing a gas
sample from the center of the flue, at the location denoted in Fig. 2, and
passing it through an analysis train.  The analytical system provided for man-
ual spot samples for smoke and for continuous analyses for 0_, C0_» CO, NO,
and unburned hydrocarbons (UHC) species remaining in the dry gases after their
passage through condensible traps, filters, and driers.  Detailed information
on the setup and operation of the train, instruments used and their ranges,
data processing, etc., has been given previously in Ref. 1 through 3.

The furnace flue thermal losses were determined by making measurements to sup-
port flue gas heat balances.  Combustion gas mass flowrate was back-calculated
from measured fuel flowrate and stoichiometric ratio (as determined from flue
gas composition measurements).  The flue gas exhaust temperature was measured
in an insulated flue pipe with a thermocouple located 0.46 meters  (18 inches)
above the centerline of the heat exchanger flue exit.  Flue draft, gas compo-
sition, and smoke measurements were taken at successive 0.0317-meter  (1.25-
inch) increments downstream of the thermocouple, respectively.

Steady-state thermal efficiencies were derived from the steady-state flue gas
temperatures and C02 concentrations (Ref. 6).  During cyclical operation in
which steady-state was not reached, values for those parameters just prior to
burner cutoff were used in the same manner to get approximations of steady-
state efficiencies.  Burner firing times of 10 minutes gave such pseudo-steady-
state efficiencies that were indistinguishable from those derived  from steady-
state measurements; those calculated from 4-minute burner firing time data
were approximately 1% higher than the steady-state efficiencies due to heat
being absorbed by the cool furnace components resulting in a lower flue gas
temperature.

Data were also measured to support calculation of cycle-averaged efficiencies.
However, the variable ambient air temperature in the outdoor facility intro-
duced considerable scatter in the data  (Ref. 3).  Indicated differences between
the mean steady-state and mean cycle-averaged efficiencies were smaller than
that scatter, so steady-state was used as the main basis for comparison.
                                      93

-------
PROTOTYPE FURNACE RESULTS

Pollutant Emissions

Prototype furnace gas concentrations of nitric oxide are plotted versus stoi-
chiometric ratio in Fig. 3.  A shaded region near the middle of the graph
indicates that a large majority of existing residential oil furnaces release
between 1.3 and 2.2 g NO/kg fuel burned.  An estimated existing furnace aver-
age of 1.8 g NO/kg fuel may be used for evaluating the potential impact of
applying candidate NO  reduction techniques.
                     JV

Measured NO emissions from the stock furnace fell on the high side and above
that typical range; at a nominal 50% excess air operating point, it produced
2.2 g NO/kg fuel burned.  Measured NO emissions from the prototype furnace
were much lower, falling between 0.5 and 0.75 g NO/kg fuel over the stoichio-
metric ratio range of interest.  Tuned  to the intended normal operating con-
dition with only 15% excess air, the unit produced 0.63 g NO/kg.  That corre-
sponds to reductions of about 72 and 65%, respectively, from NO  emissions
produced by the stock furnace (at its nominal operating point) and by the
average estimated for all existing installed units.

Carbonaceous emissions  from the prototype furnace unit also were acceptably
low at those conditions, as indicated by the lower than No. 1 smoke.  A com-
parison of values in Table I shows that CO and hydrocarbon emission levels
from the prototype furnace were somewhat higher than those measured fpr the
stock furnace but were  quite comparable with the average tuned values measured
in the field survey of  Ref. 7.

Efficiencies

Pseudo-steady-state efficiencies for the prototype furnace are compared with
those for its stock predecessor in Fig. 4 by superimposing values  calculated
from 4th-minute data in cyclical runs on an efficiency decrement plot.  The
performance curve for the  original stock furnace fell well below  (i.e., higher
efficiencies) the shaded band representative of a large majority of existing
                                     94

-------
installed residential heating units.   The stock unit could be tuned to a mod-
erately low 50% excess air nominal operating condition where its net flue gas
temperature averaged only 180 C (325  F).   The resultant steady-state gross
thermal efficiency was 82.5% (i.e., the stock furnace was among the higher
performing units in the existing population).

Thermal efficiency levels achieved by the prototype furnace were qualitatively
the same as those of the stock furnace.  However, as is evident in Fig. 4,
flue gases leaving the prototype unit were 40 to 55 C (75 to 100 F) hotter
than those from the stock furnace, and the efficiency decrement due to the
higher net flue gas temperature was offset by operating the prototype unit
at substantially lower stoichiometric ratio.  This apparently anomalous be-
havior was believed to be caused by warm-air jets outside of the firebox
bypassing some of the main heat exchanger, a condition which was thought to
be relatively easy to correct.

The 82 to 83% pseudo-steady-state thermal efficiency exhibited by the proto-
type furnace was close to the maximum achievable in noncondensing flue gas
residential systems.  Taken alone, this is not unique, since comparable effi-
ciencies are attained by some current commercially available units  (as exemp-
lified by the stock furnace that was converted into the prototype).  What is
unique and important about it is the demonstration that near-maximum steady-
state efficiency and near-minimum NO  emissions can be obtained simultaneously.
                                    X

                    INTEGRATED FURNACE DESIGN OPTIMIZATION

The prototype furnace test results confirmed the feasibility of applying the
several newly developed, low-emission oil burner and firebox design criteria
to residential space heating equipment.  The experimental prototype unit came
very close to satisfying all of the pollutant emission and efficiency objec-
tives for which it was designed.  Operationally, its behavior was quite com-
parable with current commercially available furnaces.  A 500-hour duration
test, equivalent to about one-tenth of an average heating season, indicated
that the unit might serve through an entire winter heating season without
requiring substantial maintenance and without exhibiting appreciable shifts
                                      95

-------
in operating conditions or pollutant emission levels.  Before undertaking such
a next logical step in the proof-of -concept demonstration of the furnace design
criteria, however, it was appropriate to further improve some features of the
prototype unit.

MODIFICATIONS FOR THE INTEGRATED FUKNACE

Optimum Burner

The standby draft control device for the combustion air inlet to the burner
was redesigned.  The butterfly valve was removed and air flow control was
accomplished by providing a mechanical stop to limit the opening of the
weighted flap damper.  A microswitch was incorporated in the stop to provide
a positive electrical indication that the draft flap opens when the burner
is turned on; the burner control circuitry was redesigned to take advantage
of this safety feature.

An internal clutch was inserted in the burner's drive shaft for the fuel pump.
The reason was to give a quicker cutoff of fuel flow through the oil nozzle
and better eliminate post-firing dribble.
A stamped annular sheet metal ring was added to the sheet metal burner head
and welded to the external face of its choke plate.  In addition to improving
the burner head's appearance, this improved the alignment of the head with
respect to the blast tube and oil nozzle and also eliminated leakage pf ~15? of
the combustion air through cracks and holes at the head's perimeter.  The
modified head and combustion air inlet assembly are illustrated in Fig. 5.
Finned. Air-Cooled Firebox
The fabricated steel firebox, with its large external steel fins welded to a
0.64-cm (0.25-inch) thick steel shell was inordinately expensive to make,
heavier than desirable and, as well, extracted more than the designed-for
quantity of heat from the flame zone.  Therefore, it was replaced with a
lighter weight, cast-formed combustion chamber.  Serious consideration was
                                     96

-------
given to making a cast aluminum firebox.  The coefficient of thermal expansion
for aluminum is high enough, however, that rather high stresses were calculated
at the roots of the fins and at the firebox-to-heat-exchanger joint.  In prin-
ciple, yield conditions could be avoided by maintaining low wall temperatures
but, even then, cycle-fatigue failure was predicted to lead to an unacceptably
short firebox lifetime of 3 to 5 years.  Therefore, a cast iron was selected
as the firebox construction material.

A photograph of the cast-iron firebox is shown in Fig. 6.  The internal dimen-
sions and burner insertion port duplicate those of the fabricated steel combus-
tor; otherwise, the designs are quite different.  To facilitate welding the
stock fabricated steel heat exchanger to the cast-iron firebox, a rolled steel
ring is cast integrally into the wall of the firebox discharge opening.  This
provides a positively sealed joint as well as a much smoother surface for the
external air to flow over.  All of the cast external fins are shorter than
were the fabricated ones, and both their heights and spacings are graduated
to promote more-nearly uniform wall temperatures.  The shorter fins were
designed to extract a lower fraction of the overall heat transfer from the
firebox; the cast wall temperatures were expected to be about 56 C  (100 F),
on the average, higher than those of the former finned steel firebox.  The
cast firebox, together with its support legs and burner mounting plate, has
a mass of 34 kg (75 Ibm).

Warm-Air Coolant Distribution

The vertical panel baffles positioned close to the fin tips in each side of
the prototype furnace's firebox were reshaped to improve the air-flow distri-
bution to the outside of the main heat exchanger.  The baffles were bent
slightly at about the midsection of the firebox to begin a gradual  flaring
from the constricted cross section to the full heat exchanger cross section.
The location of each baffle remained the same; however, since the fins were
shortened, this placed them about 0.064 meters (2.5 inches) from the nearest
fin tip, as opposed to nearly touching in the prototype furnace.  Also, new
baffle panels were added adjacent to the outer side panels of the heat ex-
changer section.  This was done to reduce the flow gaps between the heat
                                     97

-------
exchanger and the wall by 0.025 meter (1.0 inch), thereby reducing bypass
(i.e., no heat transfer) air flow.  These changes were made to restore the
coolant air to the outer sections of the heat exchanger to improve the overall
heat transfer of the furnace.

LABORATORY TESTS OF THE INTEGRATED FURNACE

A first specimen of the integrated furnace was built by converting the protQr
type furnace via the foregoing modifications.  It was tested in the same labr
oratory facility and with the same procedures as was the prototype unit.  A
basic test matrix provided for measurement of pollutant emissions and perform-
ance as functions of burner firing rate, overall stoichiometric ratio, combus-
tion air supply temperature, firebox draft, and warm-air flowrate.  The test
plan also allowed for retesting in the event that one or more components needed
further refinements.  Discussion of the extent to which this option was needed
is included with the description of test results, below.

In addition to hot-firing the integrated furnace, cold-flow tests of the burner
were used to measure the effects on the combustion air fan characteristics of
several complications added to the stock* burner's combustion air circuit.

Laboratory Test Results

Burner Blower Characteristics.  The combustion air blower's characteristics
were measured to establish the operational limits of the combustion air system.
Tests were made with the air flow throttled either at the air inlet or at the
exit.  This was done by mounting the burner inside or outside, respectively,
of a large sealed compartment and measuring the  compartment pressure and air
flow into or out of it.  A calibrated laminar flow element was used for accu-
rate measurement of air flows.
Throttled outlet characteristics are shown in Fig.  7.  The uppermost curves
show that the presence of the quiet-stator has very little effect on the
*The R. W. Beckett Co.'s Model AF burner was the stock unit underlying the
 optimum burner.
                                     98

-------
blower's basic throughput characteristic.  Successively lower curves show how
the achievable air flow is impacted by adding, in sequence, the inlet draft
flap assembly, a static disc within the blast tube, and the optimum head to
complete the burner assembly.  Dashed lines indicating pressure drops of the
head, static disc, and draft flap were obtained by subtraction of appropriate
characteristic curves.  The nominal design operating point, with a firing rate
of 1.05 ml/s (1.00 gph) and a stoichiometric ratio of 1.15, is indicated by
                                                                        :h
                                                                        2
an arrow at 0.0125 m /s (26.5 scfm).   To achieve that flowrate without throt-
tling the exit, the firebox would have to be pressurized to about 60 N/m
(0.24 inch water column).  The usual residential furnace practice, however,
is to throttle the inlet to the blower, rather than pressurizing the firebox
or throttling the exit.

Effects of throttling the blower inlet (i.e., limiting the draft flap opening)
are shown in Fig. 8 along with the effects of imposing a negative (suction)
pressure condition on the air supplied to the inlet.  The latter condition is
encountered sometimes in conventional furnace practice by operating a furnace
in a fairly leak-tight room or house, so that the burner blower reduces the
pressure in the furnace enclosure.  In the integrated furnace, pressure in the
sealed burner vestibule also will be below ambient outdoor pressure by the
pressure drop of the air supply system.

The throttled inlet characteristic curves show that, if the burner inlet is at
sea level pressure, the nominal design conditions can be achieved with the
draft flap only one-eighth  open; opening it further should allow tuning for
stoichiometric ratios (SR) from 1.15 to about 1.35.  Reducing the inlet pres-
sure exerts a linear reduction effect on the achievable flowrate; again, the
margin of adjustability from the design point would be eliminated by about
      2
60 N/m  (1/4 inch water column) suction at the blower inlet.  This condition
is not likely to occur unless the combustion air filter is allowed to become
very dirty.

What the burner requires is a given weight of air for a given firing rate and
stoichiometric ratio.  What the blower supplies is a given Volume of air at
its local (nominally inlet) conditions.  If the furnace is installed at a high

                                     99

-------
altitude, a greater volume must be delivered to obtain the required weight of
air.  The optimum burner, fired at 1.05 ml/s (1.0 gph) and a 1.15 SR, would
have a fully open inlet and no margin of adjustability at about 1500 meters
(4900 feet) elevation.  In practice, the furnace would have to be downrated
to a lower firing rate or supplied with a higher capacity blower to allow
servicemen at those altitudes some margin of tuning control.

Pollutant Emissions.   During most of the testing of the initial integrated
furnace, obtained by conversion of the former prototype unit, the measured
emissions of air pollutant species were not significantly different than those
reported earlier (Table I and Fig. 3) for the prototype furnace.  However, as
those tests were nearing completion and a second, newly constructed integrated
furnace was being readied for verification testing, it was discovered that the
Bacharach smoke tester was malfunctioning due to a cracked piston cap and a
faulty check valve.  Thus, the indicated smoke readings were low for the then
current tests and for previous tests over some indeterminant time period.
Because of the conversion sequence from the original stock Lennox to the pro-
totype to the integrated furnace, it was not possible to go back and retest
those earlier configurations.  However, the original stock Lennox burner was
tested in a second stock Lennox furnace and several exploratory modifications
were made to the integrated furnace and its optimum burner to simulate the
prototype configuration.  It was concluded from the results that the problem
with the smoke meter probably developed sometime between the stock Lennox
furnace tests and the prototype furnace tests reported in Ref. 3.  That is,
the prototype unit's smoke emissions actually must have been appreciably
higher than were reported.  This put a serious question on the integrated
unit's capability to be tuned to the nominal (15% excess air) design point.

The integrated furnace was retested, therefore, to determine operating condi-
tions which produce acceptably low smoke.  Some of the results are plotted in
Fig. 9.  At the 1.05-ml/s (1.0-gph) firing rate, smoke emissions were sensi-
tive to excessive over-fire draft conditions and in all cases exceeded No. 1.
(Achievable upper values of stoichiometric ratio were restricted by combustion
air flow limitations, discussed earlier).  At the 0.79-ml/s (0.75-gph) firing
                                     100

-------
rate, there was almost no sensitivity to over-fire draft variations and less
than No. 1 smoke was generally measured if excess air exceeded 20%.  It was
also found that, at this firing level, smoke emissions could be further reduced
by insetting the burner head a small distance—0.025 meters (1.0 inch)—into
the combustion chamber.

Ostensibly, 1.15 can be achieved with the latter combina-
tion of firing level and burner/firebox configuration.  However, the gradient
in smoke number with decreasing stoichiometric ratio is quite steep so that
attempting to tune the burner to precisely 1.15 SR involves the risk of pro-
ducing excessive smoke if it is missed, or later drifts, by even a small amount.
From these results, it is apparent that the integrated furnace should be rated
as a nominal 0.79-ml/s (0.75 gph) unit, rather than the previously stated 1.05
ml/s (1.0 gph), and tuned to burn with about 20% excess air.  At these rede-
fined nominal design conditions, the measured steady-state and cycle-averaged
emissions of CO were 0.25 and 0.60 g/kg, respectively, and of UHC 0.02 and
0.04 g(as CH,)/kg, respectively.  Measured emissions of NO were somewhat higher
than at the former nominal design point, principally because of the stoichio-
metric ratio change.

Thermal Efficiency.  Relocation of the baffles in the warm-air passage around
the finned firebox did not produce the expected ~2% increase in thermal effi-
ciency.  Several different baffle arrangements were tried, including some in
which small local baffles diverted parts of the air flow to better cool hotter
sections of the main heat exchanger.  Only small variations (~l/4 to 1/2%) in
efficiency were indicated and the initial integrated furnace configuration was
as good as any other.  It was restored, therefore, and attention was turned to
the internal distribution of combustion gases within the heat exchanger.
The heat exchanger construction involves six vertical flat-panel elements,
three on each side of a central cylindrical dome extension of the firebox
cylinder.  Combustion gases flow vertically up into the dome, turn 90 degrees
into a rectangular exit on the dome's rear wall into a splitter manifold which
turns them through 180 degrees and distributes them among the six panels. Flow
out of the panels is collected in a manifold at the front of the furnace, from
                                     101

-------
which it is ducted into the vertical flue.  A set of holes was drilled in the
flue gas collector manifold to permit measurement of the temperatures of the
                                                         \-
gases leaving the several panels.  Traverses made with a long chrome1-alumel
thermocouple revealed quite large temperature differences among the panels and
strong gradients within each panel.  In general, the gases were hottest in the
outermost panels and coolest in the innermost ones and, in each panel, were
much cooler at the bottom than at the top.

Anticipating that making the combustion gas temperatures leaving the panels
uniform would maximize overall heat transfer, long vertical vane dampers were
designed for insertion in the panel inlets.  Initially, only two such dampers
were installed, one in each of the outermost panels.  With just those two
dampers in their most open positions, the range of measured temperatures was
reduced from a 269 C (484 F) difference [418 (784) minus 149 C (300 F)] to a
139 C (250 F) difference [343 (649) minus 204 C (399 F)] and the temperature
gradients from the panel bottoms to tops were virtually eliminated.  Nonethe-
less, the average flue gas temperature was lowered by only 2 C (3.6 F); this
represents less than 0.1% gain in thermal efficiency.  This approach was not
pursued further.

The experimental evidence suggested that  the thermal efficiency of the inte-
grated furnace cannot be increased by improving either the warm air or the
combustion gas flow distributions, i.e.,  a larger heat exchanger is needed to
achieve higher efficiency.  Because the integrated furnace does have more heat
exchange surface (the finned firebox) than the stock Lennox furnace, there
remained a question as to why that stock  unit exhibited higher thermal effi-
ciency.  In an attempt to answer that question, tests were made with a second
stock Lennox unit.  Although it was manufactured approximately 2 years later,
this second unit appeared to duplicate the first one except that the burner
head had been redesigned.

The second stock Lennox furnace was fired both with its own burner and with
the burner unit from the original stock Lennox unit.  In neither case did it
achieve anywhere near the low flue gas temperatures (high thermal efficiency)
measured earlier with the original stock  furnace.  In fact, this second stock

                                      102

-------
Lennox and the integrated furnaces achieved very nearly equal thermal efficien-
cies at a given stoichiometric ratio.  Thus," although the question regarding
the performance of the first stock Lennox remains unanswered, it was decided
to proceed with preparing field test units of the integrated furnace'design
without further alterations.

Measured steady-state efficiencies of the integrated furnace at two firing
levels are illustrated in Fig. 10.  A value of 85 to 86% is indicated at the
0.79 ral/s (0.75 gph) firing rate and 1.20 SR.

                              FIELD VERIFICATION

As-a final step in the proof-of-concept verification process, it is planned
to install and field-test six integrated furnaces in selected host residences
during the 1977-78 winter heating season.  Detailed plans for the field testing
have been completed but the test period will not commence until approximately
1 month after this paper is presented.  This section, therefore, does not pre-
sent data but merely describes the approach to and specific plans for field
testing.

APPROACH

The basic approach to the field verification testing is to utilize a limited
number of integrated furnaces as the space heating sources in as wide a variety
as possible of existing oil-fueled residences.  Oil heating is concentrated in
three general regions of the U.S.:  the New England states, the Great Lakes
states, and the northern mid-Western states.  Because these are all far from
Rocketdyne's main plant in California, consideration of several logistics
aspects also was involved in selecting test locales.

Test Locale Selection
Three field test units are to be located in each of two test locales where oil
heat is used in a substantial fraction of single-family homes and which have
distinctly different winter climates.  As a start toward selecting the two
                                     103

-------
locales, winter climatic data averaged over several years were reviewed for
approximately 35 cities, most of which are distributed throughout the three
geographic regions mentioned above.  Data considered were:  annual and monthly
degree days; monthly average temperatures and precipitations; and average
diurnal temperature ranges.  All cities considered have normal accumulative
degree days, based upon 18 C (65 F), exceeding 4500 for the 9-month period of
September through May.  Monthly average temperature data were plotted directly
and on a weighted basis using U.S. census data on the number of households in
each city.  Mean curves were drawn through these two sets of data and were
found to differ by only  1C (2 F).  Comparing the individual cities' weighted
monthly averaged winter temperature profiles to the corresponding mean curve,
30 of the 35 cities had annual temperature profiles within ±5% of the mean.
Thus, it was inferred that the weighted mean temperature profile is represent-
ative for a majority of the households in the colder regions of the 48 contig-
uous United States.

Based upon their closeness to the weighted mean temperature profile, seven
cities were selected as being most representative; they were:   Boston, MA;
Cleveland and Columbus, OH; Indianapolis, IN; Providence, RI; Scranton, PA;
and Springfield, IL.  Comparisons of total winter precipitation and diurnal
temperature ranges revealed that these cities experience two broadly distinc-
tive winter weather patterns.  The maritime climate, represented by Boston
and Providence, encompasses the eastern seaboard.  It probably extends inland
for no more than a few dozen miles.  Because of its proximity to Lake Erie,
Cleveland's climate also falls in that climatic category.  The continental
climate, typical for the inland cities of Springfield, Indianapolis, Columbus
and Scranton, is characterized by wider diurnal temperature variations and
less total precipitation than are experienced in the maritime climate.  It
was concluded that one test locale should be selected within each of those
two general climatic categories.

Attention was then directed to nonclimatic factors, such as oil usage patterns,
predominant types of residential construction, state and local code require-
ments, and availability of local support.  Data concerning pertinent housing
characteristics for 11 candidate locates are listed in Table II.  Because of
                                    104

-------
the predominant use of oil for residential heating in New England, an assess-
ment was first made of metropolitan Boston as the test locale with a maritime
winter climate.  From several standpoints, this area was found to be very
attractive:  there are very broad ranges of house ages, architectural styles',
construction materials, degrees of insulation, and wind exposures; nearly all
of their central heating systems are oil fueled; the only code requirements
imposed on an experimental furnace is that the burner have a Massachusetts
approval number; there are reliable, well-established furnace equipment dis-
tributors willing to provide installation and service support for field test-
ing; there are several environmental testing laboratories in the area; and,
finally, there are regular and convenient travel and shipping connections to
Los Angeles, CA.  Therefore, Boston was tentatively selected as the maritime
climate test locale.  Emphasis then shifted to making subcontract arrangements
for local installation and service support and laboratory support in Boston
and, simultaneously, arriving at a tentative selection of a locale with a
continental climate.

It appeared from the initial Boston contacts that laboratory support might be
more difficult to arrange than either service support or host homes.  It was
considered that the larger metropolitan areas were most likely to have quali-
fied, well-instrumented laboratories, so the order of examining cities with
continental climates was established as from the larger to the smaller.  Thus,
even though oil has only a small fraction of residential heating there,
Pittsburgh got first consideration.  One or two apparently qualified labs
were identified and an interested service contractor was found who has a
substantial oil-heating background and several oil clients as potential host
homeowners.  Therefore, Pittsburgh was tentatively selected as the continental
climate test locale.
However, as the details of field testing in these specific cities were being
formulated and negotiated it was found that the competitive bid responses from
analytical laboratories were much higher priced than had been anticipated.
Subcontracting of the flue gas analysis effort was then deemed impractical,
and an option involving a mobile Rocketdyne instrumentation unit was formu-
lated.  It was apparent that with a single mobile unit to cover both test
                                     105

-------
locales, many advantages could be realized with a reduction of the distance
between the two locales.  A review of the climatic data revealed Albany, NY
as an excellent continental climate test site, separated from the Atlantic
maritime climate by the Allegheny mountain range, and less than half as far
from Boston, MA as Pittsburgh, PA.

The climate in Albany, NY during the heating season is somewhat colder than
that in any of the other 10 cities considered in the initial evaluation.  The
mean temperature of Albany is 5.1 C (41 F) during the heating season as opposed
to 6.5 C (44 F) for Boston.  The difference between these mean temperatures is
actually a little smaller than the difference in the numbers of degree-days
listed in Table II.  However, both criteria indicate that the thermal load
is substantially greater for Albany than for Boston.

Support in Test Locales
Service Support.  The selection of the local furnace service support started
as a general evaluation of any furnace service contractor in the selected
metropolitan areas.  However, the process of evaluation quickly revealed that
there are advantages to working with distributors of the same brand name
(Lennox) furnaces from which the integrated furnace was derived.  These serv-
ice contractors were more knowledgeable in the specifics of the construction,
installation, and operation of the stock predecessor unit.  The availability
of spare common parts through these service contractors reduces the likelihood
of unexpected delays.  Moreover, many of these franchised service contractors
were informed of some Lennox studies on fuel conservation devices, e.g., sealed
air systems, which relate to the integrated furnace unit.  The final selections
of service contractors were made by standard contract bidding procedures and
confirmed by visits of Rocketdyne personnel to the candidate organizations.
Selection of Host Sites.  A list of potential host dwellings was supplied by
each of the two service contractors from their existing customer lists.  The
selection of the six host sites (single-family dwellings) was based on crite-
ria that included consideration of:   (1) dwelling construction,  (2) dwelling
location, (3) existing dwelling furnace installation,  (4) fuel usage history,

                                     106

-------
and (5) host family.  The dwellings with basement installations are preferred
simply for the ease of accessibility and the space available for instrumenta-
tion packages.  Other dwelling construction factors (e.g., building materials,
age, number of levels, type of insulation, etc.) were considered to obtain a
variety of dwellings with varying thermal characteristics.  The thermal char-
acteristics of a dwelling can also be influenced by its location where there
could be significant localized differences in wind, insolation, heat retention
(ground or water), etc.  The integrated furnace configuration has been tested
                                    3                        3
at return air flowrates from 0.425 m /s (900 scfm) to 0.800 m /a (1700 scfm)
and return air temperature varying from 10 C (50 F) to 32 C (90 F) with little
or no effect upon its operation.   Therefore,  the existing furnace installations
in the candidate dwellings were evaluated primarily on the basis of geometric
compatibility with the integrated furnace and the field test instrumentation
requirements.
The fuel usage history considered was data for the two previous heating seasons
so that correlations with degree-day data could be made to normalize all the
thermal efficiency data to one baseline.  This correlation may also reveal the
necessity for an adjustment factor for changes in lifestyle (e.g., lowering
of thermostat settings) induced by recent energy conservation campaigns.  The
requirement for the host family themselves is stability of the number of occu-
pants during this and the previous two heating seasons.  If, for example, there
is a college student in say a small family of four, that will not be residing
in the house this season, the thermal load upon the heating system may be
affected significantly.  However, if the student had also been gone in the
previous two heating seaons, or if the dwelling is constantly occupied (e.g.,
kept heated) by another member of the house, the effect would be negligible.
The ages of the children were noted as host families without children from
1 to 8 years of age were preferred.  The concern here was the likelihood of
inquisitive tampering with the instrumentation (e.g., pulling the power cord)
by children in this age bracket.  Although it is not listed as a criterion,
a major influence upon the final selections were the attitudes of the host
families toward the field test.  Since the field test ia of long duration
(9 months) during adverse weather conditions and in two different cities,
it is very probable that delays will arise, and these could be compounded

                                     107

-------
greatly by an uncooperative host.  Therefore, enthusiasm and cooperation are
valued traits in the host families.

Integrated Furnace Installations.  The installations of the six integrated
furnaces are scheduled for the month of September (1977) and all are to be
operational by the first day of October.  Each installation will involve the
removal and storage of the existing furnace and the mounting of the integrated
furnace.  The warm air ducting will require minor modifications to allow the
temporary insertion of a 0.45 x 0.45 x 1.22 meters (18 x 18 x 48 inches) lami-
nar flow element section and various thermocouple probes for cyclic thermal
efficiency measurements.  A full complement of instruments to evaluate effi-
ciency will be installed on one of the furnaces while the remaining five fur-
naces will have time event recorders to monitor furnace operations and store
this information on magnetic tape (cassette).  The thermal efficiency measure-
ment apparatus will be rotated through the other five host sites during the
heating season.  The field test will terminate at the end of May 1978, at which
time the integrated furnaces will be removed and the dwellings restored to
their original configurations.

Test Measurements?

The field test evaluation will involve periodic measurement of flue gas emis-
sion concentrations and continuous event monitoring of cyclical operations.
The primary objective of the test will be to observe the long term pollutant
emissions characteristics of the integrated  furnace design.  A secondary objec-
tive is to establish the overall season-averaged thermal efficiency of the
furnace so that the pollutant emissions results can be evaluated in light of
the present concerns for energy conservation.

Air Pollutant Emissions.  The long-term monitoring of flue gas pollutant emis-
sion concentrations will be conducted with measurements at monthly intervals.
The pollutants to be measured are:  nitric oxide  (NO) and total oxides of
nitrogen  (NO ) by chemiluminescence; carbon  monoxide  (CO) by nondispersive
            A
infrared; unburned hydrocarbons  (UHC) by flame ionization and smoke by the
Bacharach method.  In addition, carbon dioxide (C0_) by NDIR and thermal

                                      108

-------
conductivity, and oxygen (0 ) by polarography will be measured to determine
                           *                         i
the operating conditions by stoichiometry.  All of the instruments, with the
exception of smoke pumps, will be installed into and operated from a 3/4-ton
capacity van mobile laboratory.  The van will transport the instruments to
each hose site and a 0.0064-meter (0.25-inch) diameter FEP Teflon line will
be connected to the flue at a point near the furnace to conduct the flue gases
to the analyzers.  The control of the furnace will then be locked into a 4-
minute-on/8-minute-off operating cycle by a repeat cycle timing device. Sample
transport pumps within the van will draw samples continuously and the gases
will be analyzed throughout the 4 minute firing cycle.   CO and UHC concentra-
tions characteristically "spike" on startup, and therefore, the concentration
profiles for these species will be recorded on continuous drive paper charts.
These pollutant emissions will then be averaged over the 4-minute firing and
compared from month to month in terms of their cycle-averaged values.  The
integrated furnace has consistently operated in the laboratory at SR = 1.20,
smoke 
-------
continuously monitoring, automatic data acquisition methods were devised to
rocord a large number of data points for a valid statistical analysis.
                                .»
A data logging system with a programmable central controller will be used on
one furnace to monitor all the temperature and flow parameters once every
second and data from each firing will be stored on magnetic tape.  This sys-
tem will result in cycle-averaged thermal efficiency data that is correlated
to cycle timing parameters and, eventually will be used to evaluate season-
average performance.  For this reason, cycle timing will be carefully moni-
tored on the remaining five furnaces by automatic time recording data loggers
and these data will also be stored on magnetic tape.  All the stored informa-
tion will be retrieved and evaluated monthly.  The data from the primary pro-
grammable data logger will be reduced on-site by the central controller while
the cycle timing data cassettes will be transported back to the Rocketdyne
facility for reduction on a timesharing computer system.

The steady-state thermal efficiency of the integrated furnace at 0.79 ml/s
(0.75 gph) firing rate is 85%, which is about the maximum achievable for non-
condensing flue gas furnace systems.  The cycle-average efficiency in actual
homes is expected to be approximately 5% lower at about 80%.  The season-
average efficiency for the integrated furnace will be determined by the field
test.  In addition, the fuel conservation effects of the sealed air system,
which does not appear in the above heat transfer evaluations, will be deter-
mined by comparison of the fuel  consumption of the two previous seasons and
that of the field test season.   Overall reductions in fuel consumption result-
ing from the installation of the integrated furnace are expected to be well
over 10%.
                                     no

-------
                                   CONCLUSION

Specific criteria for optimizing the designs of burners and combustion chambers
for residential oil-fueled space heating systems with respect to pollutant
emissions and thermal efficiency have been presented earlier (Ref. 1 through 4).
Developed for laboratory testing of commercial and research burners in a variety
of research combustion chambers, those criteria are believed to be applicable
to many types of residential heating equipment, e.g., either warm-air furnaces
or hydronic boilers with either side-fired or tunnel-fired combustors.  They
are sufficiently general that any manufacturer might, in principle, utilize
them to create his own embodiment of a low-emission, improved efficiency space
heating unit which is compatible with his own existing design concepts and
manufacturing techniques.  However, until the criteria's applicability and ad-
vantages have been demonstrated clearly, it is deemed unlikely that the industry
will give them serious consideration.

The integrated residential furnace and field verification activities described
herein are intended to serve as proof-of-concept demonstrators.  Within the
scope of this program, only one specific embodiment of the design criteria
could be carried through the entire process from design concept to field veri-
fication testing.  Therefore, the integrated furnace was intentionally made to
conform to the most prevalent heating system type:  a warm-air furnace with a
side-fired firebox.  It is anticipated that the field test results will docu-
ment substantial reductions in emissions and gains in performance for the in-
tegrated furnace and that these will encourage manufacturers to consider apply-
ing the underlying criteria to their products.  At that point, it will be im-
portant to remember that the integrated furnace is only one of many unit designs
which might be conceived as satisfying the several design criteria.   (In fact,
it is indicated in Ref. 1 and 2 that preferred configurations would embody
tunnel-fired and/or water-cooled combustors.)
Application of the low-emission design criteria to a different basic  type pre-
sumably would entail an R&D program with scope and magnitude at least compara-
ble with that described herein for the integrated furnace demonstrator.  This
would provide both confirmation that the general criteria are indeed  applicable

                                     111

-------
and an opportunity to respond empirically to unusual experimental results,
such as the downgrading of the integrated furnace's firing rate to eliminate
its excessively smoky operating condition.  That particular problem, which
arose from undetected errors in the smoke measurements sometime during the pro-
totype furnace testing period (i.e., after establishment of the general design
criteria), is perhaps an extreme example of empirical adjustment that might be
required.
                                     112

-------
                                  REFERENCES

1.  Dickerson, R. A. and A. S.  Okuda:   Design of an Optimum Distillate Oil
    Burnerfor Control of Pollutant Emissions,  EPA-650/2-74-047, Final Report
    on Contract 68-02-0017, Rockwell International/Rocketdyne Division, Canoga
    Park, CA, June 1974.
2.  Combs, L. P. and A. S. Okuda:   Residential Oil Furnace System Optimization-
    Phase I, EPA-600/2-76-038,  Phase Report on Contract 68-02-1819, Rocketdyne,
    Canoga Park, CA, February 1976.
3.  Combs, L. P. and A. S. Okuda:   Residential Oil Furnace System Optimization-
    Phase II, EPA-600/2-77-028, Phase Report on Contract 68-02-1819, Rocketdyne,
    Canoga Park, CA, January 1977.
4.  Combs, L. P. and A. S. Okuda:   "Design Criteria for Reducing Pollutant
    Emissions and Fuel Consumption by Residential Oil-Fueled Combustors,"
    Paper No. 76-WA/Fu-10, presented at the 97th ASME Winter Annual Meeting,
    New York, NY, December 1976.
5.  Peoples, G.:  "Sealed Oil Furnace Combustion System Reduces Fuel
    Consumption," Addendum to the Proceedings, Conference on Improving Effi-
    ciency in HVAC Equipment and Components for Residential and Small Commer-
    cial Buildings, Purdue University, Lafayette, Ind., October 1974, pp A66-
    A72.
6.  American  National Standard Performance Requirements for Oil-Powered
    Central Furnaces,  ANSI Z91.1 1972, American National Standards Institute,
    Inc., New York, NY, June 1972.
7.  Barrett, R. E., S. E. Miller,  and D. W. Locklin:  Field Investigation of
    Emissions from Combustion Equipment for Space Heating, EPA-R2-73-084a
    (also API Publ. 4180), Environmental Protection Agency, Research Triangle
    Park, NC, June 1973.
                                    113

-------
TABLE I,  COMPARISON OF POLLUTANT EMISSIONS FROM
          VARIOUS RESIDENTIAL FURNACES

STOICHIOMETRIC RATIO
CARBON MONOXIDE, g/kg
FUEL
UNBURNED HYDROCARBONS,
g/kg FUEL
SMOKE, BACHARACH NUMBER
NITRIC OXIDE, g/kg FUEL
TUNED
AVERAGES
FROM REF. 7
1.85
0.6
0.07
1.3
1.8
ORIGINAL
STOCK
FURNACE
1.50
0.27
0.015
0
2.2
PROTOTYPE
FURNACE
1.15
0.55
0.055
<1
0.63
 TABLE II.  PERTINENT HOUSING CHARACTERISTICS FOR
             ELEVEN CANDIDATE CITIES
CANDIDATE
CITY
BOSTON
PROVIDENCE
HARTFORD
CHICAGO
TOLEDO
CLEVELAND
INDIANAPOLIS
SCRANTON
COLUMBUS
PITTSBURGH
ALBANY
SEPT. THRU MAY
DEGREE-DAYS
<°FDAYS)
5586
S926
6314
6093
6105
6088
5561
6067
5681
5881
6818
THOUSANDS OF
HOUSING UN ITS
891
263
212
2,291
186
676
369
79
296
789
98
WITH WARM
AIR FURNACE
24%
82%
22%
45%
68%
71%
70%
11%
81%
69%
25%
HEATING USE
FUEL OIL
66%
66%
68%
13%
12%
6%
31%
46%
8%
6%
43%
WITH BASEMENT
94%
94%
26%
83%
62%
84%
53%
89%
77%
92%
90%







COMMENT
ATLANTIC
MARITIME
CLIMATE
GREAT LAKES
MARITIME
CLIMATE


CONTINENTAL
CLIMATE


                       114

-------
                                                          W
                                                          4J
                                                          c
                                                          0)

                                                          g

                                                          I-
                                                          o
                                                          u

                                                          -d
                                                          0)
                                                          •H
                                                          IH
                                                          •H
                                                          "O
                                                          O
                                                          S
                                                          a
                                                          n)
                                                          C
                                                          ^
                                                          0)
                                                          a
                                                          o
                                                          4J

                                                          O
                                                          M

                                                         Oi
                                                         .1
115

-------
1U

3 I
                                                                                                    6
                                                                                                    cu
                                                                                                    c
                                                                                                    o
                                                                                                   •H
                                                                                                    I
                                                                                                    0)
                                                                                                    o
                                                                                                    <*-!

                                                                                                     O
                                                                                                     O
                                                                                                    •H
                                                                                                     O
                                                                                                    crt
                                                                                                    (N


                                                                                                     01
                                                                                                     t-1


                                                                                                     00
                                                      116

-------
              3.0
           Q
           111
           E
           CD
              2.0
           I
           3
           I
           u
           CC
                    	SMOKE 2 NO. 1
                                                   SMOKE < MO. 1
                            ORIGINAL
                            STOCK FURNACE
                                 APPROXIMATELY 80 PERCENT OF EXISTING
                                 OIL FURNACES PRODUCE CYCLE AVERAGED.
                                 NOx EMISSIONS IN THIS RANGE f7|
                            - PROTOTYPE FURNACE WITH
                             RETENTION HEAD BURNER
                                  • PROTOTYPE LOW-EMISSION
                                   FURNACE
                                      - TARGET LEVEL
                 1.0
                                        1.5
                                                               2.0
                               STOICHIOMETRIC RATIO

             Figure  3.   Cycle-Averaged NO Emissions
z
ui
U
cc
Ul
a.

U
LU
<
z
I-
ui
UJ

-------
Figure 5.  Optimum Low-Emission Oil Burner
                     118

-------
Figure 6.  Cast-Formed Air-Cun.l.od  Combustion  Chamber
                       119

-------
        350
                         0.010            0.020

                         AIR FLOWRATE. STD. m3/j
0.030
  Figure 7.   Throttled  Exit Air Flow Characteristics
    •300
                                     © DESIGN FLOWRATE: 1.06 ™lrt Oil FLOW
                                                  • S.H. -1.15
                     0.010           0.020            0.030
                             AIR FLOWRATE. STD. m3/t
Figure 8.   Inlet  Throttling Air Flow Characteristics
                            120

-------
          1
                  1
                                      lit
                                      *
                                      K
1.0       1.1      1.2       1.3

         STOKHIOMETRIC RATIO
                    1.0      1.1       1.2      U

                            STOCHIOMETRIC RATIO
1.4
   Figure 9-   Integrated Furnace  Smoke  Emission Characteristics
    UJ
    O 26
      20
      15
    u» 10

    UJ
    ff
    

                                             Ei
                                          90
       15    14    13    12    11     10     9      8      7

                 VOLUME PERCENT CO2 (DRY BASIS) IN FLUE GAS

          I    .   I   .  I   .   I   .   r  .  I ' .  I  . I  . I .  I . I . I . f
          1.0
1.2         1.4      1.6     1.8   2.0  2.2

       STIOCHIOMETRIC RATIO
Figure 10.   Thermal Efficiencies of  the Integrated and  the Stock
              Furnaces
                                  121

-------

-------
     PERFORMANCE OF A THERMAL AEROSOL
                OIL BURNER
                   By:
J. E. Janssen, J. J.  Glatzel, and U. Bonne
          Honeywell Incorporated
          Minneapolis, MN  55413
                      123

-------

-------
                                   ABSTRACT
     Recent research has shown the detrimental effect of excess oil burner
capacity on seasonal efficiency of residential heating systems.  A thermally
generated aerosol offered the promise of reduced firing rate and superior
atomization wihout the need for extremely small burner nozzle orifices.  Thus
the objective of this study was to determine the relationship between fuel
temperature and pressure for a clean burning, efficient, small oil burner.

     The results showed that clean, efficient combustion can be achieved at a
firing rate of about 0.42 cm3/s (0.4 gph) with a thermal aerosol burner.
Lower pressure — (about 200 kPa g (29 psig)) and heated fuel (about 150°C)
reduced nozzle capacity 60 percent and improved atomization.  An experimental
burner is described in which heated fuel flashing in the nozzle produces a
mixture of vapor and fine droplets.  Provision was made for independent con-
trol of atomizing pressure, fuel temperature, and flowrate.

     Data are presented on the effects fuel  temperature, fuel pressure  and
excess air on emission of hydrocarbons,  carbon monoxide, soot and  oxides of
nitrogen for both No. 1 and No. 2 fuel oil.  Clean efficient combustion was
achieved with low excess air.  Increased excess air was needed in  the
research burner to reduce NOX.  Fuel temperature was  limited by  its  tendency
to precipitate carbon.  Regions of reduced fuel pressure with elevated  tem-
perature are defined which yield reduced firing rate,  efficient  combustion
and  low emissions.
                                      125

-------

-------
                            INTRODUCTION

     Recent studies (1,2^3) have shown that excess capacity in residential
heating systems leads to excessive stack losses and reduced seasonal  system
efficiency.  A study of 26 oil burners in the Boston area (4) revealed an
average excess capacity of 147% (i.e. 247% of design load).  The capacity
of these burners was then reduced an average of 27.1% by installing smaller
nozzles.  It was found, however, that excess air had to be increased an
average of 36.9% to prevent smoke.  The increased losses associated with
increased excess air largely cancelled the improvement to be expected from
reduced capacity.
     Many conventional oil burners are too large to heat a residence effi-
ciently.  Greater use of insulation compounds the oversize problem.  There
appears to be a real need for a burner with a capacity of 0.4 to 0.5 gph
                o
(0.42 to 0.53 cm /s).  Pressure atomizing nozzles of this size have such
small orifices and passages that they are prone to becoming plugged after
unacceptably short operating times.
     It is well known that improved atomization, i.e., smaller drops, per-
mits better mixing of the air and fuel and reduces the tendency to form
soot.   It would appear, therefore, that a burner that chould achieve good
atomization without resorting to very small orifices in the nozzle for a
flow rate of about 0.4 cm /s should give better efficiency in residential
heating applications.
                                     127

-------
     A droplet of fluid is held together by surface tension.   The surface
to volume ratio increases when a given volume of fluid is divided into a
larger number of smaller drops.  Energy input is needed to overcome the
surface tension force when the surface area is increased.  It can be shown
that the radius of a drop is given by (see Appendix)
                              „ _ "3
                              r "AEV
(6)
where:   o = surface tension
 i
         E = energy input
         V = fluid volume
     Equation (6) gives the average drop size consistent with energy consid-
eration.  Any spray, however, will have drops with a wide range of sizes.
Thus while it may not be possible to compute actual drop size from equation
(6), it does show the benefits of heated fuel.  Heating the fuel both reduces
the surface tension and increases the energy content.
     W. Tenney patented a "Fuel Aerosolization Apparatus" based on a thermal
principle (5).  In this device fuel oil under pressure is heated and then
allowed to expand through a nozzle.  The hot oil flashes to a vapor as the
pressure drops in passing through the nozzle.  Upon cooling, the vapor conden-
ses into very small (less than 1 micrometer) drops which produce a dense white
smoke.  The U.S. Navy used this device to produce smoke screens in World War
II.  This paper presents test results on a residential size oil burner using
this means of atomizing fuel oil.
     The objective of this study was to show that a thermal aerosol generator
                                                              3
could be used as an oil burner with a capacity of about 0.4 cm /s (0.39 gph)
and to define the operating parameters needed for clean, efficient combustion.
                                      128

-------
                       EXPERIMENTAL PROCEDURES
APPROACH
     The approach was to design a burner that would permit separate control of
each of the operating variables.  These were fuel  temperature,  fuel  pressure,
fuel flow rate and air flow rate.  Since fuel flow is a function of both fuel
pressure and temperature, some form of variable area nozzle was needed to per-
mit independent control of fuel flow rate.  Three  standard oil  burner nozzles
rated at 0.85, 1.0 and 1.25 gph were used.  The highest fuel pressures of in-
terest was used with the small nozzle and the lower pressures with the large
nozzle.  Thus it was possible to stay within the flow range of interest while
exploring temperature and pressure effects.
     Combustion efficiency was computed from measurement of the oxygen con-
tent and temperature of the flue gases.  The heat  exchanger coupled to the
combustion chamber was not designed for efficient  energy absorption but
rather to simulate the temperature quenching that  normally occurs.  There-
fore, efficiency calculations are based on an assumed flue gas temperature
rise of 300°C (540°F).
     The flue gas analysis was made with instruments discussed under the
section on instrumentation
Burner Design
     The burner consisted of a steel pipe 178 mm {7 in.) i.d. by 508 mm
(20 in.) long.  The first half was insulated with 13 mm (0.5 in.) thick mold-
able ceramic fiber insulation.*  The remainder of the chamber including the
*WRP Ceramic fiber insulation, Refractory Products Co., 12 W. Main st.
P.O. Box 309, Carpentersville, IL 60110.
                                     129

-------
Inlet end was insulated with 6 mm (0.25 in.) of the same material.   These
i                                                              3
dimensions gave a volumetric heat release rate of 1,928 kJ/s-m  (186,000
         3
BTU/hr-ft ) which was consistent with standard practice.
     The nozzle was screwed into a fitting attached to the fuel valve.   The
nozzle in turn sprayed fuel into a vortex mixing chamber attached to the end
of the combustion chamger.  This vortex chamber was 76 mm (3 in.) dia.  by
25 mm (1 in.) long.  A 38 mm (1.5 in.) orifice separated the vortex chamber
from the combustion chamber.
     Air was admitted tangentially to the vortex chamber through four 13 mm
(0.5 in.) steel tubes spaced 90° apart around the vortex mixing chamber.
Four additional air tubes were also installed parallel to the burner axis.
Use of these tubes provided air flow with no swirl.  The air was supplied
through a single orifice meter to a header.  The layout of the combustion
chamber is shown in Fig. 1.
Fuel Handling System
     Fuel was supplied from a weigh tank mounted in a pressurized chamber.
The weigh tank was mounted on a cantilevered beam equipped with strain
gauges.  The chamber was pressurized with compressed nitrogen which supplied
the atomizing pressure.  Fuel flow rate was determined by observing the
change in weight of the weigh tank over a measured time interval.  The weigh
                             3
tank had a capacity of 700 cm  which gave a running time of 20 to 30 minutes
before refueling was needed.
     The fuel was piped from the weigh tank chamber to the electric powered
heater located directly above the weigh tank.  This was a 1690 watt heater
mounted in a section of 1.5 in. pipe.  The  high powered heater was needed to
keep the heat flux density at the heater surface to the recommended level of
          2
9 watt/in. .  The fuel valve and burner nozzle were mounted directly on  top
of the fuel heater.  Layout of the fuel measuring and heating  system are
shown in Fig. 2.
                                     130

-------
Instrumentation
     The Instrumentation layout is shown in Fig.  3.   Instruments used
were as follows:
     Strain Gauge on Fuel  Weigh Tank - Baldwin  Lima  Hamilton  SR4
       Strain Indicator, Tyne N.
     Oxygen Measurement - Westinghouse, Hagan Oxygen Monitor.
     Nitrogen Oxides - Aero Chem Chemiluminescence Monitor
       for NO, NO .
                 A
     Hydrocarbons -  Beckman Model 400 Hydrocarbon Analyzer.
     Carbon Monoxide, Carbon Dioxide, Methane - Honeywell
       Non-dispersive Infrared Analyzer.
     Particle Size Measurements - Thermo Systems Inc. Model
       3030 Electrical Aerosol Size Analyzer for 0.01 to 1.0
       micrometer range and ROYCO Particle Analyzer for the
       0.5 to 10 micrometer range.
     Fuel Temperature - Copper vs Constantan thermocouple and
       Honeywell Class 19 Recorder.
     Fuel Pressure - Heise precision pressure gauge, 0-50 PSI range.
     Air Flow - 0.5 in. dia. orifice and water filled "U" tube mano-
       meter.
     Smoke Measurements - Bacharach Smoke Tester.
TEST PROCEDURE
     The test procedure was to fill the fuel weigh tank with No. 1 or No. 2
oil.  Properties of the fuel oil are presented in Table  I.
     The desired pressure on the fuel would be set.   The fuel valve was opened
slightly to bleed air from the system and fill  the fuel heater.  The valve was
then closed and power to the heater was adjusted to give the desired fuel tem-
perature in the heater.  Since the nozzle tended to pick up considerable heat
from the flame, it was necessary to compensate for this in adjusting the fuel
heater temperature.   When the desired temperature was reached, the air supply
                                    131

-------
was turned on, fuel  was turned on and ignited with a propane torch.   After
about 2 or 3 minutes the burner insulation was hot and the flame was stabilized.
The air flow was then adjusted to give the desired flue oxygen level.  The
various instruments were then read.   Following this the air flow or fuel pres-
sure was adjusted to a new value and a net set of readings was taken.  Usually
about four sets of readings were taken before it was necessary to refuel.
     The large number of permutations of the variables made it impractical to
test all possible combinations.  Instead operating parameters were selected
to cover a wide operating range and additional detail when results of special
interest were observed.  Tables II and III show the combinations of firing
conditions studied.   A brief test of the effect of swirl and a transient test
were carried out in addition to the steady state tests.
                                     132

-------
                         RESULTS AND DISCUSSION

      The experimental results are presented in Figures 4 through 17 and
Tables IV and V.
AEROSOL SIZE DISTRIBUTION
      Figure 4 shows how the aerosol drop size distribution is affected by
fuel temperature and pressure.  Figure 5 shows the relative volume distribu-
tion as a function of drop size.  Fuel temperature of 200°C and above pro-
duced substantial numbers of drops in the 0.1  micrometer range.  However the
fuel temperature had to be 250°C or higher to  produce a large volume concen-
tration.
      The data in Figure 5 cannot be used to estimate air/fuel ratio because
the measurements ignore droplets bigger than about 5 micrometers.  The larger
droplets tended to settle on the walls of the  sampling tube and sample chamber.
While the number of larger drops was small compared with smaller drops, the
mass of fluid in the larger drops probably was substantially greater.  Figure
5 is useful for comparing the relative volume  of fuel in the 0.03 to 3.0 micro-
meter range.
      As expected, increased pressure and lighter weight {NO. 1) fuel increased
the number of small drops.  When converted to  a volume or mass distribution the
distribution peaks at around 3 micrometers.  The apparent discontinuity at around
1.0 micrometers was at least partially due to  the use of two different measuring
instruments.  A Thermo-Systems Inc. model 3030 Electrical Aerosol Size Analyzer
was used in the range 0.01 to 1.0 micrometers.  This instrument based on a meas-
urement of the charged particle current carried by the collected particles after
they have been charged.  A ROYCO Particle Analyzer was used to measure the con-
centration of drops in a 0.5 to 1.0 micrometer range.  This instrument is based
on an optical principle.
                                     133

-------
      These instruments required a sample of the fuel  spray to be fed into
a.separate sampling chamber.   An aerosol  sample was then fed into the instru-
ment from this chamber.  Only the smaller air-borne drops would stay in sus-
pension through this sampling system.  Hence the larger drops in the fuel
spray were ignored.
FIRING RATE
      Figures 6 through 11 show the effect of fuel temperature and pressure
on flow for each of the three nozzles and each fuel.  Our objective was to
explore conditions in the flow range around 0.42 cm /s (0.4 gph).  To do this
the pressure range was lowered from 207 to 345 kPa (30-80 psi) with the .85
gph nozzle to 103-207 kPa (15-30 psi) with 1.25 gph nozzle.   Fuel pressure
was very stable and was controlled by the regulated nitrogen pressure in the
fuel chamber.
      Fuel flow measurements were made with the burner operating.  The nozzle
picked up considerable heat from the flame.  Consequently the nozzle tempera-
ture was substantially higher than the fuel heater temperature.  The nozzle
temperature was measured at the fitting into which the nozzle was screwed.
The nozzle orifice temperature probably was a few degrees higher.  Because
of these complications the nozzle temperature and therefore the fuel tempera-
ture varied somewhat as a function of flow rate.  We endeavored to compensate
for this but a 10 C uncertainty in the temperature of fuel entering the nozzle
probably occurred.  Since we were looking for the effect of a 100 C or higher
fuel temperature change, the uncertainty was not serious.
      The flow decreased with decreasing pressure as expected.  We expected
flow to increase slightly with increasing temperature due to lower viscosity.
When the boiling range was reached the flow was expected to decrease rapidly
due to vapor formation in the nozzle orifice.  However, in all cases except
that with NO. 1 oil in the 0.85 gph  nozzle, flow decreased with  increasing
temperature.  We believe this was due to the fact that fuel oil  boils over a
rather wide temperature range.  The  light ends probably produce  some small
bubbles in the nozzle orifice at fuel temperatures as low as 100 C.  The
restriction produced by bubble formation increases with increasing fuel tem-
perature.
                                    134

-------
      The one case where the 0.85 gph nozzle was operated at 205-215  C  and
276 kPa (40 psi) with No.  1  oil  (Fig. 6)  did show a substantial  flow  restric-
tion.  The No. 1 oil probably produced a  substantial  amount of vapor  under
these conditions.  No. 2 oil (Fig. 10) did not show any break, but probably
would have shown a break in  the  curve at  some higher temperature.
      In general we found we could operate reliably at 150°C and lower, but
cokeing become a problem during  shut down at higher fuel  temperatures.   Appar-
ently, when the fuel was shut off, heat from the combustion chamber raised
the nozzle temperature even  higher than its operating temperature.
      The 1.0 gph nozzle seemed  to operate with a characteristic different
from the other nozzles.  The curvature of the flow curve was opposite from
that of the other two nozzles.  This shows more clearly in Figure 9 where
the three nozzles are compared.   The curves have about the right relationship
in the 50-75 C range but the 1.0 gph nozzle curve drops much lower at tempera-
tures above 125°C.  Only one set of data  were taken for the 1.0 gph nozzle.
It is quire possible the nozzle  was partially restricted by carbon particles
generated in the fuel heater.
      Figure 10 shows the flow curve for the 0.85 gph nozzle with No. 2 oil.
More data were taken with this combination than the others.  The breakdown
point in the flow curve caused by boiling in the nozzle occurred at about
250°C.  However, at that temperature there were problems of nozzle fouling
due to cokeing.
EFFICIENCY
      Combustion efficiency was  calculated from flue oxygen concentration
and an assumed flue temperature rise.  The actual flue temperature rise in
any real furnace would be a  function of heat exchanger design.  For any given
set of combustion conditions a heat exchanger could be designed to achieve
the assumed flue temperature rise.  A flue temperature rise of 300°C (540°F)
is commonly found.
      Figure 12 presents a correlation among flue oxygen, flue carbon dioxide
and excess air for the two fuels studied.  A 0°C (32°F) dew point was used  in
preparing Figure 12 since we used an ice bath water trap to remove moisture
from the flue sample.
                                     135

-------
      Figure 13 gives the gross steady state efficiency (based on higher
heating value of fuel) as a function of flue temperature and oxygen content.
      Swirl in combustion chambers has been defined by different authors in
somewhat different ways (6^, 7j.  We have used a simplified form of Beer's
definition which is derived in the appendix.  The swirl parameter so defined
for this combustion chamber was 4.5.  This would be equivalent to a swirl
number of about 2.3 for the dimensions of this combustion.
      The swirl produced clean combustion under almost all firing conditions.
It was feasible, therefore, to assume a burner of this type could be operated at
2% flue oxygen (10% excess air).   This would give a steady state efficiency
of slightly over 81% with a 300°C flue temperature rise.
EMISSIONS
Nitrogen Oxides
      Figures 14 and 15 show the effect of operating parameters on nitrogen
oxide.  The nitrogen oxides were found to be essentially all in the form of
nitrous oxide, NO, but are reported as NO .  Both NO and NO  were measured.
                                         A                 A
The concentrations were always the same.
      As expected, NO  increased as the flue oxygen decreased due to the higher
                     /\
flame temperature associated with more nearly stiochiometric mixtures.  The
NO  level at a given flue oxygen decreased as the fuel temperature increased.
In most, but not all, cases the NO  also decreased with increasing atomization
                                  n
pressure.  Improved atomization generally reduced NO  levels.  This is consis-
                                                    A
tent with measurements and calculations {(>, 8_) which show that increased swirl
tends to reduce NO .  The explanation is that improved mixing reduces flame
volume and therefore residence time.  This  is, of course, offset by higher
flame temperatures.
      The fuel nitrogen was measured as shown in Table I.  If this is all con-
       to NO. the
            A
given in Table IV.
verted to NO , the contribution of fuel nitrogen to the NO  concentration is
            A                                             A
                                      136

-------
      Figures 14 and 15 show that at 8% flue oxygen the NO  all  can  be
                                                          A
attributed to fuel nitrogen.  However, at lower flue oxygen levels,  some
of the NO  must come from atmospheric nitrogen.
         A
Unburned Fuel
      Smoke, total hydrocarbons and carbon monoxide were undetectable for
all firing conditions except flue oxygen levels well below 1%.   Apparently,
the high degree of swirl  dominated other effects.
TRANSIENT PERFORMANCE
      Some smoke, hydrocarbons, carbon monoxide and even methane were usually
observed right after ignition.  It seemed that these disappeared more quickly
when hot fuel was used.  Although the instrumentation was not well  suited to
transient measurements, one transient experiment was conducted.   The 0.85 gph
nozzle was used with No.  2 oil at a pressure of 345 kPa (50 psi).  Two runs
were made; the first was with unheated oil and the second was with oil heated
to yield a 150°C nozzle temperature.  The results are presented in Table V.
      After 1.5 minutes of operation the unheated oil gave a smoke No. 5 whereas
the heated oil was already down to zero smoke.  The nozzle with unheated oil had
already risen to 80°C in 1.5 minutes whereas the nozzle temperature with heated
oil was at 121°C in 1.5 minutes.  Within 5 minutes both tests were down to 0-1
smoke.  The flue oxygen was about 3% for the unheated oil case whereas it was
about 5% with heated oil.  This difference in flue oxygen had no effect during
steady state runs since smoke was unmeasurable unless flue oxygen was reduced
well below 1%.  For some unexplained reason, smoke increased again after 15 to
20 minutes.  This phenomenon was not observed in any other tests.
      The NO  level was higher with unheated oil.  The increase was undoubtedly
            X
due to the lower flue oxygen.
      While this test was not conclusive, it supported the qualitative observa-
tion that heated fuel produced a shorter transient.  Better data would require
faster sampling, less instrument delay and contlnous recording.
                                     137

-------
EFFECT OF SWIRL
      The high swirl seemed to mask any effect of improved atomization with
hot fuel.  We therefore modified the burner to eliminate swirl.   The objective
was to try and answer the original  hypothesis that improved atomization achieved
through the use of hot fuel would reduce emissions and improve efficiency.  Four
additional air tubes were installed in the end of the mixing chamber to admit air
parallel to the fuel jet axis.  The tangential air jets were plugged.  The results
of a comparison using No. 1 oil and the 1 gph nozzle are presented in Table VI.
When the swirl was eliminated there was a correlation between smoke and fuel
temperature.  A nozzle fuel temperature of 114°C produced No. 9 smoke, 118 ppm
CO, 5 ppm CH., 31 ppm hydrocarbons and 11 ppm NO  even with a flue oxygen of
4.9%.  Increasing the fuel temperature to 179°C with the same flue oxygen low-
ered all emissions except NO  .  Increasing the fuel temperature to 199°C and
                            ^
lowering flue oxygen to 4.1%  increased the smoke slightly but lowered the other
emissions further.  All emissions except NO  disappeared when swirl was restored
                                                       f\
even though fuel temperature was varied from 121 to 178 C and flue oxygen was
varied from 8.1 to 2.1%.  The NO  levels were consistent with previous measure-
                                y\
ments.
      With swirl present the  flame was very compact and nonluminous.  Elimina-
ting swirl produced a much larger volume luminous flame.  The pictures in Figures
16 and 17 show this difference.
      The total pressure of the air supplied when operating at 0.42 cm /s (0.4
gph) and 2% flue oxygen was about 1.2  kPa  (5 in. water).  This undoubtedly could
be reduced by  increasing the  area of the air inlet passages and by reducing the i
swirl parameter to perhaps 2.  Swirl does, however, add to the pressure require-
ment of  the blower.
                                     138

-------
                               CONCLUSIONS
      The use of heated fuel oil to produce a thermal aerosol makes possible
the atomization of fuel oil at lower pressures and reduced firing rates.
Heating the fuel oil, both No. 1 and No. 2, to 150°C greatly increases the
number of droplets in the 0.1 micrometer range.
      The formation of bubbles as fuel flows through the swirl passages in the
nozzle restricts flow.  This permitted standard 0.85 gph to 1.25 gph hollow
                                              o
cone spray nozzles to be fired at only 0.42 cm /s (0.4 gph).  Atomization
achieved in this manner reduced soot, hydrocarbons and carbon monoxide.
      When combined with air swirl (swirl parameter as defined with appendix
equal to 4.5), non-luminous flames were achieved at flue oxygen levels of 2.0%
(10% excess air).  The Bacharach smoke was zero and there were no measurable
hydrocarbons or carbon monoxide.  Luminous flames were present without swirl
but the thermal aerosol was beneficial in reducing emissions.
      The thermal aerosol combined with swirl reduced NO  formation in most
                                                        /v
cases.  This effect tends to disappear when the flue oxygen is reduced below
2% (10% excess air).
      The use of heated fuel improves combustion during burner start up.  Fuel
heated to 150 C achieved zero smoke in less than 1.5 minutes.  Fuel at 80 C pro-
duced a No. 5 smoke at 1.5 minutes but achieved a zero smoke within 5 minutes.
                                                                   3
      The flow rate of 1.0 gph nozzle was varied from 0.2 to 0.6 cm /s by vary-
ing fuel temperature.  Clean combustion with low excess oxygen was achieved at all
firing rates.
      Thus the combination of the thermal aerosol generator with substantial air
swirl gives a non-luminous flame with low emissions at a flue oxygen level of 2%.
If the flue temperature rise is held to 300°C this results in 81% combustion
efficiency.
                                    139

-------
                             RECOMMENDATIONS

      The results of these experiments show promise for a residential oil
burner of 0.4 to 0.5 gph rated capacity.    A prototype burner of this size
should be built.  The burner should have a blower with sufficient pressure
rise to supply the vortex mixing chamber.  Provision would be made for re-
generatively heating the fuel.  A small electric powered heater should be
provided for initial heating of the fuel.  Excess air should be held in the
5 to 10% range.  Recirculation should be investigated as a means of further
reducing NO .
                                      140

-------
                                APPENDIX
Drop Size
     Surface tension is defined as the work required to displace a unit area
of fluid surface.
                                 »= WK/A                                (1)
For a spherical droplet,
                             WK = <^A = 
-------
     Air enters the vortex chamber tangential  to the axis and at a radial  dis-
tance, r..  The air is accelerated due to the conservation of angular momentum
as it goes through the mixing orifice which has a radius, r .  Thus the tangen-
tial  velocity at the orifice is:
y  =
                                    Q  r.
                                    _ L
                                    Airo
where:   Q = air flow rate
        A. = area of air inlet tubes
        r. = radial distance of tubes from axis
        r  = radius of orifice
The axial velocity, Va is
                     a
                                            (8)
                                                                        (9)
Then
V
  Qr.
                                        irr
                                          0
                                          Q
                                         (10)
For the four air inlet tubes
                               A- - 4,r-<
                                                                        (ID
     r, = 1.5 in.
     r  = .75 in.
     r1 = .25 in.
                                    4r-
                            -0-5)  (.75?
                                4(.25

                                           =4<5
                                          (12)
                                      142

-------
                               REFERENCES

1.   Bonne, U.  and A.  Johnson.   Thermal  Efficiency in  Non-Modulating  Com-
     bustion Systems.   Conference on Improving  Efficiency in  HVAC  Equip-
     ment and Components in Residential  and Small  Commercial  Buildings,
     Purdue University, October 1974.
2.   Bonne, U., J.E.  Janssen, R.H.  Torborg, and A. Johnson. Digital Simu-
     lation of the Seasonal Efficiency of Combustion Heating  Systems.
     Seminar on Efficient Fuel  U tilization in  Heating Systems,  ASHRAE,
     June 1975.
3.   Bonne, U ., J.E.  Janssen and R. H.  Torborg.  Efficiency  and Relative
     Operating Cost of Central  Combustion Heating  Systems, IV.   Oil-Fired
     Residential Systems.  Presented at ASHRAE  1977 Semi-Annual  Meeting,
     Chicago, Illinois, February 1977.
4.   Bonne, U., L. Katzman, and G.E. Kelly.  Effect of Reducing  Excess
     Firing Rate on the Seasonal Efficiency of  26  Boston  Oil-Fired Heating
     Systems.  Conference on Efficiency of HVAC Equipment and Components  II,
     Purdue University, Indiana, April 12-15, 1975, Proceedings, p. 81.
5.   Tenney, W. Fuel  Aerosolization Apparatus.   U.S.  Patent No.  4,013,396,
     22 March 1977.
6.   Baldwin, J.D.C.  and C.H. Long.  Effect of  Swirl  on NO  Emissions from
     a Gas-Fired Burner.  ASME paper 74-WA/FV-3, 1974.
7.   Lilley, D.G. Swirl Flow Modeling and Prediction For Combustor Applications,
     presented at CSS/CI 1977 Spring Technical.Meeting: Fluid Mechanics  of
     Combustion Processes, Cleveland, Ohio, March  28-30,  1977.
8.   Bonne, U. Source Emissions Control  - The Key  to Clean Air.  Honeywell
     Computer Journal, 7(3):199, 1973.
                                     143

-------
         TABLE I.   FUEL PROPERTIES
(Measured by Twin  City Testing Laboratory)
Property
API Gravity, 15.5°C (60°F)
Heating Value
(0/m3)
(BTU/gal)
Carbon (%)
Hydrogen (%}
C/H
Sulfur (%)
Nitrogen (%)
Nickel, ppm
Vanadium, ppm
Lead, ppm
BTU/lb
Atomic H/C
Value
No. 1 Oil
42.0

538,862
134,924
86.26
13 58
6.35
0.19
0.0064
26
23
28
—
—
No. 2 Oil
34.8

545,661
136,627
87.15
12.68
6.87
0.37
0.012
32
24
42
—
—
                       144

-------
TABLE II. FIRING CONDITIONS FOR NO. 1 FUEL OIL
Nozzle
(GPH)
0.85




1.00


1.25





Fuel
Tggp
20
20
20
150
150
20
20
20
20
20
20
150
150

Fuel
Pressure
kPa gage
207
276
345
207
276
138
207
276
103
138
207
103
138
207
Excess
Oxygen
%
2,4,8
2,4,8
<1, 2,4,8
2,4,8
2,4,8
2,4,8
2,4,8
4,8
2,8
2,8
2,8
2,8
2,8
2,8
                         145

-------
TABLE III.  FIRING CONDITIONS FOR NO.  2 FUEL OIL
Nozzle
(GPH)
0.85








1.00


1.25




Fuel
Tgnp
20
20
20
90
90
90
150
150
150
20
20
20
20
20
150
150
150
Fuel
Pressure
kPa gage
207
276
345
207
276
345
207
276
345
138
207
276
103
138
103
138
207
Excess
°2
r
2,4,6,8
<1 ,2,4,8
<1, 2,4,8
2,4,8
2,4,8
<1, 2,4,8
2,4,6,8
2,4,8
<1»2,8
2,4,8
2,4,8
<1, 2,4,8
2,4,8
8
2,4,8
2,4,8
<1, 2,4,8
                          146

-------
      TABLE IV.  CONTRIBUTION OF FUEL NITROGEN TO NO
       Fuel
       No.
  Fuel
Nitrogen
NO  Normalized
tox3% Flue 09
     PPM    *
       1

       2
.0064

.0120
     8.1

    15.5
TABLE V.  EFFECT OF FUEL TEMPERATURE ON TRANSIENT PERFORMANCE

             0.85 GPH NOZZLE, 345 kPa, No. 2 OIL
Time After
Ignition
Min.
0
1.5
5
10
15
20
0
1.5
5
10
15
20
23
Nozzle
35
80
88
100
100
100
71
121
150


150

Flue
Oy Smoke
% No.

5
0
3.1 0
6
3.0 5

0
5.0 1
4.7 0
4.5 0
3
4.5 1
PPM"



40

43


26
28

29
29
                               147

-------
TABLE VI.  EFFECT OF FUEL TEMPERATURE  AND FLUE  OXYGEN
          ON SMOKE - No.  1  OIL,  1  GPH  NOZZLE
Fuel
114
179
199
121
163
159
178
Without Swirl
Fuel
Press.
kPa
138
138
138
With Swirl:
138
138
138
138
Flue
Oxygen
4.9
4.8
4.1
8.1
5.1
5.4
2.1
Smoke
No.
9
3
4
0
1
0
0
CO
PPM
118
45
10
0
0
0
0
CH4
PPM
5
0
0
0
0
0
0
THC
PPM
31
5
3
0
0
0
0
X
11
23
19
18
25
29
63
                            148

-------
                                             0)
                                             C
                                             C
                                             0)
                                             M
                                             01
                                             Q.
                                             X
                                             Ed
                                             01
                                             M
                                             3
                                             60
149

-------
                                                  TO NOZZLE
             HEATERHOUSING
HEATER CARTRIDGE
     PRESSURE CONNECTION
                                                      POWER CORD
                                                         DELIVERY TUBE
                                                                   WEIGH BEAM
STRAIN GAUAGE  f=Z±
                                                                PRESSURE VESSEL
                    Figure  2.   Fuel handling system.
                                     150

-------
  UJ
       Q.
     ct:
     u
  QD

  yj
  o
cc
UJ CL
H <
< OC
J
        LU
        O
  cr
  £
                                                             s
                                                             02
                                                             XUJ
                                                             px
                                                             cno
                                                             UJ
                                                                         IT
                                                                         UJ
                                                                             *
                                                                            I
                                                             UJ
                                                                      O O  01
                                                                       O
                                                                       O
                                                                                      c
                                                                                      o
                                                                                      n)
                                                                                      u

                                                                                      a)
                                                                                      CO
                                                                                      C
                                                                                      en
                                                                                      QJ
                                                                                      Vi
                                               Z)
                                               O
                                               UJ
                                               ct:
                                        UJ-
                                        >
                                      151

-------
  10'
  10"
  10
o:
<
a.

u.  4
O 10

cc
UJ
m
UJ
a:
  10
   10
  10
     .01
                            0 260°C, 207 k Pa,No.I FUEL

                               Q85 GPH NOZZLE
           200°C,207kPa,

           No-1 FUEL

           0.85 GPH NOZZLE
               20°C,207k Pa, No.I

               FUEL, 0.85 GPH NOZZLE /
      l50°C.207kPa,

      No.I FUEL, 0-85 GPH NOZZLE
                                 300°C.I38kPo,
                                 No. 2 FUEL
                                 VARIABLE AREA

                                  NOZZLE

                                  0.23 cm3/S

                                    (0.22 GPH)
.0316
.1
.316
iO
             DROPLET  DIAMETER, MICROMETERS
                 Figure 4.  Spray size distribution.


                            152

-------
  10
  10
o:
UJ
uj  A
110
cr
o
o
o
UJ  9

i
UJ
cc
  10
   0
  10
                  260°0,207 kPa,No. I FUEL
                  0.85  6PH NOZZLE
        200°C, 207 kPa
        No. 1,0.85 GPH NOZZLE
                        207 kPo,
                        No I. FUEL
                                                  300*0,138 kPa
                                                  No. 2 VARIABLE
                                                   AREA NOZZLE
                                                   0.23 cm3/S
    .01
           -0316
.1
.316
1.0
3.16
10
              DROPLET DIAMETER, MICROMETERS
              Figure 5.  Spray mass size distribution.

                              153

-------
    0.6-
    0.5 -
        -0.5
1=
 o
UJ

-------
   0.6-
       -0.6
Cfl
u
 *»
LU
   05-
   0.4-
       - 0.5
       - O.4gph
   0.3-
  0.2-
             1.00 GPH NOZZLE, No. I  OIL
              o 138 kPa (20psi)
              •f 207 kPa (30psi)
              A276 kPo (40psi)
              X345 kPo (50psi)
    0.1-L
              25
          Figure 8.
                                                     15O
175
2OO
50      75      IOO      125
     NOZZLE  TEMPERATURE ,'C

Effect  of fuel temperature and pressure on flow.
    0.6
       -0.6
    0.5
       -0-5
o
ID
i-
K
o
                                                                 1.25 GPH NOZZLE
    0.4
    O3
    0.2
       • 0.4 gph
       - 0.3
       - O2
                                                             0.85 GPH NOZZLE
              NO. I FUEL OIL
              207kPo(30psi)
                                                                      .00 GPH
                                                                            NOZZLE
    0.1
              25       50     75      100      125      150
                               NOZZLE TEMPERATURE,°C
                                                              175
         200
          Figure  9.   Effect  of  nozzle size  and temperature on flow.
                                         155

-------
   0-8-
       -0.8

X
E
o

-------
a?
x
o
5

O
CD
cr
UJ
    15
    14
    13
II
    10
     8
                10% EXCESS AIR
                             . 2 FUEL OILtH/Cs 1.7467
       No. I FUEL OIL
                    «

                1.8898
                                       50
                                            60
                                               70
                                                  80
                                                     90
                                                       100
                        4       6

                      FLUE  OXYGEN,%
                                       8
IO
               Figure 12.  Correlation of flue parameters.

                             157

-------
o
z
UJ
o
UJ
                 FLUE  TEMP RISE
             NO. 2 FUEL  OIL
     68
                           4        6
                        FLUE  OXYGEN, %
                Figure 13.  Gross steady state efficiency.
                               158

-------
  E
  Q.
  0.
 o
 z
 g
    100
     80
     60
     40
     20
                                        0.85 GPH NOZZLE,NO.2  OIL
                                        NOX NORMALIZED TO 3% 02
        276 kPo,
                       207kPo,5Q''C
                          / 276 kPo,50"C

                      ^•*
                     2O7kPo
Figure
                              3456

                                FLUE  OXYGEN, %

                          Effect  of  operating parameters on
   100 h
    60
E
o.
a.

d*  60
z
tu
    40
    20
                                  1.25 GPH NOZZLE No.2 OIL

                                  NOX NORMALIZED TO 3%02
          •138 kPo.J780C

                  7 kPo. 172»C
                                    ^   103 kPo.65''C
                                                        103 kPo. I50*C
                                      138 kPo.J501C_

                                      138 kPo. IQO'C
               I       2       3       4       5       6       7
                                  FLUE  OXYGEN,%


             Figure 15.   Effect  of operating parameters  on NO

                                     159

-------
Figure 16.  Nonluminous flame with swirl.   Dark center spot is inlet
            orifice with mixing chamber in background.
                               160

-------
Figure 17.  Luminous flame without swirl.
                   161

-------

-------
             EFFECTS  OF  FUEL  AND  ATOMIZATION  ON  NOX  CONTROL  FOR
                  HEAVY  LIQUID  FUEL-FIRED  PACKAGE  BOILERS
                                    By:
                                 M.  P.  Heap
               Energy and Environmental Research Corporation
                            Santa Ana,  CA  92705
This paper was not received in time for publication, and therefore will be
included in Volume  V.
                                  163

-------

-------
               NOX  CONTROL TECHNIQUES FOR  PACKAGE  BOILERS:
                    COMPARISON OF  BURNER DESIGN, FUEL
                       MODIFICATION AND COMBUSTION
                               MODIFICATION
                                    By:

                             J.  E.  Cichanowicz
               Energy and Environmental  Research Corporation
                            Santa Ana,  CA  92705
This paper was not received in time for publication, and therefore will be
included in Volume  V.
                                   165

-------

-------
EVALUATION OF EMISSIONS AND CONTROL TECHNOLOGY
        FOR INDUSTRIAL STOKER BOILERS
                     By:

                R. D. Giammar
                   Battelle
            Columbus Laboratories
             Columbus, OH  43201
                    167

-------

-------
                                 ABSTRACT
          Battelle-Columbus is conducting a program to (1)  characterize
the spectrum of emissions from industrial coal-fired stoker boilers firing
several types of coal under various stoker firing conditions,  (2)  investi-
gate control methods to reduce these emissions,  (3) determine  the  effect
of these control methods and variations in stoker-boiler operation on the
overall performance of the stoker boiler, and (4) assess the environmental
impact of new technology on the future acceptability of stoker boilers.

          The program is divided into two phases.  In Phase I, the com-
prehensive emissions characteristics will be determined from the combustion
of untreated, reconstituted, and processed coals in Battelle's 20-hp
stoker-boiler small-scale stoker.  Appropriate data will be collected to
determine if criteria pollutants are responsive to changes  in  fuel, stoker
operation, or combustion parameters.  From the results of the  Phase I
research, an assessment will be developed that will include defining the
complete pollution potential for untreated, reconstituted,  and processed
coals for improving the environmental acceptability of stoker-boiler
technology.  This assessment will also relate the results of the small-
scale stoker-boiler investigation to larger scale stoker boilers.

          Phase II will be conducted to identify and evaluate  potential
control concepts for control of emissions from full-scale industrial
stoker boilers.  The Battelle steam plant 600-hp spreader-stoker boiler
will be utilized and modified to evaluate potential control concepts.  This
phase will provide a systematic evaluation of control concepts while
firing several representative stoker fuels.  Also, the most promising
treated coal identified in Phase I will be evaluated and compared  with
alternative control techniques.  Based on the results of the research
program, an assessment will be developed to define the potential of
emissions control on the overall acceptability of industrial stoker
coal-fired boilers to meet our nation's energy needs.  In addition,
recommendations for future applications of stoker modification control
will be made.
                                    169

-------

-------
             EVALUATION OF EMISSIONS  AND CONTROL TECHNOLOGY
                     FOR INDUSTRIAL STOKER BOILERS
                                   by
                            Robert D. Giammar

                              INTRODUCTION

          The firing of boilers for process steam,  space heating, and
on-site power generation accounts for about half of all the fuel used
by American industry—nearly as much as that consumed in the generation
of electricity by utilities.  Coal was once the dominant fuel for such
industrial boilers; it now provides only about one-quarter of their
fuel.
          Economic and enviroronental factors over the past 30 years
made coal relatively unattractive for industrial boilers.  Both the capi-
tal and operating costs of coal-fired boiler installations have been
inherently higher than those of equivalent facilities designed for gas
and/or oil firing.  Thus, with the widespread availability of oil and
gas in the late 1940's and early 1950's, there was  a drastic decline in
the demand for the industrial stoker.  Only in some of the largest boiler
installations could the savings resulting from the  somewhat lower fuel
cost of coal offset the greater capital and operating costs.  As a con-
sequence, there has been little incentive for manufacturers to improve
stoker technology through research and development.
          Stoker-fired boilers are now at a further disadvantage with
respect to environmental considerations, as their emissions of the
criteria pollutants (specifically particulate and SO™) are high in com-
parison to gas- and oil-fired boilers.  In addition, because of the
chemical composition of coal, other  potentially harmful  species  such as
                                   171

-------
POM and trace elements are emitted during stoker firing.   The establish-
ment of stringent air-pollution regulations in recent years (with national
standards for large new installations and local standards for many other
installations) has accentuated the economic disadvantages of stoker boilers.
Costs of installing and operating the downstream control equipment that
is needed to control emissions of fly ash and SO  for industrial boilers
are often a significant portion of the boiler facility and its operation.
          However, with the impending national shortage of natural gas
and fuel oil, industry must again consider the relatively abundant source
of coal to meet its energy needs.  For the commercial and small- and
intermediate-size boilers, the stoker has been demonstrated to be the most
effective method of direct firing of coal.  Although the economic aspects
involved with coal firing are becoming less dominant, some environmental
problems need to be resolved before stoker utilization can gain wide
acceptance.
          Accordingly, EPA has funded a research and development program
that will identify and demonstrate improvements in stoker coal firing
that can provide an incentive for greater industrial use of coal—a step
that enables the greatest flexibility in the nation's fuel resources.
The subsequent sections of this paper discuss this as well as pertinent
background information.  Because this program was only recently activated,
there are no data to report at this time.
                      OBJECTIVES AND SCOPE OF PROGRAM
OBJECTIVES
          The overall objectives of this program are:
             (1)  To characterize the spectrum of emissions from
                  industrial coal-fired stoker boilers firing
                  several types of natural and processed coals
             (2)  To investigate control methods to reduce these
                  emissions
             (3)  To determine the effect of these control methods
                  on the overall performance of the stoker boiler
             (4)  To assess the environmental impact of new technology
                  on the future acceptability of stoker boilers.
                                    172

-------
SCOPE
          The  scope  of this program can be summarized as follows:
               Coal Types
                • Untreated or raw
                •  Reconstituted
                •  Processed
             Pollutants of Interest
                •  Criteria pollutants  (SO , NO  , CO, HC, and particulate)
                                          L,    X
                •  Level  1 pollutants
                •   Selected Level 2 pollutants
             Experimental Units
                •  Existing 20-hp  stoker-boiler with  stack  sampling
                   facilities  (EPA  owned; located at  Battelle-Columbus)
                •  Battelle's  existing  600-hp  spreader  stoker boiler with
                   provisions  for  stack sampling
             Systems Operating Parameters
                •  Excess air  levels
                •  Boiler loads
                •  Overfire air rates
                •  Fly-ash reinjection  (600-hp boiler only)
            Modification of Stoker  Operation
                •  Overfire air modifications
                •  Steam  injection above bed
                •  Optimization of fly  ash reinjection  rates
                •  Optimization of primary/secondary  air flow
                   rates
                •  Other  modifications  evolved during perfor-
                   mance  of  program
            Modification of  Stoker  System
                 Candidate modifications include:
                 •   Zoned  air  admissions (to  control primary air
                    to different  parts of bed)
                 •  Modified coal  feeding at  low-load (as by hy-
                   brid  spreader-overfeed approach)
                 •   Small  oil burners fired above bed as after-
                   burners
                                     173

-------
                   •  Flue-gas recirculation
                   •  Staged-combustion
                   •  In-furnace control of SO- by limestone addition
                      to fuel bed
                   •  Other modifications as deemed appropriate.
                         TECHNICAL BACKGROUND

          A slow trend back to coal as a fuel for industrial boilers is
now evident.  Potential future disruptions in foreign oil supply, rapidly
rising oil and gas costs, political and economic decisions arising out of
balance of payment problems, and dislocations in domestric gas and fuel
oil supply appear certain to force this continuing trend back to coal.
          The concept of stoker firing has important attributes for com-
mercial and industrial boilers that have not yet been fulfilled by alter-
native methods of coal firing or by coal-derived fuels.  However, because
of the relatively small market for stokers in recent years, there has been
no incentives for manufacturers to improve stokers designs that have
remained unchanged in the past 20 years.  Therefore, to meet current
emission standards with stoker firing, expensive (or not fully proven
in the case of SO  control) flue-gas control equipment or premium coals
(low sulfur, low ash and in limited supply) will have to be utilized.
Modifications in stoker design and operation offer possible approaches
for improvement in the overall acceptability of stoker firing.  Alterna-
tively, consideration should be given to  the use of coals that have been
treated or  have undergone a change in their chemical properties  to make
them more environmentally acceptable.

TRENDS IN COAL FIRING

          Recent Battelle studies directed to the characterization of
the industrial boiler population provide  some insight as to trends in
coal-firing methods by boiler size range.  Information on the  industrial
boiler population and design  trends was developed in a special survey  of
                                     174

-------
                                   Size  Range,  pph
                                      10 -   16
                                      17 -  100
                                     101 -  250
                                     251 -  500
boiler manufacturers conducted jointly by Battelle and the American
Boiler Manufacturers Association (ABMA) and reported in Reference 1.
These results were later refined based on the basis of an analysis of
recent sales data for industrial-size water-tube boilers, plus overall
                                    O\
judgement of Battelle and ABMA staffv  .
          Firing methods are boiler-size dependent.  Thus, different  size
categories are defined as follows  for the presentation of trend  informa-
tion:
                  Size Category
                        A
                        B
                        C
                        D
Smaller boilers are placed in the "commercial" class and larger boilers
are considered to be "utility boilers".
          Figure 1 shows trends in coal-firing capability estimated for
boilers in these four size categories installed in 1930, 1950, and 1970
                       (2)
and forecasted for 1990   .  The data apply to boilers designed to fire
coal or having a capability to fire coal as a secondary fuel.  This dis-
tribution is shown as a percentage of all industrial boilers.  The pro-
jection for 1990 was revised by Battelle from the earlier survey on the
basis of the following broad assumptions:
          •  Oil and gas supplies will be limited
          •  Oil and gas will be utilized in smaller equipment
             and for high-priority uses  (but not fully by man-
             datory allocation)
          •  Coal will dominate new installations in the larger
             sizes
          •  Clean liquid and gaseous fuels from coal conversion
             processes will not be available in large quantities
             by 1990
          •  Firing refuse as a supplementary fuel will increase,
             but will not become significant in terms of percent-
             ages.
The coal projection includes the firing  of chemically refined solid fuels,
but does not include future synthetic liquid or gaseous  fuels derived  from
coal.
pph = pounds (of steam) per hour.
                                 175

-------
          The trends in Figure 1 indicate that coal was the dominant fuel
for industrial boilers in the pre-World War II era and that most boilers
had a coal-firing capability.  By 1950, percentage of units with a coal-
firing capability dropped dramatically for boilers being installed in the
smaller size categories; in the largest size categories, 65 percent of the
new boilers had coal firing capability.  By 1970, coal firing nearly dis-
appeared from the smaller boiler size categories, but 40 percent of the
largest boilers were being installed with a capability to fire coal.
          For the future, a return to the capability for direct firing of
coal is forecasted for new installations in the next 15 years, especially
in larger sizes, where about 70 percent of the new boilers are expected
to have coal firing capability.  In the event of new legislation and/or
fuel priority allocations that prohibit firing gas and oil in large boilers,
all new industrial boilers may be required to have a coal firing capability.
Some large industrial companies have already made the decision to provide
for a coal capability in new installations as a measure of insurance to
keep manufacturing plants operating in the face of an uncertain fuels
situation.

TRENDS IN STOKER FIRING

Industrial Coal Firing Methods
          Figure 2 shows trends in coal firing methods for industrial
boilers in each of the four size categories as developed in the studies
cited earlier.     Firing methods included in these data are
          •  Stoker Types
                    - spreader
                    - underfeed
                    - overfeed (chain grate, traveling grate,
                      or vibrating grate)
          •  Other
                    - pulverized coal
                    - miscellaneous  (cyclone, refuse,
                      sawdust, wood, etc.)
          The following overall observations can be made from Figure 2:
                                    176

-------
          •  Before the introduction and  commercialization  of  the
             spreader stoker  in the  1930's,  underfeed stokers
             dominated the smaller size categories  and had  an
             appreciable share of the market in the larger  size.
             The overfeed types were significant in all sizes,
             especially larger sizes.   Firing of pulverized coal
             was significant  only in the  largest sizes.

          •  By 1950, the spreader stoker had gained a signifi-
             cant share of the market below 250,000 pph, mainly
             displacing overfeed stokers  below 17,000 pph and
             underfeed stokers from  17,000 to 100,000 pph.   Pul-
             verized coal made some  inroads in the  size category
             from 101,000 to  250,000 pph  during this period.

          •  In the 1950's and 1960's, the penetration of spreader
             stokers into the share  of the underfeed stokers market
             continued to the point  that  it was the dominant type
             for the range of sizes  encompassing 17,000 to 250,000
             pph.  Pulverized-coal firing gained further, nearly
             dominating the market above  250,000 pph by 1970.

          •  For the near future, spreader stokers  can be expected
             to continue to gain popularity in the  smaller sizes
             with underfeed stokers  holding the major share below
             17,000 pph.  Spreader stokers can be expected to
             decline slightly in the market share above 100,000
             pph as smaller coal-fired units make gains.
It should be noted that introduction and  successful commercialization of

the fluidized-bed combustion concept to industrial  boilers could  begin
to displace the more conventional firing  methods during the late  1980's
and, thus, significantly alter the boiler population beyond 1990,

especially for units above 100,000 pph.
          The spreader stoker has gained  a significant share of the stoker
market for several reasons.  This type of stoker is a highly versatile

device that has greatly simplified the industrial utilization of coal by
being relatively insensitive to  coal properties.  Thus, the spreader
stoker can fire both caking and  non-caking coals and  coals with a broad

range of ash and moisture contents.   It is also the most flexible and
responsive stoker firing method  for fluctuating steam loads.  However,
the spreader stoker also has some limitations  that include  inherently

high fly ash carryover due to  suspension burning and  a  tendency to pro-
duce smoke at low boiler loads.
                                   177

-------
          Although the underfeed and overfeed stokers are have been declining in
popularity, there still will be a market for these types of stoker,
particularly in the smaller size categories.  In addition, many boilers
originally built with these type stokers but later converted to gas or
oil firing can be converted back to stoker firing.  Accordingly, any
assessment of the economic and environment feasibility of stokers should
include the overfeed and underfeed types.
          The design and operation of each of the three basic types of
stokers (overfeed, underfeed, and spreader) are significantly different
with the result that each has its specific operational characteristics.

Status of Stoker Technology

          Stoker development continued until the 1950's when the market
decreased drastically with the widespread availability of oil and gas.
Present stoker-boiler designs are essentially based on those of the
1950's.  Incentives for further technical development of stokers has
been nonexistent.  Until recently, the economic factors have clearly
favored oil and gas firing over stoker firing.
          Manufacturers of both stokers  and boilers believe that stoker
boilers have been developed  to a level that is technically competitive
with oil- and gas-fired boilers.  Criteria have been  established to
provide practical guidelines on combustion parameters that include
furnace volume, gas passages, baffle arrangements, and heat-transfer
surface area.  In addition,  the utilization of soot blowers, fly-ash
collectors, refractory arches, fly-carbon reinjection systems, overfire
air jets, and combustion controls to coordinate air and fuel supply with
changing loads has improved  the overall  efficiency and cleanliness of
the stoker boiler.  Selection of low volatile, low-ash, and low-sulfur
coals will also minimize the low-load smoke  (from spreader stokers),
fly-ash control, and SO- control problems associated with burning coal.
                                    178

-------
Economic Status—
          Coal has in recent years failed to compete with oil and gas for
industrial steam raising because of the higher first costs,  higher oper-
ating costs, and higher maintenance costs of the coal-fired  boiler units.
Thus, even when coal was the lowest cost fuel (in terms of cost per Btu),
many users chose the cleaner fluid fuels.  Many plants no longer have
that choice.  Gas is unavailable to many, and oil is very costly with
uncertain supply.  Coal can replace oil or gas for some larger industrial
users, if some measure of control can be attained over S0_ emissions.
          Figure 3 provides a comparison of steam-generation costs for
various fuels on the basis of a load factor of 0.30, a boiler efficiency
of 0.80, and an annualization rate of 16.7 percent.  These estimated
costs include both capital and operating expenses.   It is apparent from
the  example shown that  the stoker boiler is economically attractive when
the  cost differential between stoker coal and other  fuels is greater than
$1/10  Btu  (approximately $26/ton), steam-generation costs would be approx-
             3
imately $3/10  Ib of steam.  In an oil  or gas boiler,  the fuel costs
                                     ft                               T
would have  to be no higher than $2/10   Btu to generate steam at $3/10   Ib.
In the past, the cost of all industrial fuels was much less than $1/10
Btu  and cost differentials were about $0.50/10   Btu  between fuels.
          Figure 3 also indicates the steam costs from a stoker boiler
equipped with S0_ scrubbing equipment.   Because  of  these relatively high
estimated costs, it is  unlikely that SO scrubbing will be utilized with
medium-size industrial  boilers  (unless  dictated  by  government regulation).

EMISSIONS AS RELATED TO COAL COMPOSITION
AND STOKER OPERATION

          Table I summarizes the emission factors* for criteria pollutants
                                                         (1^
established by EPA as averages for fossil-fuel combustion   .  The higher
emission factors from coal combustion as compared to those from oil or gas
combustion can be attributed to the composition of the coal itself and/or
to the design and operation of the stoker.
 *
 which  a  pollutant  is  released  to  the  atmosphere  per  unit  of  fuel  consumed
                                    179
The emission factor as used here is a statistical average of the rate at

-------
Coal Composition
                                                                      (4)
          As noted in Table I,  particulate and  sulfur dioxide emissions
are significantly higher for coal combustion especially considering  that
coal can have an ash content as high as 25 percent  and a sulfur  content
as high as 4 percent.  Sulfur retention in coal ash is relatively low
compared to the conversion of sulfur to sulfur  dioxide.  Likewise, the
conversion of fuel nitrogen to NO can be significant as some coals con-
tain as much as 2 percent nitrogen.   In addition,  the combustion of  coal
emits trace elements at levels much greater than those from oil  or gas
because the concentrations of most trace elements  are greater in coal.
          Trace quantities of POM are generated by  nearly all combustion
processes.  Although combustion of natural gas  can  lead to the formation
of POM, POM formation is favored by the presence of the cyclic structure
of high molecular weight radicals and molecules in the fuel, which are
considerably more abundant in coal.      In addition, it is conjectured
that coal has certain other properties that enhance the potential for
POM  formation upon firing.  For example, data from a 20-hp boiler fired by
an underfeed stoker  (in  cyclic operation) suggest a relation between
POM  emissions and coal rank and volatile matter as evidenced by
                                     Volatile Matter,    POM Emissions,
                                         percent	   mg/lb coal burned
      Anthracite
      High volatile bituminous
      Low volatile bituminous
      Western subbituminous
      Lignite char
                                            4
                                           40
                                           21
                                           34
                                           16
 1
70
30
10
 2
                                                               (6)
 with POM generally increasing with volatile matter in the coal.     However,
 other factors may also be relevant, such as the free swelling index (meaure
 of the caking tendencies of a coal that affects the overall performance of
 some stokers).
                                    180

-------
Stoker Operation and Design

          As noted in Table I, the products associated with  incomplete
combustion (CO and HO are higher for coal than for oil  and  gas.  These
are primarily attributed to inhomogeneity  of the air/fuel ratio  over
the entire bed.   For some industrial underfeed stokers,  fuel beds are
extremely nonuniform, ranging in thickness from 8 inches to  over  2  feet.
Also, for caking coals, the fuel bed often contains large fissures
separating masses of coke.  As a consequence, there are  regions  in  the
bed that have an excess amount of air while others have  a deficiency of
air.
          In addition, because of the coal ash, the stoker has the  inherent
problem of fly ash carryover that can result from a disturbance  of  the bed.
In the case of the spreader, approximately 50 percent of the coal is burned
in suspension with a significant amount of fly ash carried out with the
flue gas.  In some spreader-stoker systems, to recover the carbon in this
fly ash, a portion of the fly ash is reinjected into the furnace  for re-
burning.  This results in a higher fine particulate loading, necessitating
the utilization of the highest efficiency particulate collectors.

EMISSIONS CONTROL TECHNIQUES FOR COAL FIRING

          Techniques have been developed to control emissions from  coal
firing.  Flue-gas control devices (electrostatic precipitators,  scrubbers,
baghouses, etc.) have been successfully utilized to control  particulate
emissions, but control devices aimed at SO  control (notably scrubbers^
have been substantially less successful.  This equipment is  expensive
and must be maintained.  Alternatively, the firing of coals  that  produce
few emissions (low sulfur-low volatile-low ash coals), either naturally
or by treatment offers another possible emissions control technique.
Likewise, modification in the design and operation of the stoker  to pro-
vide an improved control of the combustion processes offer another  potential
emissions control technique.
                                    181

-------
Fly Ash Collectors
          Commercially available particulate control equipment has been
utilized for many years and has been demonstrated effective in controlling
fly ash emissions from stoker boilers.  Nearly all stoker boilers sold
today have a particulate removal system of some type.   The simplest
systems are mechanical dust collectors (dry cyclones)  that are effec-
tive in removing only the largest fly ash particulates.  Depending
upon the ash content of the coal and the stoker design, these mechanical
collectors are often sufficient to meet local codes.  To meet the Federal
standards for particulates, the mechanical collector must be followed by
either a wet scrubber, electrostatic precipitator, or fabric filter.
Generally, the electrostatic precipitator has been found to be the most
cost-effective device and is most often used.  Designs have been
recently improved to produce efficiencies of over 99 percent.  The wet
scrubber might gain popularity for particulate control — if it can
also reduce SO^ emissions.  In summary, particulate removal equipment is
available with collection efficiencies approaching 100 percent.  The
equipment has been proven reliable and is available from many manufacturers.
Stack Gas Cleanup Systems for SO^ Control
          There are only a few industrial boiler installations utilizing
some form of SO,, cleanup.  In general, these systems have been expensive
and unreliable.  There are numerous processes under development, but only
a few are applicable to the industrial-size boiler.  Alkali scrubbing
systems offer the most promise.  The open-loop, once-through NaOH
scrubbing system has been demonstrated to be fairly reliable, but the
scrubber effluent can present a water pollution problem.  To avoid the
sodium-salt waste disposal problem, the scrubber effluent can be reacted
with limestone and/or lime to regenerate the NaOH, which is then returned
to the scrubber.  This closed-loop process is referred to as the dual-
or double-alkali system.  This system has not yet been demonstrated to
be fully reliable.  The system does show promise as an S0_-removal
process, but further development is needed to make it more reliable and
less expensive.
                                    182

-------
Coal Selection
          Of the criteria pollutants, particulate and SO- are the two
that can best be controlled by proper coal selection.
Particulate—
          Although it is not technically feasible to select a coal that
would eliminate the need of a control device to meet particulate emission
standards,* appropriate selection would minimize these emissions and also
the need for the highest efficiency collectors.  To meet air pollution
standards, a coal with essentially no ash would be required, however, a
minimum of about 4 percent ash is needed to protect stoker grates from
being overheated.  Utilization of low ash coals (4-10 percent) washed,
dust-treated, and double screened coal (not more than 20 percent below
1/4 size) would minimize fly ash carryover.  Likewise, selection of
free burning coals will enhance overall stoker performance and, thus,
minimize particulate loading.
so2-
          The Federal standard for SO- emissions is 1.2 lb/10  Btu of heat
input.  To meet this standard without control equipment requires  that
high-ranking coal with a sulfur content of less than 1 percent be utilized,
Several states have even more stringent requirements, regardless of the
size of the combustion equipment, and because of this fact, many coal-
fired boilers were being converted to gas or oil firing until the recent
fuel shortages.
          Any technique that improves the overall performance of the
stoker will minimize smoke, particulate, CO, and gaseous hydrocarbons
emissions although it may increase NO  emissions.  Some of these tech-
                                     X
niques include
          •  Overfire air jets—to improve mixing above the
             fuel bed
                       /o\
*  The Federal standard    for particulate emissions from large boilers
   is 0.1 lb/106 Btu-hr of input while state standards^  range from
   0.1 to 0.8 lb/10° Btu-hr, largely depending upon the heat-input r
rating.
                                    183

-------
             Zoned-air admission—to provide a control of the
             air/fuel ratio over the entire grate and produce
             uniform burning of the entire fuel bed
             Reduction in excess air level—reducing air velocity
             through the bed minimizes bed disturbance and fly
             ash carryover.  Also, this technique increases boiler
             efficiency.
             Although not utilized in industrial stoker applications,
             afterburners, either oil or gas-fired, used just above
             the fuel bed can reduce emissions associated with incom-
             plete combustions.
                            RESEARCH PROGRAM

PHASE I.  EVALUATION OF ALTERNATIVE COALS
          The prime objective of this phase of the research program is to
evaluate the emissions resulting from the combustion of untreated,  recon-
stituted, and processed coal in a small-scale stoker.  Two types of stokers
will be used: (1) a commercially available underfeed stoker and (2) a
model spreader stoker.
          The evaluation of emissions from the combustion of candidate
stoker coals will be strongly related to design and operation of the
stoker.  Certain chemical and physical properties of the coal are more
critical for one type of stoker than for another.  For example, the
spreader stoker can fire most caking coals successfully while the under-
feed stoker cannot.  Likewise, a very low-volatile coal like anthracite
cannot be fired in spreaders (unless mixed with another type of fuel) but
can be fired in either an under- or over-feed stoker.  Accordingly, the
basis for the coal evaluations will be expanded by the use of two stokers.
          The spreader stoker was selected because this type will be used
during Phase II and its use in Phase I will provide data for Phase II
planning.  As previously discussed, the spreader has been demonstrated as
one of the most effective methods of direct firing of coal in intermediate
size equipment.  The underfeed stoker was selected because (1) it is
                                   184

-------
representative of smaller industrial stokers,  (2)  it  is part of an existing
stoker-boiler facility that can be operated immediately, and (3) signi-
ficant data already exists for selected coals.

Candidate Coals
          Coals to be considered are untreated,  reconstituted, and processed.
The untreated coals will be representative stoker  fuels and/or have  low
pollution potential based on natural analysis.   These coals are commer-
cially available.  The reconstituted coals (physically treated) and  the
processed coals (chemically treated) are coals that have been  treated to
obtain the desired properties of a low emission fuel.  These coals will
be prepared at Battelle-Columbus or at an ongoing  processing facility.
The reconstituted and the processed coals will require some preparation
for stoker firing (either briqueting or pelletization) as  these coals are
produced in powdered form.
Untreated Coals—
          From a pollution standpoint, properties  of  coals most suitable
for stoker firing include                 	
          •  Ash content - about 5 percent
          *  Sulfur content - less than 1 percent
          •  Nitrogen content - less than 1 percent
          •  Volatile content - less than 25 percent
          •  Moisture content - less than 10 percent
          •  Fusion point  of ash - greater than 2500 F
          •  Free swelling index - less than 2.
 There are few coals that possess  all of these properties.  In addition,
 some of these properties are not  critical to  operation of certain types
 of stokers.
          For example,  the overfeed stoker,  typified  by the chain, traveling,
 vibrating,  or pulsating grates, produces the  minimum particulate emissions
 because of the relative quiescence of the fuel  bed.  However, these  stokers are
 usually limited to noncaking or very weakly caking coals; those having  a low  Free
                                   185

-------
Swelling Index, in the range of 2 to 4.   More strongly caking coals  (such
as those of Ohio, West Virginia, and Pennsylvania coals)  produce  caked
masses on the moving grate and poor distribution and flow of primary  air
results.  Coals from Indiana, Illinois,  and farther west  are generally
weakly caking and burn uniformly, but have sulfur levels  as much  as  3 to
4 percent.
          The spreader stoker projects the coal particles into the high
temperature furnace so that the hot flame gases destroy the surface-
caking tendency.  Thus, as the heavier burning particles  fall to  the  grate
they form a porous, uncaked bed through which primary air can flow uniformly.
Thus, the spreader stoker is the most versatile of all stokers; it can burn
a broad range of coals as well as many combustible industrial wastes, such
as bark and bagasse.  Also, with a continuously moving fuel bed the  spreader
stoker can generally handle much higher ash and moisture  content  than the
underfeed stoker (which also is adapted to burning caking coals). Two
drawbacks of the spreader stoker are inherently high particulate  emission
and a tendency to smoke below about 35 to 40 percent of full load.
          However, for the highly versatile spreader stoker, all  coals are
"typcial".  Low ash content is preferable to minimize loss of fly ash.
          For the underfeed stoker, the eastern or mid-western caking or
milding caking coals are "typical".  Low ash is also preferable to minimize
the labor of clinker removal.
          A few low-sulfur, strongly caking coals are available in the East,
but most of these are captive for metallurgical use.  In general, small users
seeking low sulfur coals must look to western coals.

Reconstituted  Coals—
          Little work has been  done  in  recent years on briquetlng or
pelletizing coal,  screening  operations  at  the mine being adequate to
provide  coal of  the proper size for  industrial users.  In earlier times,
and particularly in Europe, briqueting was common, mainly for producing
domestic fuels.  In Germany,  for example,  low-rank coals were
briqueted without  the  addition  of binder by using ring-roll presses  with
                                   186

-------
sufficiently high compression forces  that  the coal  particles adhered
tightly to one another.   In the United  States during  the  1940's, a form
of briquet called "packaged fuel",  consisting of  4-in.  cubes, was prepared
on a commercial scale to utilize No.  4  buckwheat  anthracite by blending
it with bituminous coal  for hand-firing in domestic furnaces.  A small
amount of starch binder  was used to stabilize the cubes until they were
fired, when the caking characteristics  of  the bituminous-coal portion
helped agglomerate the anthracite fines into a  relatively free-burning fuel
bed.  At the end of the  war, with a return to a normal  fuel supply situation,
production of packaged fuel was stopped.   Economically, briquetting could
not compete with sized raw coal in a  normal economy.
          In Korea, the  major occurrence of coal  is a friable type of anth-
racite which is converted into a useful form by briquetting it,with clay
as the binder,into cylinders roughly six inches in  diameter and  six inches
high.  These cylinders are pierced longitudinally with half-inch holes,
leading to the descriptive name of "nineteen-hole briquet".   In  1966, more
than 7 million tons of such briquettes  were burned  in Korea, mainly for
domestic heating and for cooking.  The  clay binder  is added to maintain
integrity of the briquettes both during handling and  during burning.  The
briquettes burn cleanly, and essentially all the combustible  is  utilized
over a four- to six-hour period.
          Pelletizing to dispose of coal fines  has  been tried  in the United
Stated    to convert 28-mesh x 0 filter cake or slurry from coal preparation
plants into a product with combustion characteristics essentially the  same
as the original coal with the same size consist.  A wet binder,  typically
a lignin derivative from the paper industry, amounting to about  one percent
of the weight of the coal, produced adequately  strong pellets.   Estimated
operating costs in 1957 were $1.08 per  ton of coal  pellets, an  unrealistic
figure today.  It is evident, based on these experiences, that  reactive
additives are not necessary to pelletize or to  briquet most coals.  There
is the additional factor today of deliberately  adding a reactive component
to coal briquets or pellets to capture sulfur during combustion or to
                                   187

-------
lower the resistivity of the resulting fly ash.   The most attractive
sulfur-capturing agent is limestone which could  be easily incorporated
as a 200-mesh slurry into briquets or pellets.   To maximize compression
strength of the product, Ca(OH)   would be better but also much more
costly.  A major problem here is that for reasonable effectiveness, the
fuel-bed temperature must not exceed about 1800  F because CaSO,  decomposes
into CaO and S0_/S0, when the temperature is higher.  Hence,  there would
be no advantage in having lime present if normal fuel-bed temperatures
were reached.  Another problem is that lime is an effective flux for the
ash in coal, reducing the fusion temperature markedly and, thus, leading
to increased clinker formation.
          Addition of Na CO  to coal has been shown to catalyze gasification
reactions, and it is evident that the alkaline can improve combustibility.
Also Na^CO. lowers the electrical resistivity of fly ash thereby making
fly ash easier to collect in an electrostatic precipitator.  But the alkalies
are also potent fluxes for coal ash and they are the major factor in causing
fouling of heat-receiving surfaces, such as wall tubes, superheater, and
reheaters.  In large boiler furnaces, as for the public utilities, increase
in the alkali content of coal would be objectionable; it is not known whether
this same objection would apply to spreader or to underfeed stokers in in-
dustrial plants.
          No fuel treatment systems are known at present that apply coatings
to fine sizes of coal.  Surface treatment of screened coal has been used
in the past to minimize dust and to prevent freezing in transit, but these
have been typically light mineral oils and calcium chloride.   Alkaline earth
and alkali metal compounds might be applied to coal fines with the reserva-
tions noted earlier about increased clinker formation and fouling tendency,
particularly if appreciable amounts of such treatment were applied.  Small
quantities of treatment definitely can control a dust nuisance during
handling, but the limits are not closely defined for today's technology
and pr ob1ems.
                                   188

-------
          The characteristics of coal ash  vary so widely  that  it  is difficult
to generalize about such factors as acid  leaching.  Low-ash-fusion coals
often contain appreciable quantities of pyrites, FeS  , which may  remain
only partially oxidized in clinker from industrial  boiler furnaces.  Such
clinkers, when exposed to weathering, may  release sulfuric acid by further
oxidation of the pyrites, just as acid mine water if  formed.   Also, SO
in flue gas can condense on fly ash, leading  to acid  leaching.  There is
at least a fair likelihood that chemical treatments can be derived to
neutralize clinker and fly ash to eliminate acid leaching, but the most
effective route would seem to be by treating  the products of combustion
rather than the coal.  Applying sealing coats to the  surface of the clinker
or fly ash would appear to be less desirable  than chemical neutralization,
but this needs to be evaluated.

Processed Coals

          Processed coal is defined as a coal or solid fuel derived from
coal which has been chemically cleaned for the purpose of removing at least
a part of the pollutant forming constituents  such as  sulfur, nitrogen, and
ash values including the toxic metal values.   During  the  treatment, the coal
structure may also be modified or altered  which may improve the combustion
characteristics of the coal.
Solvent Refined Coal—
          An example of a chemically cleaned coal  is  solvent  refined
coal.  In this case, a significant portion of the  sulfur  and  ash
including the toxic metals are separated  from the  coal by selective
solubilization of the coal in an organic  solvent.   After  separation of  the
solubilized coal from the residual ash and a sulfur,  a solid  fuel referred
to as solvent refined coal is obtained by evaporation of  the  solvent.
This solvent refined coal with a melting  point of  several hundred degrees
Farenheit, represents a solid fuel in  which the coal structure has been
modified or altered since the coal was actually solubilized during the
chemical cleaning operation.
                                   189

-------
          Solvent-refined coal (SRC)  offers a  unique opportunity  for  con-
trolling sulfur and particulate emission,  but  its  handling and  combustion
characteristics are still not well enough  known to permit  prediction  of  its
suitability in industrial boiler furnaces.  Preliminary work done for the
Electric Power Research Institute by  Babcock & Wilcox     demonstrated that
the ignition of pulverized SRC should be similar to a high volatile bituminous
coal while the burnout temperature is similar  to a low volatile bituminous
coal or an easily burned anthracite.   Hence combustion in  pulverized  form
should pose no serious problems.  Because  SRC  tends to become tacky at
temperatures as low as 280 F, special precautions  are necessary in pul-
verizing it.  Emission of NO  has been high because SRC has a high fuel-
bound nitrogen content, but SO  emissions  are  low  since SRC contains  much
                              i    *
less sulfur than the original coal.  Emission  of particulates also is low
because the ash content is generally  well  under one percent.  However, about
4 to 5 percent ash is desired because the  coal ash protects the stoker
grate from overheating.
          Probably, because of its low melting point and  low-ash  content,
SRC will not be adaptable to stokers, either spreader or underfeed.   However,
at 410 F, SRC has a sufficiently low  viscosity for pumping like residual
fuel oil, and at 540 to 580 F it can  be atomized successfully.  Thus, since
its combustibility apparently is equivalent to that of pulverized coal,  and
since it can be handled much like a heavy  residual fuel oil,  there is an
excellent chance that SRC can be adapted to industrial boiler  furnaces as
a liquid fuel.
   Apparently SRC does not remove any fuel nitrogen.
                                    190

-------
TRW Coal Meyer's Process—
          Another example of a processed coal is that produced  by  the
TRW process.  In this process, the sulfur and possibly some  of  the ash  is
selectively extracted and separated from the coal matrix in  contrast to
solvent refining in which the coal is selectively extracted  from the
ash.  The resulting chemically cleaned coal, i.e., that produced by the
TRW process, may behave quite differently in stoker firing because its
structure has not been altered to the same extent as the solvent refined
coal.  However, because of the removal of a portion of the pollutant
forming constituents, the TRW coal should have a lower emission factor
than that of raw (untreated coals) and may also be more reactive.

Battelle-Columbus — Hydrothermal Treated Coal (HTT)—
          The Battelle-Columbus'  hydrothennal-treated coal process includes
heating an aqueous slurry of coal and chemical leachant at moderate tern-4
peratures and pressures to extract sulfur and some of the ash from the
coal.  Further applicability of this process may be achieved since in
one alternative of the process the solid fuel is Impregnated with  lime,
a sulfur-getting chemical in the desulfurization operation.   During the
combustion process, the alkali — part of the coal structure — reacts  with
part of the remaining sulfur in the clean solid fuel to produce nonvolatile
sulfur compounds which remain with the fuel.  Another feature of the HTT
process is the extraction of a number of potential toxic or  hazardous
metal values from the coal, many of which might normally be  discharged
into the atmosphere with the products of combustion.

Briqueting/Pelletizing—
          Stoker fuel may be prepared from solvent refined coal by one
of several methods.  The simplest method would entail briqueting in
which the coal is compacted into briquets of the desired size.   Because
of the nature of the coal — low melting point — the addition of  a
binder will most likely not be necessary.  Grinding or milling prior  to
briqueting may be required depending on the size of the particles  or
                                   191

-------
pieces from the solvent removal operation.   Stoker  fuel may  also be
prepared by pelletizatlon either during or  after  removal  of  the solvent.
Pelletization may be achieved during  removal of solvent by conducting
the .solvent removal operation in a rotating drum  arrangement.  In  this
case, the addition of a binder would,  most  likely,  not be required.
Furthermore, grinding or milling of the fuel would  not be required as
the pellets would be formed during the solvent removal step.
          The solvent refined coal may also be pelletized after removal
of the solvent.  In this case, the coal would be  ground to a desired particle
size, probably about 100 percent minus 100  mesh.  The ground coal  would  then
be tumbled in a rotating drum to form the pellets of the  desired size.   The
addition of a binder will not be necessary  if the coal  is heated slightly.
On the other hand, a binder such as water or an organic solvent may be
mixed with the coal prior to pelletization.
          Stoker feed may be prepared from  TRW coal and the  HTT coal by
briquetting and/or pelletizing.  In both cases, the addition of a  binder
may be necessary.  Briquettes of sufficient mechanical  strength may be
prepared by compacting a mixture of the chemically  cleaned coal and a binder
such as water into the desired size.   Other candidate binders include sugar
solutions and coal tar.  Pellets of the TRW and HTT coal  chemically cleaned
coal may be prepared by the same method as  used  in  pelletization of the
solvent refined coal.  In this case, the chemically cleaned  coal before
drying may be tumbled in a heated rotating  drum to  form the  pellets and
evaporate the water.  If stronger pellets are desired,  the coal may be
heated to a temperature just below or near  the softening  point of  the coal,
approximately 500 to 600 F.  On the other hand,  strong  pellets may be formed
by the use of sugar solutions or coal tar as the  binder.
                                   192

-------
Experimental Systems
          A commercial 20-hp hot-water stoker-fired boiler capable of
firing coal at rates up to 75 Ib/hr will be utilized during Phase I.
Currently, the boiler is fired by an underfeed stoker.  This facility
has suitable controls so that the stoker-boiler can be operated
unattended under conditions typical of commercial stoker operation.
Combustion gases are vented through a 14-in. insulated stack that
includes provision for sampling (4 ports) at a location 10 stack diam-
eters downstream of the boiler outlet.  The stack has additional ports
for gaseous emissions and smoke determination.  Bottom ash residues
can be collected through the firebox door.  If this procedure is found
not to be satisfactory, a minor modification in the boiler setting
will be made so that ash samples can be collected without disrupting
the operation of the stoker.
          It is considered difficult  to project the effect of some  important
coal properties on combustion in a spreader stoker, based on underfeed stoker
data.  The two types of stokers provide radically different burning methods
 (suspension plus bed burning versus entirely bed burning) and even have
opposite  coal and air  flow patterns (counterflow versus parallel flow).
Thus, as  discussed earlier, to make the results of Phase I more relevant to
Phase II  and more applicable to industrial  stokers, this small stoker will
be modified to operate as a spreader  stoker for most  of the Phase I studies.
A  feed system similar  to that of a spreader stoker will be installed through
the  firebox door and the boiler setting will be modified to accommodate a
new  grate.  Other elements of the  facility will remain the same.

Environmental Impact
          From the results of the Phase I program, an assessment will
be developed that will include defining the complete pollution potential
for untreated,  reconstituted,  and processed coals for improving the
environmental acceptability of stoker-boiler technology.   This assessment
will also relate the results of the small-scale stoker-boiler investi-
gation to larger scale stoker boilers.
                                   193

-------
PHASE II.  CONTROL TECHNOLOGY EVALUATION PROGRAM
          The prime objective of this phase of the research program is to
identify and evaluate potential concepts for control of emissions for full-
scale industrial stoker boilers.  In this phase of the program,  control
strategies are limited to modifications in the design and operation of
the stoker and to firing of low pollution coals (either naturally occurring
or those that have been chemically or physically treated).  Research on
flue gas clean-up equipment will not be investigated.
          The experimental research will be conducted in the Battelle-
Coluiribus steam plant 600-hp spreader stoker boiler facility.  In addition,
Battelle-Columbus plans to consider the evaluation of potential  control
techniques in the small scale system, if deemed appropriate, prior to
evaluation in the larger stoker boiler.  If technically feasible or
practical, utilizing the smaller system to screen potential control
concepts can be accomplished at a significantly lower rate of expenditure
than in the larger system.

 Experimental Boiler  and Control Equipment

          The investigation and demonstration will be conducted in a water-
tube boiler  (approximately 600 hp) having a 25,000 Ib/hr steam capacity
that has been recently converted to stoker firing.  This boiler, a Keeler
Type MKB, was designed for coal firing but until recently has been fired
on oil and gas.  The stoker is a Hoffman spreader-type with a vibrating
grate.  The  facility is designed with a fly-ash reinjection system to
promote carbon burnout and, thus, to increase boiler efficiency.  In
addition, the installation includes a UOP Multiclone mechanical dust
collector for particulate control and complete coal-handling and ash-
disposal systems.
          The stoker installation was designed to include provision for
a moderate amount of modification, thus minimizing modification costs.
Additional instrumentation such as pressure gages, thermocouples, flow
meters, will be  installed in the stoker-boiler system where needed to
provide for  measurement of pertinent stoker and boiler operating parameters.
Damper and valve positions will be calibrated to determine  flow rates.

                                    194

-------
Ports for gaseous emissions, smoke opacity, and particulate loading
measurements will be provided.  In addition, provisions will be made to
sample fly ash from the dust collector and from the settling chambers
ahead of the dust collector, the grate-ash residue, and the grate-ash
siftings.

Candidate Fuels

          Providing that one or more of the treated fuels gives definite
pollution control results with reasonable economic potential, a quantity
of the best formulation ("designed fuel") will be prepared in sufficient
quantities to permit Level 1 and Level 2 environmental assessment
sampling and analysis work.  The 600-hp stoker boiler fires over a ton of
coal per hour so availability of the various treated fuels will have to
be a factor in fuel selection.
Potential Experimental Conditions
          Table II outlines potential operating variables for the Phase II
program.   As in Phase I, criteria pollutants can be utilized to screen
the various operating conditions.  It is envisioned that this experimental
program can be conducted in the following five parts.
          1.  Baseline Performance.  Emission (and efficiency) measure-
ments, plus other observations, will be made and analyzed for this
system firing several coals and size distributions under a variety of
operating conditions to establish characteristic baseline performance.
These tests will establish baseline data and document the effect of
coal parameters on performance of this stoker, particularly as they
relate to modern emission standards.
          Table II includes an outline of pertinent variables (and the
resulting potential operating conditions).  Of these, the coal type will
be the most dominant in affecting the performance of the stoker-boiler
system.  Consideration will be given to high volatile, caking coal as
this type of coal is one of the more difficult coals to burn with a
stoker.  From a research viewpoint, it would be of interest to investigate a
low-volatile coal, about 20 percent volatile matter, but these coals are
                                   195

-------
generally used for metallurgical purposes and are not fired In industrial
boilers.
          Generally, coals available for stoker firing will have a narrow
range of volatile matter.  However, other pertinent coal properties that
affect smoke generation, fly ash emissions, and slag or clinker formation
are
          •  Coal size
          •  Ash content
          •  Ash softening temperature.
These coal properties will be investigated to determine their effect on
boiler performance and air pollutant emissions, while varying overfire
air rates, boiler loads, excess air levels, and fly-ash reinjection rates.
Because low-load smoke is an acute problem of the spreader-stoker, an
additional effort will be expended at this condition.
          Because the facility will be well-instrumented, gaseous and
smoke emission (as well  as boiler operating performance data) can be
recorded continuously and encompass many test points.  In addition, it
is planned to collect particulate samples at selected test conditions.
These particulate samples, the continuously recorded emissions and boiler
operating data, plus various coal and ash analyses, will provide ample
baseline data from which to characterize this system under a variety of
conditions.  From these, the most practical conditions will be selected
for baseline Level I/Level 2 measurements,

          Task 2a.  Modification of Stoker Operation.  Based upon the
results of Task  1  (Baseline Performance), modification in the operation
of  the  stoker  (specifically, the overfire  air and  fly-ash reinjection
systems) will be investigated under those  conditions that are found to
be  environmentally unacceptable.  These  findings will provide a basis
for suggesting modifications in the stoker operation to achieve a reduction
in  emissions  and an improvement in overall stoker-boiler performance.
However, it is anticipated  that modifications in stoker operation to reduce
low-load smoke will need consideration.  Accordingly, an optimum coal will
be  selected to explore  the  effect  of:
                                    196

-------
          •  Overfire jet modifications
             - Jet temperature           i
             - Jet quantity and velocity
             - Jet orientation
          •  Steam Injection above bed
to find their effects on low-load smoke.
          An additional consideration will be to optimize fly-ash reinjec-
tion rates and to optimize the ratio of primary and secondary air flow
rates as a function of boiler load.  Also, from a boiler efficiency
viewpoint, a minimum excess air level will be established and the charac-
teristics of slag formation and emission generation will be investigated
at this excess air condition.  Criteria pollutant data will be collected
during these runs.

          2b.  Modification of Stoker System.  The stoker-boiler system
will be modified to evaluate the emissions control improvement by the
various control strategies.  Table II summarizes candidate control
strategies.  Other promising techniques identified during the program
will also be considered.  Criteria pollutant data will be collected.

          3.  Evaluation of Particulate Control Equipment.  The multistage
cyclone will be evaluated for its fly-ash control capability.  Because
the cyclone is coupled directly to the boiler outlet, it is not possible
to sample upstream of this device.  The collection efficiency of the
cyclone will be determine from a comparison of the collection of fly ash
in the cyclone to the fly ash collected in the flue gas immediately
downstream of the cyclone.

          4a.  Conditions for Level I/Level 2 Sampling and Analysis.
Emissions of all Level 1 pollutant groups and specific Level 2 pollutants
expected  to be responsive to control parameters shall be determined for
each major modification.  The criteria pollutant data (where appropriuiv ,
will serve as an indicator to determine whether Level I/Level 2 pollutant
will be responsive to modifications.
                                   197

-------
          4b.  Level I/Level2Methodology.   The industrial stoker that
will be used in this program is much more complicated than the small
stoker, as can be seen in Figure 4.  It is planned to conduct a complete
fireside Level 1 sampling and analysis run for the industrial boiler for
each coal fired.  However, as previously mentioned, adequate Level 1 (and
Level 2, if necessary) data on the waterside streams will be forthcoming
on another EPA program (Contract No. 68-02-2197).  Thus, we plan not to
sample waterside streams for this combustion study.
          Considering the fuel-side streams, only four streams actually
flow into or out of the unit.  These are the air, coal, ash, and flue gas
streams.  We will not sample the combustion air stream.  Although
there are four separate ash collection points, the collected ash is either
reinjected into the furnace or is combined into one ash stream for disposal.
The coal and ash streams actually operate on a batch basis, rather than
continuously.  Coal is fed into the hopper until the hopper is full and then no
coal is fed until the coal in the hopper is nearly all consumed.  Likewise,
ash is typically discharged to the ash holding tank about once during each
shift.  A number of coal and ash samples will be obtained during the coal
feeding and ash discharge operations to insure that the coal and ash
samples are reasonably representative of the bulk material.  Applicable
ASTM procedures will be used for this sampling.  The individual samples
will be combined into a single sample for each run for Level 1 and
appropriate Level 2 analyses.
          During the Level 1 sampling run of each fuel, Level 1 sampling
of  the flue gas will be conducted.  This will include sampling with a SASS
sampling  train  and other sampling  systems, as necessary,  to obtain complete
Level 1 data.   A complete Level 1  analysis will be conducted on the
samples.  Level 2 analysis will be conducted for pollutants expected to be
strongly  influenced by fuel  and/or operating conditions  (notably POM).
                                     198

-------
Environmental Impact
          Based on the results of the research program, an assessment
will be developed to define the potential of emissions control on the
overall acceptability of industrial stoker coal-fired boilers to meet
out nation's energy needs.  In addition, recommendations for future
applications of stoker modification control will be made.
                              ACKNOWLEDGMENT
          The research covered in this paper is currently being .conducted
under EPA Contract No. 68-02-2627.  The author would like to thank Battelle-
Columbus staff members, Richard E. Barrett and David W. Locklin and former
staff members, Richard B. Engdahl and William T. Reid for their contributions
in this paper.
                                    199

-------
                               REFERENCES
(1)
(2)
(3)


(4)


(5)


(6)
Barrett, R.  E.,  Miller, S.  E.,  and Locklin, D.  W. ,  "Field Investi-
gation of Emissions From Combustion Equipment for Space Heating",
Final Report on Contract 68-02-0251 from Battelle-Columbus Labora-
tories to U.S.  Environmental Protection Agency, EPA-R2-73-084a
(June, 1973).

Locklin, D.  W.,  Kropp, E.  L.,  "Design Trends and Operating Problems,
Part A - Industrial Boiler Population and Design Trends11, Final
Report on Grant R-802402 by Battelle-Columbus Laboratories to U.S.
Environmental Protection Agency, EPA-650/2-74-032 (April, 1974).

"Compilation of Air Pollutant  Emission Factors", U.S.  EPA, Office
of Air Programs, Publication No. AP-42 (April,  1973).

CGA-TR-75-26-G(l) Draft copy of final report of Preliminary Emission
Assessment of Conventional Stationary Combustion Systems.

Particulate Polycyclic Organic  Matter. National Academy of Sciences,
Washington,  D.C., 1972.

Giammar, R. D., Engdahl, R. B., and Barrett, R. E., "Emissions  from
Residential and Small Commercial Stoker-Coal-Fired Boilers Under
Smokeless Operation", prepared  for U.S. Environmental Protection
Agency under Contract No. 68-02-1848 by Battelle's Columbus
Laboratory, October, 1976.
(7)



(8)


(9)



(10)


(11)
"Initial Operating Experiences with Dual-Alkali S02 Removal System",
presented at EPA Symposium on Flue-Gas Desulfurization, Atlanta,
November 4, 1974.

"Standards of Performance for New Stationary Sources", Federal
Register, Vol. 36, No. 139, Part II, pp 15704-15722 (August 17, 1971).

"Analysis of Final State Implementation Plans—Rules and Regulations",
L. J. Duncan  (Editor), MITRE Corp., Washington D.C., MTR-6172,
Rev. 1  (July, 1972).

M. P. Corriveau  and T. Linton,  "Pelletizing of Fine Coals", Joint
Solid Fuels Conference, ASME-AIME, Quebec, 1957.

"Investigating the Storage, Handling, and Combustion Characteristics
of Solvent Refined Coal", EPRI  1235-1, Final Report, October,  1975.
                                   200

-------






^•^
f*">

t^
R
&j
ffi
(X,



c-
CL

1
O

H
S5
O


H
p
O
S
H



H

CO
M
S
O
O

.Jl
w
e

M
tn
en
g

fy*
o
£

o

ri

O
fl
,0
1-4

*
V
o
44
O

(24

C
»
m
1
M











(0
o
•o
•rl
X
o
c
4)
c?
(4
44
•H
JZS



to
C
o
fQ
r4

33


41
•0
1-1
X
g


c
O
rO
H
tt
O

(0
U
TJ
•H
X
O
3
IW
r-4
3
to

10
4)
44
r-4
3
U
4J

id
fl



44

(2


^ xf
in tn

0 0
CM 00
tx CM M> *D IX. O O
. • • . . 4J 4J
O C4 O O O
IX. IX,
CM CM

o o








N CM fO
fl ft xt xt fl CM CM
O O O O O O O

O O O O O O O







xt xt 00 CO CM CM CM
O O O O O O O

o o o o o o o

.





s*\
j^
X**
CO CO in CO CO CO
in CM 
Soo CM o m tn fi
o tn CM o fi o
• • • • • • •
o o o o o o o


M
4)
•£
o

B)
a 4J
p-l 41 (J r-l ft
rt *o td  H fl p. f U
M 4J o 4-> M
1-1 m 0 -o n 4)
H 3 t-C 4) 3 g
9> -O M -O g
> C 0 
M C xy vt C ..
• 3 e8 o «d fl ft
U fO U T* «
P! r-< Iv flC!4>fl O i-l
rt fi u 4>rS BO.4J 3 4) 4J fl g fid
.goo Bcoo T) a to i-i B i-* >H
fll O O *rlO*rlOO 44
rt Pn cj ffl pu "O X> O co-o
% «S| S 0 §
o
«**
 4J
•H C
fiC O
u
V
3 O

^
41
41 M
J5 ^

Ol
XU3

tf>
•o c
41 O
•rl -H
fl U>
P. 0)

4-1 E
fl CD
a
B «)
m crt
rt f )
3
•O O
ft •!-<
O H
M P. 4J
• C
i-4 41 x-v 4)
Id ^3 C U
O 4J O V)
O 4J 41
«'xw CL
t> iJ £,
js a ft 4J
4J 11) J3
O O 00
d ^* >o •<-<
•H 41 fl 01
a. ?
X 4->
01 O 3 fi
0) fl O -H
14-1 M nj •—
O -H x^ 41

4) 4J 3 C
60 £ 4J ^
fd o> 03 td
44 44
4) O O cd
O 0 r-l x^
M ^x
4) fi f> 41
(X W r-l 44
fl) O
44 xt e
.c; i> • 4J
004: o o
•H 44 O
S *O O
C 4)
ot cd * 4*
44 xT ^O •— j

44 - O 4-1
no c
.c «  >J 44  4) 3
Cs* «-4 6 c*
r-t O. O. 61
6 1-1
ji X cr (/)
H M 4) :
"ft ?
201

-------
          TABLE II.  SYSTEM CHARACTERIZATION AND COAL EVALUATION
•  Coal Types
   - Coal size distribution (double screened to 25 percent fines)
   - Ash content
   - Ash-fusion properties
   - Sulfur levels
   - Other physical and chemical properties
•  Excess Air Levels
   - Minimum levels
   - Other levels to establish optimum operating range from both an
     emissions and efficiency viewpoint
•  Boiler Loads
   - Low-load (30 to 40 percent of full load)
   - Partial loads
   - Full load
•  Overfire Air Rates
   - Jets inactive
   - Other rates to establish optimum rate for various boiler loads
     and excess air levels
•  Fly Ash Reinjection Rates
   - Maximum
   - Partial
   - None
Modification of Stoker Operation
•  Overfire Air Modifications
   - Modified velocity or placement (as determined by cold model studies)
   - Preheat temperature
«  Steam Injection Above Bed
•  Optimization of Fly Ash Reinjection Rates
•  Optimization of Primary/Secondary Air Flow Rates
Modification of Stoker System
Candidate modifications:
•  Zoned Air Admissions (to control primary air to parts of bed)
•  Modified Coal Feeding at Low-Load (as by hybrid spreader-overfeed
   approach)
*  Small Oil Burners Fired Above Bed as Afterburners
•  Flue Gas Recirculation
                                    202

-------

[~ / / /' •' •' / S •' S
1 ' /'/ '/V s '' ' ;'





\ 1 J 1 1 1 1 | | I 1 T 1 * 1 1
I 1 1 1 1 1 1 | | 1 1 U I 1 1
| I | I Ll 1 J I I 1 1 1 1 11

x"^ "x "x "X ~x "- .
v X XX X X X X
•v -x_ X X X \ X

1 / / / S S / / / /
\ 	 / /' s' / ./ /" / /

L 	
J"

El 1 ' 1 1 1 1 1 1 1 1
1 1 1 1 M 1 1 HI
| 1 1 M 1 i | H 1 j

"x"x"x"x'x ^s">
X^X ^X ^X ^X. \ ^ s.

' t i / / ,
E
i M IT i 1 1
1 1 HUM
U 1 1 l.H L

s>>x<>-

I,- S / /
[
HT
1 "x^O'vL^Olx^^
O O O O O <
^ CD i^ ^ Cy

0 o-
p o
3C
r\ .
N 2 0
* « in *H ^
. - W CO O\
O , ^-i
0 0 £ .,
iP o g S

Q O C T3
J?) - •* W
CO O)
M <0
01 U
r-l  -i
"
< — rH
0 o •»
/» 0 ^
o 3)
0 ^ £
fO
3
  SJD8A m pd||D|Sut
|D|O| p iU8DJ9d
  '|DOQ 9JJJ Oi
     DUJADH J3|iog
               203

-------
 0)
JZ
TJ
 0>

 ^
"u.
 0)
 o>
T3
 O
    a.
   CO


  i
   /y
•o
 a>
 CD



O




g

   (


.8



8
                                                      a.
                                                      a.

                                                     O
                                                     O
                                                     O.
                                                     o
                                                     o
                                                     to
                                                      I
                                                     o
                                                     o
                                                      n
       a

  i    o
       O
       O
  )    O"
       m
   U M


?    §
       O
                                                     I
                                                     o
                                                     o
                                                     o
                                                     o
M O
, to o\
                                                                          go
                                                                          oo M
                                                                      o  tu  o
                                                                      13  U  U-l
                                                                      ^  at
                                                                      rt u  TJ
                                                                      ai     tu
                                                                          a)  u
                                                                          N  W
                                                                             «0
                                                                             o
                                                                             a>
                                                                          M  M
                                                                          3  O
                                                                          O  fe
                     H
                    CQ

                    T3
                     (U
                    JJ
                     a)
                     E
                                                                   W  C  C
                                                                  W  -H  n)
                                                                   3,
                                         204

-------
      7.00
 £

*b
 O
 o

 E
 o
 o>

 tn
      6.00
      5.OO -
      4.00 -
      3.00 -
      2.00 -
       1.00
      O.OO
              Annuolization     16.7%

              Lood factor      0.30
              Boiler efficiency  0.80
                     Type of boiler system
          0.00
                 i.oo
2.00
3.00        4OO  $/!0  Btu
                                  -J2—14—li  $/bblofoi!
.2  o 0   2   4   6   8    IO

.|  _ OO    05    LO    1*5    ^0     2.5  $/IO^ft3of noturol gas
     cr a
     LU u. o    tO   2O   30  40  50   GO

                              Fuel Cost
                                 Y. $/ton of coal
    Figure 3.  Cost Comparison  of Various Types of Boiler Systems.
                                       205

-------
         Treating
         agents
Water
Water Makeup^
treatment water
1
Spent
treating
agents
Coal 	 *
Air 	 ••



JU
Rein
	 ^Slowdown
Steanr
loop
ection
Stoker


dust
collector

V V
Settlings y
	 	 	 Fly n«?h

K7
Flue go
1


                                              Grate
                                            discharge
                              Grate
                              sittings
                                                                              Ash
                                                                            discharge
                        Figure A.   Schematic  of  Industrial Stoker-Boiler.
                                                  206

-------
FIELD TESTING:  APPLICATION OF COMBUSTION MODIFICATIONS
          TO CONTROL POLLUTANT EMISSIONS FROM
                   INDUSTRIAL BOILERS

                        PHASE II
                          By:

              D. R. Bartz and S. C. Hunter
                   KVB, Incorporated
                   Tustin, CA  92680
                           207

-------

-------
                                   ABSTRACT
        The use of combustion modification has enabled many utility boilers to
meet the emission standards for NO .   Its usefulness in reducing NO  emissions
from industrial boilers (ranging from 3 MW to 147 MW heat input) has been investi-
gated during a recently completed field test program.  The gaseous and partic-
ulate emissions from coal, oil, and natural gas fuels were measured both before
and after the combustion modification.  Data were taken on particulate size as
well as concentration.  Trace species and organics emissions were measured on
selected units.
        The principal combustion modification methods that were investigated
included reduced excess combustion air, staged combustion air, recirculated
flue gas, tuned burners, and reset burner registers.  Staging was implemented
by the use of overfire air ports or by turning off the fuel to 'some burners
and increasing the fuel to others, thus creating zones of fuel-rich combustion.
All of the combustion modification methods were effective to varying degrees in
reducing the nitrogen oxides emissions, and reduction of as much as 50 percent
were obtained with several of the modifications.
        The majority of the results from this program were previously presented
in the first EPA Symposium.  This paper summarizes the overall results, presents
trace species and organics data, and discusses several reports that were pre-
pared based on the results.
        This work was conducted in fulfillment on Contract 68-02-1074 by
KVB, Inc. under the sponsorship of the U.S. Environmental Protection Agency.
This report covers the period September 1974 to  March 1977.
                                       209

-------

-------
                                   SECTION 1
                                 INTRODUCTION

PROGRAM OBJECTIVE
        Industrial combustion devices of all kinds contribute a  significantly
large fraction of the total air pollution from stationary sources.  Combustion
modifications have been demonstrated for utility boilers which can reduce pol-
lutant emissions without degrading boiler efficiency.  Application of these
modifications to industrial combustion devices could have a profound impact
on air quality and energy conservation.
        It was the objective of this program of field testing to determine
the gaseous, particulate, and trace species emissions and the efficiency of
industrial boilers ranging in capacity from 4.5 to 227 Mg (10,000 to 500,000
Ib) of steam/hr, and to determine the reduction of emissions that could be
achieved by modifying the combustion process in a systematic and controlled
manner.  An additional objective was to maintain or improve boiler efficiency.
                                                         •».
        The results of the field test program sought to establish  design
and/or operational changes that boiler manufacturers and operators could make
to reduce emissions and where future combustion research activities should
be concentrated.  The measurement of trace species emissions was conducted
only in as found (unmodified) conditions to determine if industrial boilers as
a class are a significant source of these emissions.
PROGRAM SCOPE
        The program was conducted in two phases and this paper summarizes the
results including results of trace element emission measurements.
        Phase I was one year in duration and involved the selection of 47
representative industrial boilers for testing, construction of a mobile emis-
sions measurement laboratory, and field testing of the 47 boilers for emissions
with the boilers operating normally or with minor operating changes.  The
                                       211

-------
boilers were selected to reflect the prevailing geographical distribution of
boilers and fuels throughout the continental United States.  Consequently, the
preponderance of test boiler sites was east of the Mississippi River.  The
results of Phase I were presented in a final report of that Phase (Ref. 1).
        The Phase II activities were of 14 months duration and involved the
intensive testing of 19 boilers to measure the sensitivity of the emissions
and boiler efficiency to combustion modifications that sometimes required
retrofit of the boiler.  Examples of such combustion modifications were staged
combustion air ports and flue gas recirculation.
        The results of the Phase II effort were documented in several reports.
The major portion of the test results were reported in the Phase II  final
report  (Ref. 2).  All trace species and organics test procedures and results
were presented in a separate report (Ref. 3).
        Two guideline manuals were prepared based on the program.  The first
guideline manual is intended for use by boiler operating personnel and dis-
cusses boiler adjustment procedures to minimize air pollution and achieve
efficient use of fuel  (Ref. 4).  The second  guideline manual  {Ref. 5)  is
intended for use by boiler manufacturers and discusses more extensive  combus-
tion modifications as compared with the boiler operators manual.
        A data supplement was prepared for Phases I and II  (Ref. 6).  This
supplement documents the test data in more detail than was practical in the
final reports.
        Section 2 of this paper  summarizes the baseline  (unmodified) emissions
test results.  Section  3 summarizes the results of combustion modifications.
Trace species and organics measurements are  discussed  in Section 4.  Section
5 is a  review of the boiler operators guideline manual and Section 6 is a re-
view of the boiler design guideline manual.
                                        212

-------
                                   SECTION 2
                             BASELINE TEST RESULTS

INSTRUMENTATION
        Baseline test procedures and results have been presented both at the
first Stationary Source Symposium in Atlanta, GA (Ref. 7) and in a paper to the
ASME (Ref. 8).  Therefore only the highlights of that data will be reviewed in
this paper.
        The emissions measured and instruments used for the program are listed
in Table I below.  The operation of the instrumentation is discussed in detail
in References 1, 2, and 3.
               TABLE I.  EMISSIONS MEASURED AND INSTRUMENTATION
 Emission
Symbol
       Measurement Method
                          Equipment Manufacturer
Nitric oxide
Oxides of nitrogen
Carbon monoxide
Carbon dioxide
Oxygen
Hydrocarbons
Sulfur dioxide
  and trioxide
Total particulate
  matter
Particulate size
Smoke spot
Opacity
Trace Species and
  Organics
  NO
  NO
  CO
  CO,
    4.
  °2
  HC
  SO
  so
2'
  TS &
  O
Chemiluminescent
Chemiluminescent
Spectrometer (NDIR)
Spectrometer (NDIR)
Polarographic
Flame ionization
Absorption/
  titration
EPA Std Method 5
Cascade impactor
Reflection
EPA Std Method 9
Modified EPA Method 5
Thermo Electron
Thermo Electron
Beckman
Beckman
Teledyne
Beckman
KVB Equipment Co.

Joy Mfg Co.
Monsanto
Research Appliance
                                 Train design by
                                 Midwest Research
                                 Institute
                                       213

-------
BASELINE EMISSIONS
        The objective of the baseline emission measurements was to establish
the general level of emissions from industrial boilers as a class and to pro-
vide a base level from which to determine the effect on emissions of combustion
modifications.  Data from both Phase I and Phase II are discussed below.
Nitrogen Oxides Emissions
        The total nitrogen oxides emissions were not found to be significantly
dependent upon boiler size, as is indicated in Figure 1.  However, they were
s,trongly dependent on the type of fuel being fired, and, with coal fuel, strongly
dependent on the burner design type.  This strong dependence is illustrated in
Table II which shows the range and average concentration of nitrogen oxides.
           TABLE II.  RANGE AND AVERAGE EMISSIONS OF TOTAL NITROGEN
                   OXIDES AT BASELINE AND LOW-NO  OPERATION




Fuel Type
Coal

No. 6 Oil

No. 5 Oil

No. 2 Oil

Natural Gas

Range
Baseline
NOX
ng/J
(ppm)
100-562
(164-922)
107-196
(190-350)
112-347
(200-619)
36-101
(65-180)
26-191
(50-375)
Average
Baseline
NOX
ng/J
(ppm)
290
(475)
151
(269)
164
(293)
67
(120)
71
(139)
Operation
Excess
02
%
8.7

5.3

5.8

5.5

4.8

Low-NOx
NOX
ng/J
(ppm)
225
(369)
121
(216)
142
(254)
59
(105)
57
(111)
Operation
Excess
°2
%
6.7

4.9

4.9

4.0

5.0

        The "Low-NO  Operation" column entries are the average of the emissions
                   X
when the most effective combustion modification method for that particular fuel
type was used.  Coal-fueled boilers were the greatest emitters of total nitrogen
oxides.  All nitrogen oxides measurements cited in this paper in parts per
million  (ppm) have been normalized to dry at 3% excess oxygen.
                                       214

-------
\J\J\Jm

0™ 400.
2

in

*>,
^ 400
•
a
a 200
n
| ** | I COAL FUEL
A -
1

— -ti
A A ^
AA A

A4BA ^B^
AA
A A
-A
1 1 1
0) «N
Q 2001
H
X CM
O O
z
u in
o 

. & 200
Tj

11 1
1
jm NATURAL GAS FUEL

A * *
-rf * _ •• •
i HI tt'lli *
nL 1 1 I
E*
O
E-t
cnn
300,

M
i 2oa
in

^ ioa

D^
c
n

o
* 400
• (3)
>
^ 200
• e-
•' ' ' OIL FUEL
•
0
^ * |
94^P 0
^^^fc^ ^-^ i
^•J^Bk^fc.^^^^ ^^^ ™l
E I^^H^V^B ^B ^B
a H^^

J* * i i i
0 100 200 300 40
^
10 Ib/hr of Steam

0 100 200 300 400
                               GJ/hr of Equivalent Saturated Steam
                                            TEST LOAD
Figure 1.  Total oxides of nitrogen emissions at baseload.
                                       215

-------
Solid Particulate Emissions and Size
        The solid particulate emissions were not at all dependent upon the
boiler size, but were strongly dependent upon the fuel type.  The emissions
from oil and gas fuels were usually below the Stationary Source Standard of
43 ng/J (0.1 lb/10  Btu) for boilers of greater than 74 MW, but the emissions
from coal always were well above it.
        Particulates having a diameter of 3 ym or less are believed to be
especially dangerous to public health.  The proportion of the particulate
catch that was in the fine particulate range ran from about 30% to 80% for oil
fuel.  For coal fuel the proportion was less, about 20% to 33% with one instance
of 65%.  Oil fuel also produced more particulates in the size range of 0.4 to
0.7 ym that causes reduced visibility and atmospheric haze than did coal.
Other Emissions
        Hydrocarbon  (HC) emissions  from both natural gas and oil fuels general-
ly were in the zero  to  14 ng/J  (zero to 75 ppm) range.  The two highest base-
line values measured were 35.4 and  101.8 ng/J  {200 and 575 ppm).  The baseline
carbon monoxide  (CO) emissions were normally near zero, although in a few test
cases the boiler were being operated with over 70 ng/J  (200 ppm) of carbon
monoxide emissions.  The sulfur oxides emissions for coal- and oil-fired boilers
were directly proportional to the sulfur content of the fuel except when the
sulfur content of the coal exceeded 2%.  These cases indicated that up to about
one-quarter of the sulfur was not emitted as a gas.
BASELINE BOILER EFFICIENCY
        Baseline efficiencies for coal-fired boilers ranged from 72%  to  88% and
averaged 81%.  The oil-fired boilers  exhibited efficiencies between 72%  and 88%
and averaged 83%.  Gas-fueled boilers had efficiencies  from 70% to 85% with an
average of  81%.
                                        216

-------
                                   SECTION 3
                     COMBUSTION MODIFICATION TEST RESULTS
COMBUSTION MODIFICATION METHODS
        During Phase II eleven methods of combustion modification were investi-
gated.  They are listed on Table III and the postulated effect of the modifica-
tion of combustion which caused the lower nitrogen oxides emissions is indica-
ted.  A total of 52 tests of the eleven combustion modification methods were
conducted.
MOST EFFECTIVE METHODS FOR NO  REDUCTION
                             x
        The combustion modifications that produced the largest NO  emissions
                                                                 A
reductions on each boiler in Phase II tests are shown in Figure 2.  Data are
grouped by fuel type and, for each fuel, shown in order of increased complexity
of equipment modification.  The effect of each method is discussed below.
Excess Air Reduction
        Reduction of excess air was most effective on coal fired boilers, and
relatively ineffective on gas fuel, particularly on units with no air preheat.
NO  reduction with oil fuel was limited to about 20% by excess air reduction.
Burner Out Of Service
        This method was effective in reducing NO  with all three fuel types,
with reductions ranging from 20 to 55% of baseline.  However in most cases
particulates emissions were increased.
Reduced Air Preheat
        A reduction of NO  of about 30% was obtained by reducing the combustion
                         A
air preheat by about 69 K  (125 °F).  There was of course, a loss in boiler
efficiency of about 3%.  This method was only effective in reducing thermally
generated NO , but could not reduce NO  derived from fuel nitrogen.
            •*C                         X
                                       217

-------
           TABLE III.   COMBUSTION MODIFICATION METHODS AND EFFECTS
       CATEGORY AND METHOD
               EFFECT
1.  Fuel-Air Ratio Variation
       Excess air level
       Staged air
       Burners-out-of-service
       Burner register adjustment
Varies the overall fuel to air
mixture ratio

Creates local fuel to air ratio
stratification by bypassing air
and delaying complete combustion

Creates local fuel to air ratio
stratification by bypassing air
and delaying complete combustion

Controls swirl and the local rate
of fuel and air mixing
2.  Enthalpy Variation

       Combustion air temperature


       Flue gas recirculation


       Firing rate
Influences peak gas temperature
level and duration

Reduces peak gas temperature level
and duration

Affects fuel heat release rate per
unit volume, and gas heat loss rate
3.  Input Variation

       Fuel oil temperature and
       viscosity
       Fuel type switching


       Burner tuneup


       Fuel oil atomization method
       and pressure
Controls atomization characteris-
tics; e.g., drop size and vaporiza-
tion rate

Reduces sulfur and/or nitrogen
oxides emissions from the fuel

Assures performance according to
design specifications

Controls local fuel and air mixing
rates by varying drop size distri-
bution and overall fuel spray shape
                                      218

-------
      600
       500
       400
    Cn

    c  300
    o
    z
       200
       IOC
 Test Number
                  o *tf  r- CM in f*>  cr>
                  i-itNt in in co  o
                  i-< »-t  H rH i-l H  CM
                                         cofN nr^ui  
-------
Staged Combustion Air
        Staged combustion air was one of the more effective modifications for
gas and oil firing.  Boilers on which this was implemented yielded NO  reduc-
tions of 20 to 54 percent.  There was an increase in particulates with oil
fuel.
Flue Gas Recirculation
        One boiler was modified for up to 30% recirculation of flue gas and
NO  emissions were reduced 73% with gas fuel and 30% with No. 6 oil.  Recircu-
  X
lation is more effective in reducing thermal NO , the only source of NO  in
                                               X                       X
natural gas combustion.  With oil fuel, nitrogen in the oil also forms NO  and
                                                                         X
recirculation is less effective.
Reduction Potential for Industrial Boilers
        Considering all the boiler and fuel combinations tested in Phase II as
a small sample of the entire industrial boiler population, it is of interest
to examine the NO  reductions achieved as some indication of the potential for
                 X
NO  reduction that may be possible on these devices.  Figure 3 shows the per-
centage NO  reduction achieved by the most effective control method for each
          X
boiler and fuel combination plotted as a function of the cumulative percentage
of the number of boilers and fuel combinations on which NO   was reduced to a
given amount.  For example, of the 24 boiler and fuel combinations tested
reductions of over 60% were achieved on only about 10% of the combinations.
Reductions of 30% or more were achieved on about 50% of the combinations.  Some
degree of reduction was possible on about 90% of the combinations, implying
that for the remaining 10% no reduction was possible.  More extensive use of
the complex methods such as staged combustion and flue gas recirculation would
tend to increase the reductions achievable.
        The boilers selected for test were based on a survey of industrial
boilers in use.  However, on a statistical basis the sample of boilers tested
is'rather small so that the foregoing discussion of reductions achievable
should only be used as a general guide and not as a positive indication of
reductions achievable on the entire population of industrial boilers.
                                       220

-------
       80
       70
       60
    o  50
    M
    EH

    U

    Q
     x 40

    §
    2
    w
       30
       20
       10
                   i     i      i     i—rn—i—i—i	1—i	r
             J	i	i	i	i	i   i   i    i
         12     5    10     20   30  40  50 60  70   80    90    95    98  99


                         NUMBER OF BOILERS, CUMULATIVE PERCENT
Figure 3.   NO  reductions  achieved in Phase II as a function of number of

           boilers reduced.
                                       221

-------
DATA SUPPLEMENT - PHASES I AND II

        The large amount of data collected in the Phase I and II programs

could not be presented in total in the final reports.  A data supplement has

been prepared that tabulates the entire data set (Ref . 6) .  The purpose of the

data supplement volume is to document data in greater detail than was practical

in the final reports for this contract.  It is intended to provide the neces-

sary details to other researchers who are interested in performing their own

analysis.

        For each unit tested, the data supplement contains the following

information:


        1.  Industrial Boiler Design Data

            Design data tabulations include:  unit description, installation
            date, steam conditions, furnace characteristics, burner type and
            spacing, fuels burned, and design temperatures for fuel, air and
            stack gas.

        2.  Boiler Emission Measurements and Control Room Data

            Emission measurements tabulated for each test include test type
            test load, combustion conditions, gaseous emissions, total partic-
            ulates, boiler efficiency, Bacharach smoke spot number, and plume
            opacity.  Control room data tabulated include steam pressure and
            temperature, combustion air and stack temperatures, fuel flow,
            fuel temperature and pressure, furnace pressure, and burner air
            register setting.

        3.  Cascade Impactpr
            Impactor data include impactor flow conditions, stage collection
            weights, and calculated aerodynamic particle size for each size.

            Emission Data Plots

            Plots of gaseous emissions of NOX, HC, CO, and smoke as a function
            of excess O_ at a given load are included.

            Stack Traverse
            Plots of stack traverse data for 02, velocity and temperature as
            a function of position in the stack are included.

            Fuel Analysis

            Laboratory reports of fuel analysis are included as received.
                                       222

-------
CONCLUSIONS
        The emissions of industrial boilers as a class are not significantly
dependent upon size, but they are very dependent upon the fuel burned.   The
effect of the type of burner employed is subordinate to the effects of fuel
and to the characteristics of the particular burner and boiler combination.
        The nitrogen oxides emissions of the boilers tested were at or above
the Stationary Source Standard [which actually applies only to new boilers
with,a heat input larger than 73 MW (250 x 106 Btu/hr)]  in 30% to 40% of the
cases.  The particulate emission standard of 0.1 lb/10  Btu was exceeded in
16% of the tests of oil fuel and in 100% of the tests of coal fuel.

        The combustion modification methods that have been successful in
reducing the nitrogen oxides emissions of utility boilers are also effective
in reducing the emissions of industrial boilers.  The NO  reductions achievable
range from less than 10% up to 70% with a median reduction of about 30% for
50% of the boiler and fuel combinations tested.  NO  reductions are highly
dependent on the boiler and fuel combination and the control method applied.
No one single method was found universally superior.
        The reduction of NO  was found to generally result in increased
particulates emissions.  The hydrocarbon, carbon monoxides, and sulfur oxide
emissions were relatively unaffected by the combustion modifications.  In most
of the tests, the boiler efficiency either remained unchanged or increased by
up to 4 percentage points.  In some cases efficiency was reduced by up to 5%.
        The purpose of this program was to investigate the general applicability
of combustion modifications to industrial boilers.  Cases where problems
occurred with increased particulate emissions, reduction of efficiency, etc.,
could not be fully resolved within the scope of the test program.  The actual
implementation of combustion modifications on a full time basis may require
additional or improved boiler instruments, and replacement of some burner
hardware.  The application of sound combustion and boiler engineering knowledge
can minimize the effect of many of the operational problems that were encoun-
tered in this program.
                                       223

-------
                                   SECTION 4

                       TRACE SPECIES AND ORGANICS TESTS


INTRODUCTION

        Sampling of four coal fired industrial boilers was conducted during
Phase II to determine the emissions of 19 trace and minor species and poly-
cyclic organic matter (POM).   The emissions of the trace and minor elements
were related to total quantities of each element present in the fuel by
examining the degree of mass balance and element partitioning based on
fuel input and element output in furnace deposits, fly ash and flue gas
vapor.  The tendency of volatile elements for enrichment of finer particulate
was examined by analysis of cascade impactor samples.  The primary emphasis in
the trace species tests concerned emission measurements in the field using
existing sampling trains as opposed to measurement method or sample equipment
development.

        The trace species identified and quantified were the following:
        Antimony, Sb
        Arsenic, As
        Barium, Ba
        Beryllium, Be
        Cadmium, Cd
        Calcium, Ca
        Chromium, Cr
Cobalt, Co
Copper, Cu
Iron, Fe
Lead, Pb
Manganese, Mn
Mercury, Hg
Nickel, Ni
Selenium, Se
Tin, Sn
Titanium, Ti
Vanadium, V
zinc, Zn
Chlorides
Florides
Sulfates
In addition to these elements, four polycyclic organic materials  (POM) were
measured:  (1) 7-12 dimethylbenz[a]anthracene, (2) benzofajpyrene, (3) 3-
methylcholanthrene, and (4) dibenz[a,h]anthracene.  Complete results are

presented in Reference 3.


TRACE ELEMENT AND ORGANICS SAMPLE  COLLECTION

         The  trace element sample collection apparatus that was used  for  flue
gas  sampling was developed by another EPA contractor, and it is shown schemat-
ically on  Figure 4.   The  collection train was designed for collecting both
                                       224

-------
SAMPLE COLLECTION
    SECTION

      CYCLONE
   PARTICLE
  COLLECTION
   SECTION
  VAPOR COLLECTION SECTION

  FRONT FILTER
                TENAX HOLDER
                                               lOOOOC
p-



—



r
1
^_^_«_i^^^_^b_

M^
^^^





f—




— •




-n


                                                              TO
                                                             PUMP
   Bubbler
    No. 1


       2


       3


   Middle
   filter
Dry
Tenax Adsorber
Bubbler
 No. 4


    5

    6


    7

    8
                                        , HN0
H2°2' AgN03' HN03


H2°2' AgN03' HN03

Dry


Silica Gel
             Figure 4.  Trace element collection apparatus.
                                 225

-------
inorganic elements and POM.  The chemicals that were selected for the bubblers
are indicated on the figure.  Tenax, a granular gas chromatograph column-packing
material, was placed downstream of the third bubbler to act  as an adsorbing
filter.  The Tenax was to extract all of the remaining POM from the sample
stream before it passed into the bubblers that contained the oxidant.
        The particulate size in the flue gas was measured using a Brink Model
"B" Cascade Impactor with a precutter cyclone.  To improve the accuracy of the
weighing, the particulates were collected on an aluminum foil substratum that
was placed in each steel collection cup.
        Samples of fuels and ashes in the boiler and dust collector were
obtained during the periods of stack sampling and submitted along with the
stack samples for analysis.
SAMPLE ANALYSIS
        The laboratory analysis of the trace element samples was conducted by
the Midwest Research Institute, Kansas City, MO.  The analysis methods are
summarized below.
        A portion of each bubbler solution was heated with permanganate and
nitric acid prior to analysis for mercury and selenium.  Sulfuric acid was
added to another portion of each impinger solution and heated in a reflux
condenser before analysis for arsenic and antimony.  The ash and particulate
samples were prepared for analysis by digesting 0.5 g of solid with 3 ml of
hydrofluoric acid and 1 ml nitric acid in a Parr bomb at 130 °C for 18 hours.
Boric acid was added  (1.5 g) and diluted to 25 ml.  The oxidizing bubbler
liquids were analyzed directly for the remaining elements.
        Mercury was analyzed by the cold vapor technique.  Antimony, arsenic,
and selenium were analyzed by atomic absorption following addition of sodium
borohydride.  The volatile hydrides that were formed were swept into a nitrogen-
entrained air-hydrogen flame and analyzed.  The other metals were analyzed by
conventional flame-atomization atomic absorption spectrometry.
        The POM in the ashes was isolated by continuous Soxhlet extraction with
benzene.  The POM in the benzene-soluble fraction was separated from aliphatic
and heterocyclic compounds by column chromatography with activated silica gel
                                       226

-------
as the adsorbent.  Following the isolation of the total POM, gas chromatography
 (with a Dexsil-300-packed column and flame ionization detection) was used to
quantify the individual POM.
RESULTS
        Three trace element emission phenomena were investigated during the
program:   (1) the material balance, (2) element partitioning, and  (3) particu-
late enrichment.  The size distribution of the particulates was also measured
as part of the particulate enrichment program.  In addition, the amounts of
POM in the coal, ashes, and stack gas were determined.  The results are pre-
sented in detail in Reference 3 and are summarized below.
Material^ Balance
        The degree of material balance for each of the elements of interest
was evaluated using information from the field tests, and the results are
shown in Figure S.  Mass balances were judged to be acceptable when the total
quantity of an element presented in the collected ashes and stack gas was with-
in 75 to 125% of the amount present in the fuel.  This criteria was established
based on statistical analysis performed for a previous EPA contract  (Ref. 9).
        Most elements that are not expected to volatilize during combustion
tended to be balanced or overbalanced.  Barium, iron, maganese and titanium
were overbalanced in some tests.  Calcium balanced well in three of four tests.
Cobalt was underbalanced in all tests.

        Elements that are expected to volatilize during combustion tended to
be underbalanced, particularly arsenic, selenium,  cadmium, zinc, lead and
tin.   However, copper and the most volatile element, mercury, were within the
acceptable balance range for two and three of the four measurements,  respec-
tively.
        Elements with volatility characteristics that are not clearly defined
tended to be balanced or overbalanced.  These elements include chromium,
nickel,  and vanadium.
                                       227

-------
            ELEMENT COLLECTED/ELEMENT IN COAL, %




          0        50        100      150       200
Arsenic
Barium
Beryllium
Cadmium
Calcium
Chromium
Cobalt
Copper
Iron
Lead
Manganese
Mercury
Nickel
Selenium
Tin
Titanium
Vanadium
Zinc
Test
169-3
166-9
166-10
166-11
^ ,

^^^™
	

"••^ ^— ^V
=_

Wxj
"* 1
HUBHHmm^^Mft I
.*.•.*.•.*.*.
vX-Xv
t-.*. %•.*.•.'
ix-x-x*:-
I
I
— i
p
1

»




.-.v.-.-X-
|:>>x:x::




^Sx^
.•.%"X-»"3|
****"» "•^•"•"t


:x:>x::|
-:.:-x-:-:.1
pi

pi
— rrrr-.J 	
t-"-X'X'!i
yii —








(
->

k

=4
ACCEPTABLE
BALANCE
Figure 5.  Trace element material balance.
                    228

-------
        The unacceptable balances, particularly of the more volatile elements
were attributed to low collection efficiency of the bubbler liquids for these
elements.  A significant portion of the elements that entered the boiler with
the coal evidently were vaporized and passed through the filters and bubblers
without being condensed.
Element Partitioning
        Element partitioning theory indicates that certain elements will tend
to be concentrated in certain parts of the boiler, depending on the tempera-
ture of the flue gas in that part.  The highly volatile metals will tend to
be deposited in increasing concentrations on the fly ash as the flue gas
carrying it cools in passing through the boiler.  The less volatile metals
will tend to condense early and to remain in a more uniform concentration
throughout the boiler.  The results of the field tests were in general
accordance with partitioning theory.
        The increasing concentration of the highly volatile elements in the
fly ash during Test 169 is illustrated in Figure 6.  Copper, for example, is
relatively volatile and the concentration steadily increased from the furnace
bottom to upstream of the dust collector and then to downstream of the collec-
tor.  In this instance zinc did not partition, and it was an exception to the
general trend.
        The partitioning of the less volatile metals is shown in Figure 7,
The concentration in the furnace bottom ashes increased over the concentration
in the coal and then remained relatively constant.  There was little partition-
ing between the bottom ash, the fly ash upstream of the collector, and the  fly
ash downstream of the collector,
Particulate Enrichment
        Particulate enrichment theory  indicates that those elements which are
volatile at the temperatures of combustion tend to condense or absorb on the
smaller size fly ash.  Because the finer particles have a higher surface area
per unit mass than do the coarse particles, they will contain a relatively
higher concentration of the more highly volatile elements.  Those elements
which are less volatile in the combustion zone form the fly ash upon which  the
volatiles later condense.
                                       229

-------
     Cn
     H
     J
     W
     §

     a
     §
     u
             IO

              0
              10
         Cd
            300
          Cu
            200
Pb
Se
             ZO
          Sn
   ZOO
Zn
            1000
                 COAL
            FURN.
             BOT.
                UMJLLL
                 rrrTrri
 DUST COLLECTOR
BEFORE  IN  AFTER
            uitin
                      //////?>
                      f////f
 UllJJi
                             7/77T
            7MiMm

                                  mmM///>
 Figure 6.  Partitioning of highly volatile elements, Test 169.
     0!
     a.
     w
     s
     o
     W
     8
                             DUST COLLECTOR
                            BEFORE  IN  AFTER
Figure 7.  Partitioning of moderately volatile elements. Test 169.
                           230

-------
        The particulate size results from the field tests were consistent with
particulate enrichment theory, as illustrated for the highly volatile elements
in Figure 8.  For example, there was a marked increase in the concentration of
antimony  (Sb) as the particulate size decreased.  The enrichment trends of the
moderately volatile elements are depicted in Figure  9.  The moderately vola-
tile manganese, for example, did not exhibit a well defined increase in concen-
tration as the particulates became smaller.  As with element partitioning,
there were exceptions to the general trend, e.g., titanium  (Ti) did enrich at
the collector outlet in the particular test series.
Polycyclic Organic Material (POM)
        The occurrence of POM in the coal, ashes and stack gas  was investi-
gated.  POM 7-12 diraethylbenz[a]anthracene, was not found in any signif-
icant quantities.  Benzo[a]pyrene, and 3-methylcholanthrene, were present
throughout:  in the coal, in the ashes, and in the stack gas.  Dibenz[a,h]-
anthracene was found in the coal but not found in the boiler.  The concentra-
tions decreased at successive locations in the boiler, indicating that
partitioning was not taking place.
CONCLUSIONS
    1.  Sample collection methods based on modifications of EPA Method 5
sample trains proved to be unsatisfactory for collection of volatile metals
based on mass balance criteria.  This was attributed to low collection effi-
ciencies of the liquid bubblers in the sampling train.  The sampling system
used in this program has been superceded by the Source Assessment Sampling
System (SASS) developed by EPA.  That system has promise for more efficient
collection of trace species and organics.
    2.  Material balances for trace elements varied from less than 10% to over
200% recovery of the amount of element input in the fuel.  Elements that were
most difficult to recover included antimony, arsenic, cadmium, cobalt, lead,
selenium, tin and zinc.  An excess amount was recovered most frequently for
barium, beryllium, chromium, manganese, nickel, titanium, and vanadium.
Acceptable recovery was achieved more frequently for calcium, copper, and iron.
While recovery of mercury was judged acceptable in several tests, the majority
of the mercury was found to be contained in boiler ashes rather than in flue
gas vapor.

                                       231

-------
         1000
       Sb
       Cd
          40O
       Cr
         1000
        tt
             0
             O.I
        0.3
3.0
10   20
                   AERODYNAMIC DIAMETER, \im
                 Collector Inlet          Collector Outlet
 Figure 8.  Particulate enrichment by highly volatile elements.
           100
       Co
         2000
       Mn
             0
       20000
       Ti
             0
         SOOO
        B«
10

 0
 O.I      O.3      I.O     3.0

        AERODYNAMIC DIAMETER,
      Collector Inlet
                                               IO   2O
                                          Collector Outlet
Figure 9.   Particulate enrichment by moderately volatile  elements.

                               232

-------
    3.  The results for partitioning of the elements at specific points in the
boilers were consistent with expected partitioning characteristics.   The more
volatile elements were deposited in increasing concentrations on the ash as
the combustion gases passed to successively cooler parts of the boiler system.
The less volatile elements were more uniformly distributed.
    4.  The results for particle size enrichment were also in agreement with
expected enrichment characteristics.  The more volatile elements tended to
concentrate on the smaller size particles, while the less volatile elements
were more uniformly distributed with particle size.
    5.  Polycyclic Organic Matter (POM) was found to be present in some of the
coal, ash and stack gas vapor samples.  Samples were analyzed for four specific
POM compounds.  All four compounds were found to be present but not in all
samples.  The quantity of stack gas particulate samples collected was not
sufficient for quantification of POM.  The data for presence and quantity of
each POM compound in different parts of the boiler and for replicate samples
were not sufficiently consistent to allow positive conclusions regarding the
formation or emissions of these compounds.
                                       233

-------
                                   SECTION 5
                   BOILER PERFORMANCE IMPROVEMENT GUIDELINES

INTRODUCTION
        A document entitled "Guidelines For Industrial Boiler Performance
Improvement" was prepared by KVB and issued jointly by the U.S. Environmental
Protection Agency and the Federal Energy Administration (Ref. 4).
        The guide details boiler adjustment procedures to minimize air pollu-
tion and to achieve efficient use of fuel.  The guidelines are intended for
use by.
        1.  personnel responsible for boiler operation to perform
            efficiency and emissions tuneup
        2.  plant engineers to initiate maintenance and efficiency
            monitoring practices
        3.  as a supplement to boiler manufacturers.
        The guidelines deal primarily with boiler adjustments which are
typically within the control of boiler operators and plant engineering per-
sonnel.  However, other techniques are also described in appendices which will
usually require the assistance of outside combustion specialists due to the
more critical dependence on specific boiler design, operating conditions and
fuel characteristics.
        Since efficiency and emissions are sensitive to many of the same
boiler operating parameters, it is essential that both of these areas be
simultaneously treated in one integrated set of guidelines.  The recommended
procedures are based on the results of U.S. Environmental Protection Agency
and Federal Energy Administration sponsored programs  (Refs. 2 and 10) during
which various nitrogen oxides reduction methods were evaluated in the field
and improvements in boiler efficiency were demonstrated.
                                       234

-------
        The guide begins with a simplified discussion of the fuel burning
process to provide the boiler operator with a basic understanding of NO
formation in the boiler and the major combustion-related factors important to
efficiency.  With this background, the techniques for reducing NO  and improv-
                                                                 X
ing boiler efficiency are then discussed in detail and instructions are pro-
vided to assist the boiler operator in applying them to a particular boiler
installation.  Emphasis is on reducing the boiler excess air to the lowest
practical level.  This will require that the operator have a working knowledge
of the boiler's combustion control system and the required flue gas analysis
instrumentation necessary to perform the boiler adjustments safely and
effectively.
        The major sections of the guide are:
        INTRODUCTION
        FUNDAMENTALS OF COMBUSTION
        PREPARATION FOR BOILER TESTS
        EFFICIENCY IMPROVEMENT AND NO  REDUCTION TECHNIQUES
                                     X
        MAINTAINING HIGH BOILER EFFICIENCY AND LOW NO
                                                     A
The guide contains appendices that deal with more complex NO  reduction
techniques, cost of combustion modification, efficiency improvement, and
combustion generated air pollutants.
        The major features of each section of the guide are discussed below.
FUNDAMENTALS OF COMBUSTION
        This section of the guide discusses the fundamentals of excess air,
NO  formation and boiler efficiency.
Furnace Excess Air
        Excess air control is the key operating variable within the guide.
The advantages in the use of excess oxygen (O.J rather than carbon dioxide
                                             £*
(C02) as a measure of excess air are emphasized.  Many boiler operators are
accustomed to judging boiler firing conditions based on stack CO,.  The use
of excess O  is preferred for the following reasons:
                                      235

-------
        1.   The relation between percent ©2 and percent excess air is
            only slightly affected by fuel composition whereas the level
            of CC>2 for the same excess air is significantly different
            for gaseous, liquid or solid fuels.
        2.   Measurement of C02 requires a much greater precision than for
            02 to obtain the same precision in determining excess air.
        3.   Oxygen level is a direct measure of excess air and can be
            reduced continuously toward zero as excess air is optimized.
            C(>2 increases as excess air is reduced and reaches a broad
            maximum so that optimum excess air may be somewhat difficult
            to define.
        4,   Instrumentation for 02 measurement is generally less expensive
            and more reliable than that for C02-
        The guide discusses the importance of correctly understanding the
difference between excess air in the stack and excess air at the burners in
multiple burner boilers, in boilers with overfire air, or as the result of
casing air leaks.
Nitrogen Oxides
        The fundamentals of NO  formation by thermal or fuel nitrogen paths
are described in terms of the factors that influence NO  control.
                                                       x
Boiler Efficiency
        The various sources of heat loss in a boiler are discussed and the
variation of each type of loss is illustrated as functions of excess Q  ,
Figure 10.  Achievement of minimum loss by optimizing excess air is emphasized.
The effect of changes in excess air on boiler efficiency is presented, as
shown in Figure 11.
PREPARATION FOR BOILER TESTS
        The successful accomplishment of boiler improvements requires careful
attention to test preparation and the guide emphasizes the necessary prepara-
tion  steps of boiler inspection, proper stack instrumentation, flame appear-
ance, and sampling and test procedures.  An inspection checklist is provided
in the guide, summarized on Table IV, defining inspection steps necessary to
check the burners, controls and furnace.
                                       236

-------
        25
                     Total Efficiency Loss
        20
        15
     o
     Z
     W

     0  10
                            Flue Moisture
                         Dry Flue Gas
                                     Radiation
                           Combustibles (Carbon Monoxide)
                                EXCESS OXYGEN (0_),  %
Figure 10.  Variation in boiler efficiency losses with changes in excess O .
                                      237

-------
  Z  -10
  O
  H
  E-
  U

  Q  .09
  < .08
  w
  E-i
  is
  w
  s
  w

  I
  U
  s
  w
  H
  U
  H
  En
  In
  U
  cu
     .07
     .06
.05
     .04
.03
.02
     .01
10




11.1





12.5





14.3





16.7
                                                                  50.0
                                                                      100.0
                                                                         H



                                                                         I
                                                                             H
                                                                             U
                                                                             z
                                                                             H
                                                                             H
                                                                             dP
                                                                             a;

                                                                             g
                                                                  20.0  -
25.0  H

      U
      D


33.3  I
                                                                        OS
                                                                        M
                                                                             tt
              200
                     300         400         500


                        STACK TEMPERATURE,  °F
                                                              600
Figure 11.  Curwe showing percent efficiency improvement per  every one  percent

            reduction in excess air.  Valid for estimating efficiency improve-

            ments on typical natural gas, #2 through  #6 oils  and  coal fuels.
                                       238

-------
   c
   o
tfl  -H
VI  XI
oi  id
           01  VI

           5  8,

           *  8
           O  0,
           oi
           X  rl
           o  o
              «w
                   I
                   Vt
                   o
                   Q.
                   s
                   a,
%

id
01
3
M
0


to
41
x>
id
Vi


X
XI

A:
o
i>
x
o




•
in

10

C
.3
XI
u
11

£J
•H
01
VI
1
VI
11
fj
•H
O

•O
c
id





g
-H
4-1
-H
0)
a

VI
01
a
o
vt
ft

li
X
XI
g
V|
•H
IH
C
o
u















111
VI

Q,

m
•o
en
c:


o
•rl
Xl





J

c

M
•H
n

rH
id
o
o

rl
41

O
ft

J>,
*M

l-j
6)





•
                           vi  id
                           D  in
                           0!  ID
                           O  01
                           vi  o
                           Qi  II
                                     •H -H
                                     u at
                                     »z
                                     O id
                                     •H  0)
                                      0)

                                     IH  id
                                         C

                                     rH  01
                                     U -H
                                     O

                                     fS
                                     Ha
41  I
M  id
3  rH

B£
01  41
Vi  Vi
ft
   4J
u  a
a
      at
      Vl
 . >, O rH
cn  Vi xi  4)
01  id id  3
H  in rH
   10  •
                                                        O  rH   X
                                                        XI  II   XI
                                                                   3
                                                                _  cn It
                                                                O  41  XI
4>
w
nl
j^
c
•H
rH

,_!
o
Vl
XI
C
0
0

c
•rl


id
rH
a.
s

1)
XI
Id
C
•s
•rl
rH
U
Vl
3
0)
in
01
Vl
ft

xi
41
H
C
•rl

J>^
rH
a

3

1
rH
Oi
3
IH

D
3
ifl
0)
_*
id
VI
3
10
in
id

o
X)

j-j
en
O
c
u

X
o>

X
01
Vi
Id

to
Vl
s
id
rH
3
S1
IH
cn
C
-H
Vi
• rl
in


rH
id

Vl
0
IH

to
4)
VI
3
CO
in
01
Vl
a

xl
4J
3
O
O
XI
IH

XI
id
f.
XI

10
XI
c
4)
E
U
rH
01

rH
O
VI
4-1
O
O


C
id
XI
0
41
IH
IH
S



•o
c
Id.
6

t3

g
«
01
XI
to

en
c
•rl
>,
Vi
Id
>

o

rH
X
4->
10
u
tn
3
•r~l
•o
id


rH
Vl

C
id
o

to
Vl
u
rH
rH
O
Vl

Cn
C
•rH
C
o
•rl
Xl
u
c
3
Qi xi IH
0 C
Vl
a
E
•H

>1

V
U
SO
3
Id
O
Cn
C
xl
C
3
O
u

Vl
01
XI
tn
i

o

XI
I
XI
3
id

41
VI
id

10
01
Cn
IH
-H
XI
C

•8
•rl

C
• rl

XI
a
•rH
in
in
id

O
XI
id
Cn-o

rH
f-H
n
XI
id

XI
01
tl
in
01
X)
id
Vl
X!
-rH
rH
id
O
4)
Id








Vl
3
O
g

l-^l
HI
X
XI

10
id
in
e
01
rH
2
ft

2
Xl
£

•0
c
Ifl

to

u
0
r-l
U
01
xl .
C O
•r4 rH
O

XI VI
11 QJ
a o

u
rH Vl
rH IB
id
10
U XI
Vl -rl
3 3
in u
Vl
U -H
ra u
surfaces for exces-

1
XI

41
•O
10
to
id
tn
Vl
1)
rH
•rl
Q


11
X
XI
XI
u
41
- Ql
to
H



•
U
0)
41
X
0
t}
id
01
rH
01
in



•
v.
•rl
rH
3


*g
id

Ul
XI

10
a
D
T3
41

•H




iciencies.
lu
11
Vl
4)

•H
o

^
C)

f^
•a
£
id

10
01

3
Xl
id
Vl


.1




u
41
H
•H
S
01
5
X
[0
id
c
o
•rl
XI
id
rj
41
Qi
0
rl

Q
r-t
4J
8
to


41
ft
0
Vi
ft


IH
-H

>



'
m
•H
11
•o
Q)
i

%^
j>
id
j£
Q)

i
01
Vl
0
XI

^,
rj
*
<0
01
u
g

IH




s passages and
id
Vl
01
rH
•rl
Jg

u
X
XI
e
-H
to
X
Id
41
rH

^,
C
Id


•rl

M 10
11 id
u oi
Id rH
c
V4 &
3 C
UH -rl
to
u id
X O
4-1
xl £
U id
01
Ql rJ
U -n
c id
I-H Oi
11
M
tJlHJ
C C
--H id
rH


rH
Vl
O
rH
U
o
V,
Id

10
a
L"
"3
C.

£
o
XI
O
§

O




n essential part of
id
to
•rl

c
o
•rl
4-1
id
Vi
o
10
xt
o

i •
Cj JJi
r-H C
IH -r|
XI
iu in
U 4!
a xi

tn vi
41
Xl rH
C -H
id O




9
tn
•a
c
id
-li
to id
li
•B T3
01
Vl C
01 VI
it*
Oi en
11 -rl
X 01
xi XI
IXI rH
o
M
01 41
N QI
e xi
s.
Id X
XI
x~
Xl i-H
01
VI Vl
3 O
10 IH
U 01
18 -H
£ in
.


^
Vl
XI
e
Q
&
Vl
tl
c
D
XI
41
X
XI
ri

•o
Vl
id
Vi
tl)
g
•rl
•H
Vl
o
I
10
01
3>
id
to
10
id
Oi
3-
XI
1
rH
•rH
o
XI
o
&
W)
c
HH
.
its to assure proper
m
a
4


cn
Vi
o
01
o
o
c
us
11

Vi







.
c
VI
XI
Id
s
2
a.
in
rH
•rH
o

in
•H
4)
rH
Vl
11
to
3
IH
IH
•o
VI
11
c
Vl
3
•°
01
XI
XI
S
XI
11
Vi
3
10
0)
1
.
located with respect
.-7
M
1
VI
a.
in
-rl
I

41
o>

15
•a
XI
8






f
DI
•H
XI
c
3
1
rH
-H
0
41
X
XI
5

positioned properly
10
•H
i
en

rH
o

id
£
n
01
in
S

_*
u
o
.
x>
O
xi
01
5
XI
Id
xl
•o
c
id
xl
id
O
Vl
XI

rner
3
ft
41
XI
C

*

I
XI
•rH
•O
i

•a
§
Cn
c
•H
to
•H
&
O
XI
O
id
IH
11
rl

nd temperature at the
id
0)
V4
3
in
to
4)
Qi

rH
•H
O
01
Ql
8
o.
-S V4
IH 41
•n £
Vl V4
Sri
.
                                                               01
                                                               Vl
                                                               3
                                                               to
                                                               10
                                                               4)
                                                               V4
                                                               ft
                                                               01
                                                               XI
                                                               at


                                                               k
                                                               xi
                                                               id
                                                               S
                                                               a
                                                              •rl
                                                               IH
                                                               01

rH M 1
rH H -3 in
•J 41 h 41
4J 41 o
4J -H Q, -rl
id -H O WH
X H x in
•H XI ••Id
rl H O>
« 41 id
> it u IH
in <-H o
•O o

id o • c
0) 01 -H
ID rH O C"
41 «C Id tP
O rH 3
•H ft r-l
04 • Ol
•H -O C
.4 a> — i xi
O xl C.
U 01 41
C 3 Vi >
O h id 41
•r4 XI H
xi in to Oi
a xi ft
o 0 >d 0
•n C Vl XI
C 3 4J
1 11 01 rH
tn vi H IH
id id 3 41

in in o
4-1 U -H Vl
O Cn 0 Ou
U id E
d 10 Cn
in ui -d c
C Id C -rH
M Q| Id Xl

.
m
rl
11 M
to O
3
IH IU
IH IH
•rl O
•a i
•a
IH 41
O C
Vl
§.§
•rl
4J >.
id c
xi id
C
01 14
•rl O

O

•o o
§3

c
0 •
•n u •
XI XI W)
Ifl 41 XI

O » id
rH 10 0,
Cl
M c: Vi
alt) 41
U C
0 VI
V< 10 3
ft Id X)

B ^ Cn
C - C
•r4 U --H

c. 3 10
O ft 'rH
U (D €

.
                                                                                                                01

                                                                                                                •g
                                                                                                                u

                                                                                                                ?
                                                                                                                id
                                                                                                                <0
                                                                                                                01
                                                                                                                a
                                                                                                                 id
                                                                                                                 o
                                                                                                                 o

                                                                                                                 Vl
                                                                                                                 id
                                                                                                            Di T)
                                                                                 239

-------
        The guide recommends measurement of the following parameters in the
stack:
        excess oxygen {or CO,)
        carbon monoxide (CO)
        oxides of nitrogen  (NO, NO_)
        opacity (smoke density)
        stack temperature
Instruments required to measure these parameters and selection of the stack
sampling site to obtain accurate measurements are discussed.  The ability
for visual furnace observations is noted as important to ensure acceptable
flame conditions.  Procedures for properly conducting tests are discussed,
including necessary cautions to be observed while making adjustments, checking
of safety interlocks, flame appearance, etc.  The boiler test data that should
be recorded is listed and a sample data sheet is included.
EFFICIENCY IMPROVEMENT AND NO  REDUCTION
                             X
        The guide basically deals with the procedures for obtaining the lowest
possible excess air as this boiler improvement method is the one method that
can be most readily implemented by boiler operators to reduce NO  and increase
efficiency.  Step-by-step procedures are presented for establishing the
optimum level of excess air as a trade-off between low O  and acceptable
levels of smoke and combustibles.  A maximum level for CO of 400 ppm on gas
fuel  is recommended.  For oil fuels maximum smoke numbers are recommended  for
each  fuel grade.  The following levels of excess O  are indicated as typical
minimum levels at high firing rates and serve as guides in  establishing
minimum O_ on a given boiler:
        Natural Gas
        Oil Fuel
        Pulverized Coal
        Coal Stoker
0.5 - 3.0% 02
2.0 - 4.0% 02
3.0 - 6.0% O.
4.0 - 8.0% O,
The adjustment procedure steps for low excess air operation are summarized in
Table V.
                                      240

-------
                   TABLE V.   STEP-BY-STEP BOILER ADJUSTMENT
                     PROCEDURE FOR LOW EXCESS 0  OPERATION


 1.   Bring the boiler to the desired firing rate and put the combustion controls
     on "Manual".   Make sure all safety interlocks are still functioning.

 2.   After the boiler has stabilized, observe flame conditions and take a
     complete set  of boiler and stack readings.

 3.   Raise the excess air until the stack excess 02 has increased by 1 or 2
     percent.  Take readings after the boiler has stabilized and note any
     changes in flame conditions.

 4.   Return the excess air to the normal level and then slowly reduce the excess
     air in small  steps.  Watch the stack for any signs of smoke and constantly
     observe the flame.

 5.   Continue to reduce the excess air stepwise until one of the following
     limitations is encountered:

        Unacceptable flame conditions such as flame impingement on furnace
        walls or burner parts, excessive flame carryover, or flame instability.

        High carbon monoxide  (CO) in the flue gas.

        Smoking at the stack.
        Incomplete burning of coal fuels leading to high carbon carryover or
        increased quantities of solid combustible matter in the refuse.

        Equipment-related limitations such as low windbox/furnace pressures
        or built-in air flow limits.

 6.   Obtain as many readings of CO, excess oxygen, smoke number, and nitrogen
     oxides as. necessary to establish curves similar to the samples in Figures
     2 and 3.

 7.   Establish the minimum excess 02 and the margin in excess 02 above the
     minimum that is required for fuel variations, load changes, and atmos-
     pheric conditions.  Reset the burner controls to maintain this lowest
     practical excess O  when operating on automatic.
                       £

 8.   Repeat steps 1 through 7 for each firing rate to be tested.

 9.   Verify that the new settings will be acceptable during sudden load
     changes that may occur.

10.   Perform a combustion efficiency and boiler efficiency spot-check.  Observe
     the boiler flame appearance and monitor for several weeks.  Repeat test
     sequence for alternate fuels.
                                       241

-------
MAINTAINING HIGH BOILER EFFICIENCY AND LOW NO
                                             x
        Once the optimum boiler firing mode has been implemented, there are
several steps that should be taken to assure that high efficiencies and low
NO  emissions are maintained in day-to-day boiler operation.  A periodic
  X
combustion efficiency spot-check is recommended to identify efficiency-related
problems requiring maintenance or repairs before the fuel wastage becomes
large.  Over the longer term, a boiler performance monitoring log of stack
temperature, fuel flow, air flow, steam flow, and unburned combustion indi-
cates such as smoke or CO is desirable.
        A preventive maintenance program is strongly encouraged in this era
of rapidly escalating fuel prices.  This fuel savings benefit is in addition
to primary considerations of safety and boiler reliability.  A periodic re-
view of the recommended inspection items is suggested.
CONCLUSIONS
        The implementation of a low-excess-air firing mode can result in a
significant reduction in nitric oxide emissions while also providing an im-
provement in boiler efficiency and fuel use for most industrial boilers.  A
test procedure for implementing this mode of operation has been prepared
covering topics concerning boiler test preparation, adjustment procedures,
and post-test efficiency maintenance.
        Inefficient boiler operation costs fuel and dollars in that a 2%
efficiency degradation on a 100,000 Ib/hr steam flow  (approximately 29 x 10
watts) boiler results in a $38,000 increase in annual fuel cost.
        Boiler operators should not attempt to implement a modified combustion
mode on their boiler from this paper alone.  A copy of the EPA/FEA report
 (No. EPA-600/8-77-003a) should be read thoroughly and in conjunction with the
specific boiler's operating and maintenance manuals before attempting any
boiler adjustments.  The "Guidelines for Industrial Boiler Performance Improve-
ment" referenced above can be obtained through the National Technical Informa-
tion Service, Springfield, VA  22161, order number PB 264 543.
                                      242

-------
                                  SECTION 6
                       BOILER DESIGN GUIDELINE MANUAL

        A guideline manual directed to boiler manufacturers is being prepared
as a companion  manual to the operational guidelines discussed in the previous
section.  This manual concentrates on boiler modification techniques that go
beyond the optimization of excess air that is the main thesis of the operational
guidelines.  Drafts of the design guidelines have been submitted to EPA and
to boiler manufacturers for comments and the finalized version should be avail-
able later this year.
        The design guideline is basically a summary of the combustion modifica-
tions evaluated in Phases I and II of the EPA industrial boiler field test
program (Refs. 1 and 2).  However, additional data has been included from
other studies to augment the discussions.  Injection of ammonia for gas phase
decomposition of NO  is an important NO  control process that was not evaluated
                   X                   X
in the EPA industrial boiler program but is discussed in the design guideline.
This process, called "Thermal deNO ", was first demonstrated commercially in
Japan in 1974 achieving up to 70% NO  reduction and has been applied to
                                    X
several boilers in that country.  Interest in the process is growing in the
U.S. (Ref. 11).
        The cost of combustion modifications is an important aspect in the
selection of the most appropriate method.  The actual cost of each modifica-
tion is, of course, dependent on the specific boiler design.  However some
order of magnitude estimates are provided in the manual as summarized on Table
VI  showing  the cost/reduction ratio for seven control options.  For small
boilers staged combustion, ammonia injection and reburnering have the lowest
cost per unit of NO  reduced and are about equally cost effective.  As boiler
                   X
size increases all options become less costly per unit of NO  reduced.
                                       243

-------
TABLE VI.    COSTS OF NO  CONTROL BY COMBUSTION MODIFICATION
               (Units of $/Ton NO  Prevented)
   Firing Rate
       Million Btu/Hr
10
100
1000
Control ^---^^
Option -*^^

Low Excess Air (-10%)
Staged Combustion (25%)
Flue Gas Recirculation (20%)
Reduced Air Preheat (-100 °F)

Water Injection (1 Ib/lb)

Ammonia Injection (90%)
Reburnering - Low NO

•XvX-XvX •.•.-.•.•.-.•.•. -.vj
ox:::x::1800:i>::£:$ 300 -300
1000 300 100
x??:?x-------™o:-:-:?vTl
v'xx':x'. 2300x:xXx:l 900 700
::x:x:xyx-xv.:.:.x.::::xv:: w;r!T:™-?rT v.: 7TJ
x;::x;:;:;i65G:;:;x'::;x;:"; ':: :1580:: : ]:l 1500
;x;:::x:::::;:x:;:::x:v:::v::::: XxXxv": •'h?.-'^.— .——•??
'$£??:• 2506j:':x;:>x'::: 2200:: :WS:x":v':.2QQO. x^:
^•:-x1:-:-^;:-:vX:xv.v:v;- ^ xivr-x^x-^ -J^iitii^^ii:^^
1000 390 370
800-1000 450-550 200-300






':T
<-\
•d


                              244

-------
                                  SECTION  7

                                 REFERENCES
 1.     Cato,  G.  A.,  et al. ,  "Field  Testing:  Application of  Combustion
       Modifications to Control  Pollutant Emissions  from Industrial
       Boilers  - Phase I," EPA 650/2-74-078-a, NTIS  No. PB 238  920/AS,
       October  1974.

 2.     Cato,  G.  A.,  et al.,  "Field  Testing:  Application of  Combustion
       Modifications to Control  Pollutant Emissions  from Industrial
       Boilers  - Phase II,"  EPA  600/2-76-086-a,  April  1976.

 3.     Cato,  G.  A.,  "Field Testing:  Trace Element and Organic  Emissions
       from Industrial Boilers," EPA 600/2-76-086-b, NTIS No. PB 261  263,
       October  1976.

 4.     McElroy,  M. W.  and Shore, D. E.,  "Guidelines  for Industrial Boiler
       Performance Improvement," EPA 600/8-77-003-a, NTIS No. PB 264  543/OWP,
       January  1977.

 5.     	, "Application of Combustion Modifications
       to Control Pollutant  Emissions from Industrial  Boilers—Boiler Design
       Guideline Manual," Draft  report,  KVB  6001-49, submitted  to EPA
       April 1977.

 6.     Hunter,  S. C. and Buening, H. J., "Field  Testing:  Application of
       Combustion Modifications  to  Control Pollutant Emissions  from
       Industrial Boilers—Phases I and II (Data Supplement),"   Draft
       report,  KVB 6001-491, submitted to EPA  May 1977.

 7.     Cato, G.  A.,  Muzio, L. J., and Hall,  R. E., "Influence of Combustion
       Modifications on Pollutant Emissions  from Industrial  Boilers,"
       Proceedings of the Stationary Source  Combustion Symposium, Vol.  Ill,
       p IV-163, EPA 600/2-76-152c, June 1976.

 8.     Cato, G.  A.,  Hall, R. E., and Muzio,  L. J., "Reduction of Pollutant
       Emissions from Industrial Boilers by  Combustion Modification,"
       ASME Paper 76-WA/Fu-5, December 1976.

 9.     Cowherd, C. Jr., and  Spigarelli, J.,  "Hazardous Emission Characteriza-
       tion of Utility Boilers," Interim Report  on EPA Contract No.  68-02-0228,
       Task No. 39,  Midwest  Research Institute,  Kansas City, MO, June 1974.
                                                                     »
10.     "Assessment of the Potential for Energy Conservation through  Improved
       Industrial Boiler Efficiency," Final Report - Vol.  I, U.S. Federal Energy
       Administration Contract No.  C-04-50085-00, NTIS No.  PB  262 576,  October
       1976.

11.     "A Way to Lower NOX in Utility Boilers,"  Environmental  Science and   m
       Technology, p. 226, March 1977.
                                      245

-------

-------
               FIELD TESTS

     STOKER FIRED INDUSTRIAL  BOILERS
           AS SPECIFIED BY ERDA
                   By:
              B.  C. Severs
American Boiler Manufacturers Association
           Arlington, VA  22209
                      247

-------

-------
                                  ABSTRACT

     This paper represents a field testing program on stoker coal-fired indus-
trial boilers.  The objective is to prepare design and application guidelines
for industrial stoker firing.  Among other objectives, is increased utiliza-
tion of stoker coal-fired industrial boilers compatible with emissions and en-
vironmental requirements.

     Selected site locations of the program use current stoker-boiler design
arrangements.  The capacity selection of equipment is from 75,000 Ib/hr to
250,000 Ib/hr steam flow.

     The results of the program will encourage stoker and boiler technology to
meet market needs and emission standards utilizing coal as a fuel and incorp-
orating existing, new, and/or improved air pollution control equipment.
                                    249

-------

-------
                          STATEMENT OF WORK
I.  INTRODUCTION AND OBJECTIVES

    A.   Introduction

        In recent years a heavy preponderance of industrial boiler installa-
        tions has been shop-assembled gas- and oil-fired units purchased and
        installed at substantially lower costs than conventional coal burn-
        ing, boiler-stoker equipment.  Because of this decline in coal-firing,
        little or no work has been done to improve specification data and in-
        formation for consulting engineers and purchasers of coal-burning
        equipment.  The current implementation of more rigid air pollution
        regulations has made it difficult for many coal-burning installations
        to comply with required stack emission limits, thus creating a further
        negative influence on coal-firing.

        The market for coal suitable to be fired in industrial boilers, as
        reflected by sales data, is being held back -by critical uncertainties
        in the environmental and energy areas, causing potential customers of
        coal-fired Industrial boilers to shelve plans for capital expansion
        and conversion.  This has caused a serious reduction in the number of
        installations of new industrial coal-fired units.  It is highly desir-
        able to remove these uncertainties and thereby establishing confidence
        among industrial users to order and install stoker coal-fired boilers.
        This will lead to significantly increased coal usage.

        This Test Program is an industry first where the ABMA and associated
        industries and the Federal Agencies have agreed upon a combined
        achievement to encourage design and application of stoker coal-fired
        Industrial Boilers.

    B.   Objectives

        The objectives of this Program are:

        1.  To prepare a comprehensive document, Design and Application
            Guidelines for Industrial Stoker Firing.

        2.  To advance boiler and stoker technology by testing various boiler
                                   251

-------
Statement of Work
3.
            furnaces and stoker designs for response to changes In coal analy-
            sis and sizing,  degree of fly-ash reinjection overfire air ad-
            mission, ash handling and pollutants emitted.

            To refine applications of existing pollution control equipment and
            to more closely control stack emissions under varied operating
            conditions through more accurate boiler outlet dust loading data.
        4.  To contribute to the design of new and improved air pollution con-
            trol equipment.

        5.  To facilitate preparation of intelligent and reasonable national
            emissions standards for small coal-fired boilers by the Environ-
            mental Protection Agency.

        6.  To facilitate planning for coal supply contracts by users of the
            boiler/stoker equipment by developing reasonable emission regula-
            tions .

        7.  To promote the increased utilization of stoker coal-fired boilers
            by U.S. industry by insuring the compatibility of these units
            emissions with appropriate environmental requirements.
II.  SCOPE OF WORK

     The American Boiler Manufacturers Association (ABMA), the Contractor,
     shall be responsible for the overall program.  The Program Director
     shall be the Executive Director of the ABMA.  Full-time administrative
     and technical program directions shall be provided by a Project Manager,
     selected by the ABMA, acceptable to ERDA and EPA and responsible to the
     ABMA.  General program guidance shall be provided by the ABMA Stoker
     Technical Committee which shall include the ABMA Program Director, and
     the ERDA and EPA Program Managers.

     The Contractor shall be responsible for all engineering, materials,
     equipment, services, personnel, facilities and administrative support
     for the accomplishment of the specific tasks described below.  The
     source of these capabilities shall include the ABMA, a selected sub-
     contractor, boiler site personnel, equipment supplier engineers and
     instrumentation and control technical specialists.  Where possible,
     existing facilities such as laboratories, shops, offices and data
     acquisition and processing, shall be made available for use.  The
     ABMA and its Stoker Technical Committee have chosen KVB, Inc., Minneap-
     olis, Minnesota, to be the sub-contractor for site monitoring and meas-
     uring.  The sub-contractor selection was made on a competitive technical
     evaluation basis.
                                    252

-------
Statement of Work

     A.   Task 1 - Boiler/Stoker Selection

         The Contractor shall select a total of six (6)  boiler/stoker com-
         binations for testing and evaluation.   The selection shall be rep-
         resentative of the spectrum of industrial boilers in a steam capac-
         ity range of 75,000-250,000 Ib/hr.   The selection criteria shall
         follow the guidelines provided by the  Stoker Technical Committee.
         All selections shall be reviewed and approved by the Stoker Tech-
         nical Committee and the ERDA and EPA Program Managers.

     B.   Task 2 - Coal Selection

         The Contractor shall select representative coals for use In the
         selected boiler-stoker combinations.  The selection process shall
         involve the Contractor, ABMA member firms, regionally and nationally
         situated coal producers and suppliers, and applicable technical
         groups, such as the National Coal Association.   The selection cri-
         teria shall include current and projected availability, geographical
         occurrence, accessibility to the selected boiler/stoker combinations
         and coal properties and analyses.  All ranks of coal shall be con-
         sidered, including lignite and anthracite.

         The coal selections shall be reviewed  and approved by the Stoker
         Technical Committee.

     c*   Task 3 - Test Program Development

         The Contractor shall develop test programs for the six (6) selected
         boiler-stoker combinations burning the selected coals. The test pro-
         gram shall include test duration, instrumentation needs, test pro-
         cedures, overall schedules and sampling requirements.

         The test program shall be reviewed and approved by the Stoker Tech-
         nical Committee.

     D.   Task 4 - Site Preparation

         The Contractor shall prepare working drawings for each test unit to
         show location and design of necessary instrumentation and of sampl-
         ing connections.  The Contractor shall also design and produce fab-
         rication drawings for any necessary platforms and ladders to insure
         safe access to sampling connections and for adequate instrumentation
         placement.  All surveys, design information and drawings shall be
         submitted to the Stoker Technical Committee and the Owner for check-
         ing and approval.
                                     253

-------
Statement of Work

         Upon authorization to proceed by the Owner,  the Contractor shall
         fabricate and install the necessary access and instrumentation re-
         quired for the specific site.

     E.  Task 5 - Boiler/Stoker Testing

         The Contractor shall conduct the test program, as approved, on the
         selected six (6) boiler/stoker combinations.  Of the selected units,
         one shall be selected for initial testing;  all others will await
         the evaluation of the techniques and procedures employed at that
         site.  Following any modifications considered advisable from the
         first test, the Contractor shall perform tests on the remaining
         five (5) boiler/stoker combinations.

         Personnel involved in the test program shall include the ABMA Pro-
         ject Manager, the sub-contractor, the Owner and the applicable serv-
         ice engineers to insure that the facility is operating properly for
         valid test results.
     F.  Task 6 - Documentation

         The Contractor shall implement, maintain, and control a system for
         the identification, preparation, reproduction, distribution, and
         maintenance of all documentation, data and information necessary for
         the management of this project.

         Documentation and reports shall include, but not be limited to, the
         following:

           1.  Periodic Reports
               a.  Work plan
               b.  Monthly
               c.  Quarterly (in lieu of third monthly)
               d.  Annual (in lieu of fourth quarterly)
               e.  Final Report in two volumes
                     Vol. 1   -  Engineering Documentation
                     Vol. II  -  Data Supplement
           2.  Interim Reports

               a.  Task 1 Selections
               b.  Task 2 Selections
               c.  Task 3 Test Program
               d.  Task 4 Site Preparation
               e.  Task 5 Testing Data
               f.  Design and Application
                   Guidelines for Industrial
                   Stoker Firing
                                    254
Due date, after program start
        months
        months
        months
        months
      including in quarterly
      which follows completion
      of tests
      24 months

-------
                               TECHNICAL REPORT DATA
                         {Please read Instructions on the reverse before completing}
1. REPORT NO.
 EPA-600/7-77-073a
2.
                           3. RECIPIENTS ACCESSION NO.
4. TITLE AND SUBTITLE PROCEEDINGS OF THE SECOND
STATIONARY SOURCE COMBUSTION SYMPOSIUM
Volume I. Small Industrial, Commercial, and
Residential Systems
                           6. REPORT DATE
                           July 1977
                           G. PERFORMING ORGANIZATION COOi
7. AUTHOR(S) .
         Symposium Chairman J.S. Bowen, Vice-
Chairman R.E. Hall
                                                     8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
                                                     10. PROGRAM ELEMENT NO.
NA
                                                      EHE624
                                                     11. CONTRACT/GRANT NO.
                                                      NA (Inhouse)
12. SPONSORING AGENCY NAME AND ADDRESS
 EPA, Office of Research and Development
 Industrial Environmental Research Laboratory
 Research Triangle Park, NC  27711
                           13. TYPE OF REPORT AND PERIOD COVERED
                           Proceedings: 8/29-10/1/77
                           14. SPONSORING AGENCY CODE
                            EPA/600/13
15. SUPPLEMENTARY NOTES TERL_RTp project officer for these proceedings is R.E. Hall,
Mail Drop 65, 919/541-2477.
is. ABSTRACT Tjie proceedings document the 50 presentations made during the Second
 Stationary Source Combustion Symposium held in New Orleans, LA, August 29-
 September 1,  1977.  Sponsored by" the Combustion Research Branch of EPA's Indus-
 trial Environmental Research Laboratory—RTF, the symposium dealt with subjects
 relating both to developing improved combustion technology for the reduction of air
 pollutant emissions from stationary sources,  and to improving equipment efficiency.
 The symposium was divided into six parts,  and the proceedings were issued in five
 volumes: Volume I—Small Industrial,  Commercial, and Residential Systems; Volume
 II--Utility and Large Industrial Boilers; Volume HI—Stationary Engine, Industrial
 Process Combustion Systems, and Advanced Processes; Volume  IV—Fundamental
 Combustion Research; and Volume V--Addendum.  The symposium was intended to
 provide contractor, industrial, and Government representatives with the latest infor-
 mation on EPA inhouse and contract combustion research projects related to
 pollution control, with emphasis on reducing nitrogen oxides while controlling other
 emissions and improving efficiency.
17.
                             KiY WORDS AND DOCUMENT ANALYSIS
                DESCRIPTORS
                                         b.lOENTIFIERS/OPEN ENDED TERMS
                                        ',. COSATI Field/Group
 Air Pollution, Combustion, Field Tests
 Combustion Control, Coal, Oils
 Natural Gas,  Nitrogen Oxides, Carbon
 Carbon Monoxide, Hydrocarbons, Boilers
 Pulverized Fuels, Fossil Fuels, Utilities
 Gas Turbines, Efficiency
                Air Pollution Control
                Stationary Sources
                Combustion Modification
                Unburned Hydrocarbons
                Fundamental Research
                Fuel Nitrogen
                Burner Tests
13B
2 IB  14B
21D  11H
07B
07C  13A
13G  14A
18. DISTRIBUTION STATEMENT
 Unlimited
                                          19. SECURITY CLASS (TMi Report)
                                          Unclassified
                                        21. NO. OF PAGES
                                             260
               20. SECURITY CLASS ITMspage)
                Unclassified
                                       22. PRICE
EPA Form 2220-1 (9-73)
            255

-------

-------
 EP 600/7  EPA
\77-Q73a   Ind.  Env.  Res. Lab.
 AUTHOR
      Proc. of the second stationary
 T|TLE source combustion symposium.
      V.l:  Small industrial, commer-
           &
 OAVLORO 48

-------
DATE  DUE
                                            BORROWER 5  NAME

-------
DATE DUE

-------

-------