950R80044
U.S. Environmental Electric Power IERL-RTP-1087
Protection Agency Research Institute October 1980
[Proceedings of the Joint
symposium on Stationary
Combustion NOx Control
Volume V
Addendum
SEPA A EPRI
SER^CjEPM
&baISeph
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine 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
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the MISCELLANEOUS
REPORTS series. This series is reserved for reports whose
content does not fit into one of the other specific series.
Conference proceedings, annual reports, and bibliographies
are examples of miscellaneous reports.
EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental
Protection Agency, and approved for publication. Approval
does not signify that the contents necessarily reflect the
views and policy of the Agency, nor does mention of trade
names or commercial products constitute endorsement or
recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
IERL-RTP-1087
October 1980
Proceedings of the Joint
Symposium on Stationary
Combustion NOx Control
Volume V
Addendum
Symposium Cochairmen
Robert E. Hall, EPA
and
J. Edward Cichanowicz, EPRI
Program Element No. N130
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, D.C. 20460
and
ELECTRIC POWER RESEARCH INSTITUTE
3412 Hillview Avenue
Palo Alto, California 94303
-------
PREFACE
These proceedings document more than 50 presentations given at the
Joint Symposium on Stationary Combustion NOx Control held October 6-9,
1980 at the Stouffer's Denver Inn in Denver, Colorado. The symposium was
sponsored by the Combustion Research Branch of the EPA's Industrial
Environmental Research Laboratory-Research Triangle Park and the Electric
Power Research Institute (EPRI). The presentations emphasized recent
developments in N0X control technology. Cochairmen of the symposium
were Robert E. Hall, EPA, and J. Edward Cichanowicz, EPRI. Introductory
remarks were made by Dan V. Giovanni, Program Manager for Air Quality
Control, Coal Combustion Systems Division, EPRI, and the welcoming address
was given by Roger L. Williams, Regional Administrator, EPA Region VIII.
Stephen J. Gage, Assistant Administrator for Research and Development,
EPA, was the keynote speaker. The symposium had 11 sessions:
I: N0X Emissions Issues
Michael J. Miller, EPRI, Session Chairman
II: Manufacturers Update of Commercially Available Combustion
Technology
Joshua S. Bowen, EPA, Session Chairman
III: N0X Emissions Characterization of Full Scale Utility
Powerplants
David G. Lachapelle, EPA, Session Chairman
IV: Low N0X Combustion Development
Michael W. McElroy, EPRI, Session Chairman
Vas Postcombustion NOx Control
George P. Green, Public Service Company of Colorado,
Session Chairman
Vb: Fundamental Combustion Research
Tom W. Lester, EPA, Session Chairman
VI: Status of Flue Gas Treatment for Coal-Fired Boilers
Dan V. Giovanni, EPRI, Session Chairman
VII: Large Industrial Boilers
J. David Mobley, EPA, Session Chairman
VIII: Small Industrial, Commercial, and Residential Systems
J. David Mobley, EPA, Session Chairman
IX: Environmental Assessment
Robert P. Hangebrauck, EPA, Session Chairman
X: Stationary Engines and Industrial Process Combustion Systems
Robert E. Hall, EPA, Session Chairman
XI: Advanced Processes
G. Blair Martin, EPA, Session Chairman
ii
-------
VOLUME V
TABLE OF CONTENTS
Page
Introductory Remarks, D. V. Giovanni 1
Welcoming Address, R. L. Williams 2
Keynote Address, S. J. Gage 5
Luncheon Address, L. M. Henry 18
"Regulatory Pressures for Increased N0X Controls,"
R. E. Wyzga 24
"Development and Revision of Air Quality Standards with
Special Attention to the NO2 Standard Review,"
M. H. Jones 36
"Acid Rain Issues," R. A. Luken 55
"State of California Perspective on N0X Control for
Stationary Sources," A. Goodley 57
"Fossil Steam Generator N0X Control Update,"
J. A. Barsin 69
"Current Developments in Low N0X Firing Systems,"
T. Kawamura and D. J. Frey 94
"An Evaluation of N0X Emissions From Coal-Fired Steam
Generators," R. A. Lisauskas and J. J. Marshall 130
"N0X Emissions Characteristics of Arch-Fired Furnaces,"
T. W. Sonnichsen and J. E. Cichanowicz 152
"Relationship Between N0X and Fine Particle Emissions,"
M. W. McElroy and R. C. Carr 183
"Commercial Evaluation of a Low N0X Combustion System
as Applied to Coal-Fired Utility Boilers," S. A. Johnson
and T. M. Soraner 200
"The Development of Distributed Mixing Pulverized
Coal Burners," D. P. Rees, J. Lee, A. R. Brienza,
and M. P. Heap 249
"Japanese Technical Development for Combustion N0X
Control," K. Mouri and Y. Nakabayashi 274
iii
-------
Page
"Empirical Evaluation of Selective Catalytic Reduction
as an N0X Control Technique," J. E. Cichanowicz
and D. V. Giovanni 322
"Development of Flue Gas Treatment in Japan,"
Y. Nakabayashi, H. Yugami, and K. Mouri 350
"Treating Flue Gas from Coal-Fired Boilers for N0X
Reduction with the Shell Flue Gas Treating Process,"
J. B. Pohlenz and A. 0. Braun 400
"The Development of a Catalytic NOx Reduction
System for Coal-Fired Steam Generators," T.
Sengoku, Y. Todo, N. Yokoyama, and B. M. Howell 412
"Applicability of Thermal DeNOx to Large Industrial
Boilers," B. E. Hurst and C. E. Schleckser, Jr 441
"Utility Boiler Environmental Assessment — the EPRI
Approach," M. W. Zengerle 471
"Single-Cylinder Tests of Emission Control Methods for
Large-Bore Stationary Engines," R. P. Wilson, Jr 486
"Emission Reduction by Combustion Modification for
Petroleum Process Heaters," R. J. Tidona, W. A. Carter,
J. R. Hart, and S. C. Hunter 536
Addendum to "Subscale Tests of Combustion Modification
for Steel Furnaces," R. L. Tidona, W. A. Carter, and
S. C. Hunter 563
List of Attendees 569
iv
-------
INTRODUCTORY REMARKS
Dan V. Giovanni, Manager
Air Quality Control Program, EPRI
It is my pleasure today to welcome you on behalf of EPRI to the Joint Symposium
on Stationary Combustion NOx Control. This is the third in a series of
symposia with which EPRI has been associated and it may prove to be the
most important.
NOx emissions from stationary sources have been controlled to date through
manipulation of the combustion process using conventional furnace and
burner designs. To achieve significantly lower NOx levels it will be
necessary to employ heretofore unconventional combustion systems or post-
combustion treatment processes. New technologies such as these inherently
increase the risks undertaken by manufacturers and operators of stationary
combustion equipment. Capital and operating costs will increase and overall
system reliability may be jeopardized.
To help offset these risks, successful research and development programs
are necessary. During the next few days we will be hearing and discussing
results from ongoing EPRI and EPA projects, as well as, many private pro-
grams undertaken in the U.S., Japan, and Europe. I encourage your active
participation in the symposium. Thank you.
1
-------
WELCOMING ADDRESS
Roger L. Williams, Regional Administrator, VIII
U.S. Environmental Protection Agency
Good Morning. It is my pleasure to welcome you to Denver and to this
symposium on nitrogen oxides control. The joint sponsorship of this symposium
by the Electric Power Research Institute and the Environmental Protection
Agency provides the clear message of the need and desire for a close working
relationship between energy and environmental interests. Later this week
another workshop will be held here in Denver that will have a broader theme,
designed to assist the Energy Community in understanding the many programs
and authorities administered by the EPA. This seminar beginning on Wednesday
is titled, "Conference on Environmental Regulations Relating to Energy."
Hardly a week goes by that there isn't a major energy conference or
workshop here in Denver. Estimates indicate that there are more than 600
energy companies now residing in Denver. Denver is rapidly becoming known
as the "Energy Capital" of the West. The development of Western United
States vast energy resources is a vital ingredient in the National Energy
Plan. EPA is committed to doing its part in support of the Nation's goal
of energy self-sufficiency. These cooperative workshops are an important
part toward that end.
Conversion of oil and gas fired utility boilers to coal is an important
part of the energy supply equation and should proceed expeditiously. Almost
3 million barrels per day of oil equivalent could be displaced. Conversion
of industrial boilers where possible should also proceed. EPA's role under
the Power Plant and Industrial Fuel Act of 1978 and the Clean Air Act
Amendments of 1977 is one of assuring that these conversions occur in an
environmentally acceptable manner. Continued research into cost effective
methods to reduce NOx emissions from stationary source combustion, such as
papers presented at this symposium are a key to the environmental acceptance
of these conversions.
2
-------
All new steam electric power plants constructed in the United States
will be coal fired or nuclear. The abundance and availability of cheap
surface mineable coal in the West has provided an attraction for the utility
industry. This attraction has resulted in a substantial increase in EPA
Region VIII Office activities in the form of EIS reviews, issuance of
surface water discharge permits, and air quality (PSD) permits to construct
and operate. Since January 1, 1978 we have issued more than 308 energy
related permits. We are presently averaging one Regional energy related
action per working day and this pace is projected for the next five years.
Most permitting applications have been approved with reasonable environ-
mental safeguards, a few permits have been denied. Since the PSD concept
was adopted in the Clean Air Act Amendments of 1977, my Office has received
applications for and issued PSD permits to eleven power plants with a total
combined generating capacity of almost 10,000 Megawatts. This will raise
the Rocky Mountain Regions generating capacity to about 29,000 MW. I
believe that this indicates that while environmental standards impose tough
requirements in the form of new source performance standards, best available
control technology, and PSD ambient air quality increments, the program
cannot be labeled as a major inhibiting force against energy development
as some people have chosen to do. Appropriate siting combined with good
controls result in adequate energy development coupled with proper environ-
mental protection.
The protection of environmentally sensitive areas must share an equal place
at the table along with the development of energy resources. Region VIII
states have been blessed with a dichotomy. While we have abundant energy
resources -- 50% of the Nation's coal and uranium resources and essentially
all of the oil shale and tar sands resources -- we also have the cleanest air
and water in the Nation, 10 million acres of Class I National Parks and
3
-------
Wilderness areas, wild and scenic rivers, unique wetlands, prime
agricultural land, endangered species and habitat and some of the Nation's
clearest visibility. It is heartening to see increasing emphasis on
energy conservation measures and on the development of renewable energy
resources. It is also encouraging to note that electric power demand
increased by only 2.1 percent in 1979 over 1978. That is a major
improvement over the industry's historic 7 percent annual growth rate.
It is clear from a review of your program and the experience assembled
in this room, that this conference will be a major success and have a
significant impact in the field of combustion research and environmental
control.
We are very pleased to have the opportunity to co-sponsor this
seminar with the Electric Power Research Institute here in Denver and
look forward to the continued cooperation between your Industry and
EPA towards a better life.
Welcome and I wish you much success for your seminar.
Thank you.
4
-------
SOLVING THE NOx CONTROL DILEMMA
(Keynote Address, October 6, 1980)
Dr. Stephen J. Gage
Assistant Administrator for Research and Development
U.S. Environmental Protection Agency
To someone who is inclined to believe that the Federal Government
spends its time searching for v/ays to make life difficult for the private
sector, it would be easy to assume that we get a perverse delight out of
instructing industry to burn more coal, while insisting at the same time
that air quality be maintained. Such instructions sound almost like a
Zen koan...one of those paradoxical word puzzles a Zen Master will use
to confound his students and spur them toward enlightenment. "Without
getting your robe wet, remove a stone from the bottom of the sea."
"Without moving a muscle, stop the clanging of the bell across the
river."
There's no getting around the fact that coal is a dirty fuel,
contaminated with a variety of substances found to be harmful to human
health and the environment. How on earth can we possibly triple the
burning of coal by the year 2000, yet keep our air breathable?
We in EPA are aware of the dilemma industry faces. Far from
sitting back with an inscrutable smile as you scramble to find solutions,
we consider your problem our own. We are not in the business of assigning
koans...that job wi 1-1 be left to the Zen Masters. But in the past
decade, we have become enlightened about one thing: achievement of
these twin goals ijs possible. ..at a reasonable cost...through the
development of effective control technologies. Environmental degradation
need not follow on the heels of a dramatic increase in coal use. The
nation can have economic growth, achieve energy independence, and at the
5
-------
same time maintain a high quality of life that so closely depends on the
quality of our environment. Working together with industry, EPA has
made and is continuing to make considerable progress in developing cost-
effective ways of controlling coal's pollutants.
One success story has been sulfur oxide control, thanks to the
development of flue gas desulfurization techniques. Although the first
commercial use of FGD control systems took place in the early 1930's,
the technology is only now becoming widely used, after ten years of
extensive research, development and demonstration programs sponsored by
EPA, the Tennessee Valley Authority, and the Electric Power Research
Institute. As a result of these programs, FGD systems can now remove up
to 90 percent of S0£ emissions reliably and economically. Industry and
utility companies, which in the early days of FGD development viewed the
technology with a skeptical eye, are now adopting the process. At last
count, 73 FGD units were in operation, with 127 units in design or under
construction. Once all these units are operating, over a quarter of the
current total U.S. coal-fired capacity will be equipped with FGD.
Because of this growing acceptance of FGD, the total amount of sulfur
oxides emitted to the atmosphere will remain constant or even decrease
slightly from 1975 to 2000. This represents a tremendous victory for
pollution control technology.
The history of particulates control is similarly cheering. Although
the total amount of particulates generated between 1975 and 2000 is
projected to double because of increased coal use, particulate emissions
from coal combustion by electric utlities and industrial boilers will
decline. This decline will be made possible by the widespread use of
6
-------
modern, high-efficiency electrostatic precipitators. In the future,
more advanced ESP1s will likely attain the same level of performance at
half the cost. And, while there is still work to be done in improving
removal methods for fine particles and the difficult-to-remove ash
typically produced from Western coals, the already established technologies
of wet scrubbing and baghouses may provide the answers we're looking
for. The bulk of the particulates problem has been licked.
What I mainly want to talk to you about today is a third air
pollutant from coal combustion that may present us with our biggest air
pollution challenge during the 1980's and beyond: nitrogen oxides.
While sulfur oxide emissions are expected to decrease or remain constant
between now and the year 2000, and while particulate emissions will
decrease, nitrogen oxide emissions could easily double during this
period...unless more effective control methods are developed. At
present, about half the current N0X emissions come from stationary
sources, but by 1985, due to the trend toward greater combustion of
coal, stationary sources may be responsible for 70 percent. Of the
emissions from stationary sources, over half are contributed by utility
and large industrial boilers alone. These large boilers emit an estimated
6 million tons of N0X every year. Without controls, a single coal-fired
boiler may spout 120 tons of N0X into the atmosphere every day.
EPA is extremely concerned about the N0X trends in this country.
As a result, we have mobilized a significant portion of our research and
development machinery toward combatting such emissions. In each of the
last three years, through 1980, EPA has spent $12 million on developing
methods of controlling N0X pollution. In 1981 and 1982, funding will
7
-------
continue at this high level, for a total expenditure of $60 million over
the five-year period from 1978 to 1932. This level of funding should
leave no doubt of EPA's dedication to the NO control effort.
A
Until recently, the effects of N0X on humans and their environment
has not been sufficiently appreciated. At last year's conference on
EPA's Interagency Energy/Environment R&D Program, someone stood up and
asked why EPA was devoting so much time and money to N0X control. "N0X
makes your eyes burn," he said, "but sulfur oxides can kill you." Well,
as a matter of fact, we now know that FI0X can do a great deal more than
cause eye irritation. Long-term exposure to even low levels of N0X can
reduce resistance to respiratory infections such as bronchitis, pneumonia
and influenza, and high-level, short-term exposure can have a variety of
pronounced adverse health effects. The available evidence indicates
that human health can be impaired by exposure to nitrogen dioxide,
concentrations approaching or falling within the range of recorded
ambient air NO2 levels.
Also of great concern are the environmental effects of N0X once it
has been transformed into other pollutants in the atmosphere. N0X is a
precursor of photochemical oxidants...the air pollutants most damaging
to agriculture and forestry in the United States. Elevated concentrations
of ozone, which is the chief ingredient of smog, have become a regional
problem throughout the country, causing widespread damage to crops on
both coasts. Oxidants have caused harm to agricultural crops, forest,
and native vegetation in Southern California as well as widespread
damage to crops in the East, and have become a new stress on ecosystems
8
-------
in the Southwest.
Probably the most alarming environmental effect of N0X is its role
in the formation of acid rain, which is developing into what may be one
of the most serious environmental threats of the century. As you heard,
acid rain is formed when the gases of nitrogen and sulfur oxides combine
with water vapor molecules in the atmosphere and are transformed into
nitric and sulfuric acids. These acids are returned to earth by rain or
snow or in dry form, sometimes hundreds or even thousands of miles from
their sources, and can cause extensive ecological damage.
In New York's Adirondack Mountains, for example, acid rain has
killed all of the fish in half of the high-altitude lakes. In Norway,
the losses to salmon fisheries attributable to acid rain are estimated
to be in the tens of millions of dollars. Acid rain may also be playing
a part in the decline in forest growth observed in the Northeastern
United States. In the eastern part of the U.S. and in Canada, there are
extensive areas which are particularly susceptible to acidification
because of the lack of natural buffering capacity in soils and water.
In much of the West, the alkaline nature of the soils and lakes acts to
neutralize acid rain, so the effects are not as pronounced there. But
even in the West, ominous signs of vegetation damage have appeared.
N0X accounts for 30 to 50 percent of the acid rain problem in the
East, depending on the season. In the West, acid rain may be composed
of up to 90 percent nitric acids formed from N0X. Even as sulfur oxides
are brought under control, the jump in N0X emissions ensures that the
acid rain threat will remain with us.
As required by the Clean Air Act, EPA has set a National Ambient
Air Quality Standard for nitrogen dioxide, based on the level required
9
-------
to protect the public health and welfare. This standard is currently
pegged at 100 micrograms per cubic meter of air, or .05 parts per
million, on an annual average basis. Several of our largest cities,
including Los Angeles, Chicago, Denver, New York, and Salt Lake City,
are currently exceeding this standard. In fact, many industiralized
cities have short-term levels of nitrogen dioxide that reach several
times the NAAQS.
EPA is ready and willing to work with you in attaining the ambient
air and stationary source standards. A teamwork approach was successful
when we were confronted with the need to reduce sulfur oxides and
particulates. The same approach can work with N0X control. I would
also like to make a special point of welcoming our international friends
to join us in this cooperative effort. As one who has appreciated the
significant contributions made by the Jpaanese and, more recently, the
Germans in advancing FGD technology, I am encouraged by the active
participation of the highly capable engineers from Japan and West Germany
as well as from Canada and the European Common Market.
The N0X problem is, however, a tough nut to crack. It doesn't
yield to the solutions that worked so well for sulfur oxide and particulates
control. Physical coal cleaning, which can be used on some coal to
reduce sulfur and ash content, has no effect on coal's nitrogen content,
because the nitrogen is chemically bound to the coal. "Denitrogenation" --
that is, chemically removing nitrogen from coal -- is prohibitively
expensive at present, and at any rate does not address the problem of
thermal N0X, which, is formed by molecular reaction in superheated
10
-------
combustion air. Flue gas treatment for N0X control has been used with a
fair amount of success in Japan on oil-fired boilers, but there are
major financial and technical hurdles to applying that technology to
coal-fired units. Even the coming age of synthetic liquid fuels made
from coal, which may consume 120 million tons of coal in 1990 and
300 million tons in 2000, offers little hope for N0X control -- in fact,
the concentration of fuel nitrogen may be increased when coal is converted
to a liquid.
However, the picture is not nearly as bleak as it may at first
appear. There is an answer that is both cost-effective and energy-
efficient. By modifying the conditions under which combustion takes
place, an existing coal-fired power plant can reduce its N0X emissions
by 40 to 50 percent, which is generally enough of a reduction to meet
current New Source Performance Standards for utility and large industrial
boilers. When applied to new burner designs, combustion modification
may reduce N0X emissions by another two-thirds, yielding a total N0X
control of up to 85 percent. This means that N0X emissions from coal
burners can be reduced to the level found in oil burners. Yet, because
combustion modification involves changes only in burner design, the cost
is quite small -- less than one-half of one percent of the boiler cost.
And, because we are ensuring that the new burners are as efficient as
the older designs, the operating cost is nearly zero. Levels of other
pollutants, such as particulates and hydrocarbons, are also reduced,
because the total combustion process is optimized.
EPA's work in developing new low-NOx burner designs is the center-
piece of its N0X control effort. As early as 1971, EPA's Combustion
11
-------
Research Branch at Research Triangle Park in North Carolina was experimenting
with burners that produced a slower fuel-air mixing and a cooler flame
than the traditional turbulent diffusion flame. We discovered that by
suppressing flame temperatures and delaying the mixing of fuel and air,
both thermal and fuel N0X formation was hindered. Pilot scale work has
proven that the technology is sound -- N0X emissions were reduced to
below 200 ppm.
The next step is to evaluate the low-N0x burner's performance at as
close to practical size as possible, in order to encourage industry
acceptance of the technology. Two projects currently underway have this
aim in mind; they will allow full-scale field evaluation of the low-N0x
burner.
One project will employ two industrial boilers in the range of 30
to 150 thermal megawatts, while the other will involve two utility
boilers in the range of 100 to 300 megawatts. For each boiler, a
prototype burner will first be constructed and then tested in an experimental
facility. After the uncontrolled emissions of each host boiler has been
measured, the low-MOx burners will be installed and adjusted. Then, EPA
will conduct a long-term evaluation of the burners, including environmental
assessment and analysis of boiler operation. This will include corrosion
testing, since slagging and fouling from coal ash is always a worry when
burner design is altered. Finally, a guideline manual will be prepared
to explain the technology.
The goal of these projects, which will be completed in late 1982,
is not only to show that the new burners do indeed reduce N0X emissions,
but, that in doing so, they also equal or improve the thermal efficiency
12
-------
of conventional boilers...that carbon emissions are not increased...and,
that corrosion is prevented. One encouraging observation has been that
increasing the scale of the low-NOx burner toward practical size appears
to make the technology more effective.
To aid EPA in the evaluation of these projects, and also to help
keep the projects on the track toward commercialization, a technical
review panel composed of representatives of boiler manufacturers,
utilities, and research organizations will be on hand to offer advice
and criticism. To provide a broader perspective on potential users,
there will also be a technology transfer panel consisting of government
agency representatives and trade association members.
The low-NOx burner is being tested with a wide variety of U.S.
coals. Initial tests indicate that it is effective even with high-
nitrogen-content coals. Experiments with burning high-nitrogen residual
fuel oils using low-NOx technology provide both a worst-case scenario
for conventional oils and a preview of what may be achievable with high-
nitrogen coal and shale-derived oils. At pilot scale, the low-NOx
burner reduced N0X emissions from these residual oils by 50 to 75 percent
of what they would be if burned in a conventional boiler.
The results of all of EPA's tests with this new technology have met
or exceeded program goals, so we feel comfortably optimistic about the
kind of performance we will see when it is adopted commercially. In
fact, we're very much encouraged by the results being obtained on a 700-
mwe boiler in Vlest Germany, an experiment which you will be hearing
13
-------
about tomorrow morning.
What we've learned in applying combustion modification techniques
and innovative burner designs to utility and large industrial boilers is
also being applied to other combustion sources. These include small-
scale industrial boilers, gas turbine engines, and residential oil
furnaces. A new oil-fired furnace design we've been working on not only
reduces N0X by 65 percent, but reduces oil consumption an average of
15 percent. These are the kind of results we really enjoy seeing:
increases in efficiency as well as improved pollution control.
A particularly exciting new control technology, which can be
retrofitted to many existing coal-fired boilers with only minor
modifications and which reduces sulfur oxide emissions as well as N0X,
is the limestone injection/multi-stage burner, or LIMB for short. The
LIMB technology may be able to remove 50 to 70 percent of sulfur oxides
at the same time that it reduces NO by 50 to 80 percent. And it can
A
accomplish this at a cost for SO2 control equipment of only $30 to $40
per kilowatt, as opposed to the average of $150 per kilowatt that wet
scrubbing requires. Although the LIMB has only reached the bench/pilot
scale stage of development here in the U.S., Germany is currently
operating a 60 megawatt electric boiler using the technology, so we know
that it works on a larger scale.
The idea of combining limestone injection for S0£ control with a
low-N0x burner is not a new one. In 1967, U0P, building on earlier
limestone injection experiments by Combustion Engineering, injected
-------
limestone into an arch-fired burner, which is a naturally low-NOv
A
burner. SO2 emissions were reduced by 50 percent at a stochiometric
ratio of 1.3.
The 60 megawatt prototype limestone injection boiler in Germany,
which I mentioned earlier, has been operating for one year. It fires
West German lignite, and utilizes flue gas recirculation to minimize
peak temperature and N0X formation. At present, it is achieving 50 to
90 percent SO2 removal at stochiometric ratios of 2.5 to 5.0. Retrofit
capital costs for this technology are only $3.00 per kilowatt.
EPA has proposed a five-year research, development, and demonstration
program that will bring the LIMB technology up to commercial scale. In
the first year, EPA will characterize reactions and furnace conditions;
evaluate impacts on furnace operation; and test the technology with a
wide range of coal types and calcium-based sorbents. Next will come a
year of field evaluation, in which EPA goals will be to demonstrate
sulfur removal efficiency, optimize performance'variables, determine if
there are any adverse boiler side effects such as slagging, plugging and
corrosion, and obtain design and cost data. Both wall-fired and
tangentially-fired units will undergo testing. Another year will be
spent installing the LIMB technology on full-sized boilers, which will
then be subjected to two years ot* performance optimization and long-term
evaluation. The total tab for the LIMB program will amount to $16.5
million, which will be a bargain if LIMB fulfills its initial promise.
The development effort will be co-sponsored by EPA and the Department of
Energy.
I've been talking a lot about EPA's plans for developing N0X
control methods, and the sort of projects we have underway. But, I'm
15
-------
not forgetting the crucial role industry must play in this mission. EPA
has the resources to provide the fundamental research and the testing of
new control technologies, but we must rely on industry to provide the
host sites that allow technologies to be tested under real-life conditions.
And, we must depend heavily upon the commercial expertise and engineering
experience of boiler manufacturers if a technology is to progress beyond
the demonstration stage.
There's always an element of risk for the private sector when it
invests in new equipment and new technologies. Control processes that
look promising on the drawing board or during small-scale experiments
don't always pan out when they are put into practical use. But we at
EPA believe that with the kind of cooperation between government and
industry we have enjoyed up to now, and with an equitable sharing of the
risk which is, after all, inherent in all innovation, we can solve the
pollution control challenges that we face.
For the rest of today and in the course of the next two days, you
will be hearing about a lot of new ideas, new techniques, and new
technologies aimed at keeping nitrogen oxides out of the air. Personally,
I am very excited about these developments. A number of them will have
far-reaching effects. For example, the low-NOx burner or the LIMB
technology may provide the technical fix for what will be an increasingly
difficult political problem we will be having with our Canadian neighbors
over our acid rain export to Southern Canada. Already these control
methods have advanced from being just a gleam in an engineer's eye to
successful operation in the laboratory, in pilot plants, and in a few
16
-------
small combustors. I'm willing to bet you that when we gather together
ain in a year, many of these technologies will be starting to move into
the commercial market, working to produce electricity, generate power,
and heat homes. And we'll all be able to breathe a little easier.
17
-------
AIR QUALITY CONTROL FROM A STATE REGULATOR'S VIEWPOINT
(Luncheon Address, October 8, 1980)
L. Michael Henry, Chairman
Colorado Air Quality Control Commission
Ladies and Gentlemen, I am very honored to be here, but I must say that
I am also considerably intimidated - for two reasons.
First, as I believe you know, I am substituting as a speaker for United
States Senator Gary Hart from Colorado. I am intimidated because I
certainly have nowhere near the knowledge and stature of Senator Hart on
air quality issues.
We are very proud in Colorado that Senator Hart is respected so much in
the national government to be the Chairman of the National Commission on
Air Quality. He has also played a vital role in fighting to reinstate high
altitude standards for automobile emissions. As you know, our mile-high
altitude in Denver and much higher altitudes in our mountain communities
contribute very significantly to automotive air pollution due to incomplete
gasoline combustion at higher altitudes. In addition, Senator Hart has
helped focus national attention on Colorado's deep concern about the pros-
pect of economic and environmental problems resulting from projected
massive development of the oil shale industry in the Piceance basin in
northwestern Colorado. Senator Hart held a large public hearing in western
Colorado on the oil shale development proposals of Exxon about five weeks
ago. For some reason our Colorado Air Quality Control Commission was
not notified of that meeting and, being a feisty Commission, we did send
Senator Hart a message of complaint about this failure of communication.
I cannot quite put out of my mind the possibility that Senator Hart was trying
to punish me for this complaint by suggesting that I be thrust as a speaker
before such an august meeting of international experts in stationary source
control. Seriously, I do appreciate his suggestion that I might have some
words of interest to you.
18
-------
Secondly, I am very intimidated to be addressing you as a complete
.scientific lay person and amateur. 1 was very lucky to complete one course
in high school physics and one course in high school chemistry. I have
retained very little of those courses except for a deep respect for the
discipline of those hard sciences. I am even less knowledgable in the laws
of medical science or mechanical engineering or computer sciences, and
1 hope you will keep this in mind in any questions you might have.
Much of what I would like to talk to you about relates to the relationship
between a citizen non-scientific regulatory commission as we have in
Colorado and the regulated industries and the affected public.
In Colorado we have a 9-person part-time Air Quality Control Commission
all appointed by the Governor. We are paid the grand sum of $40 a day
for each day of work with a ceiling of .$1, 284 per year. We have regular
meetings twice a month and numerous committee meetings with groups
such as yours every month. Our Colorado statute requires the Governor
to give consideration to appointing persons with appropriate scientific,
technical, industrial, labor, agricultural, and legal training or experience,
although no specific number of members needs to be from any specific
background. The current members of our Commission include one lawyer,
one lawyer=businessman, one woman rancher, one woman Chamber of Com-
merce representative and former City Council woman, two civically-active
League of Women Voter members, one medical doctor, one professor of
Engineering, and one engineer with a large manufacturing company.
Our current political makeup, which never is reflected in any of our votes,
is three Democrats, three Republicans, and three unaffiliated members.
Our statutory obligations include the functions of preparing a comprehensive
State Implementation Plan (SIP) to assure attainment and maintenance of the
National Ambient Air Quality Standards, preventing significant deterioration
and the preparing of emission control regulations.
We are very proud that th; State Implementation Plan (required by the
Federal Clean Air Act) which was submitted by Colorado to the Environmental
Protection Agency in early 1979 was the first plan submitted by any state.
We are not quite so proud that our State Implementation Plan was the first
also to bring sanctions from the Environmental Protection Agency due to the
failure of our Legislature to pass in a timely fashion legislation for inspection
and maintenance of automobiles in our non-attainment areas. These Environ-
mental Protection Agency sanctions have since been lifted after the Legislature
did finally pass such Inspection and Maintenance legislation.
19
-------
In the two and one-half years since I have been a member of the Colorado
Commission, I have been extremely impressed with the quality and diligence
and good common sense of the members of our Commission. Many have
suggested that the State would be better served by having full-time paid
members of the Regulatory Commission who would either be scientists or
at least persons who can afford time to be scientifically and rigorously
trained on the job. I certainly do wonder sometimes myself whether a
deeper scientific training would have helped me be a better policy maker,
particularly in the middle of public hearings when acronyms and medical
phrases and chemical formulas and computer modeling results and engineering
terms are thrown at us with scarcely any interconnecting English sentences.
On reflection, however, I believe that we do have a very workable system
of regulatory air quality control in Colorado for four reasons:
First, we do have very competent technical permanent staff for the analyses
of permit applications, for air monitoring, for planning and analysis of
overall state strategies and for inspection and enforcement of our regula-
tions. The trouble is that we do not have enough staff or enough funding to
retain competent staff, partly due to a seven percent annual increase in
Stale spending limitation which has boon passed by our Legislature. The
entire State Air Pollution Control Division has approximately seventy full-time
staff, some of whom provide administrative support for the technical person-
nel. We have a budget of approximately 2. 6 million dollars of which 10%
is passed through to local health departments.
In addition to Colorado's spending limitations in a time of double-digit
inflation we are also very concerned about future uncertainties in Federal
Environmental Protection Agency budgeting in the next several years. We
are very pleased however, with the staff resources which we have, given
the limited staff numbers and the fiscal limitations we must live with.
Second, we have an enormous reservoir of public concern in Colorado about
air quality control to spur us on. Very many people first came to Colorado
around the turn of the last century on doctors' orders so that our pure air
would ease the problems of people with tuberculosis and other respiratory
diseases before they died. The fact is that many survived for very many
years in our clean air. We have had a population boom for the last many
years with many tourists and military short-term visitors attracted by our
great weather and scenery and air to remain permanently in Colorado. None
of these people are going to sit quietly by while our air deteriorates. Many
times at our hearings, unfortunately my opinion, we hear very little from
these people - as most of them have neither the scientific background nor
the economic resources to spend time in testifying. However, we do feel
a deep obligation to heed those who recall us to our heritage of clean air and
20
-------
respect for our "purple mountains majesty" (as one public witness told us
a couple of months ago). Many environmental groups do come however, and
provide significant detailed analyses to us, which serves as a useful counter-
balance to the great scientific resources of industrial representatives.
Third, we do have what I believe to be a very cooperative and mutually
respectful attitude in Colorado between oir Regulatory Commission and
our major industries. All of our industries have a stake in maintaining
Colorado as a healthy and attractive environment and all, with very few ex-
ceptions have recognized this obligation.
Fourth, we do have what I believe to be a very capable and constructive and
common-sense Commission even given some of our technical weakness. We
consult prior to drafting regulations with affected industry and with citizens
and local governments and are very thoughtful about all of our final regula-
tions. We are attempting now to develop a new approach to rule-making to
respond to an EPA's regulation to protect visibility in Class 1 mountain parks
areas which will come down in November. We have concluded that we would
all be better served by a regulation which has been drafted jointly bv the
State and affected industry and the public at the beginning rather than having
adversarial polarized groups at the public hearing yelling not only at each
other, but at the Regulatory Commission. We are attempting to bring the
resources of environmental groups and industries into the early stages in a
calm atmosphere not only to defuse the process, but to capture the energy
and resources and creativity of all groups, especially industry, when the
resources of the State technical staff are limited, as I have said.
I think we all must recognize that when regulated industry has a role in finding
solutions, certainly the outcome is much more practical and sensible. Those
working deeply in an industry certainly know better than anyone else the
unique problems of that industry. I am very favorable to the "bubble" con-
cept, which is just now being applied to several industrial sources to allow
the industry itself to find creative ways to reduce the overall emissions from
that industrial site without outsiders attempting to establish regulations for
every individual source inside the location.
1 certainly commend the Environmental Protection Agency and the Electrical
Power Research Institute lor conducting this first joint symposium, which
also reflects an understanding and need for cooperation between industry and
regulators.
I will give you just a few comments on some of our specific Colorado problems:
At the time that our State Implementation Plan was submitted in early 1979,
different parts of Colorado violated all of the National Ambient Air Quality
21
-------
Standards except for sulfur dioxide. In Denver our 1977 average annual of
NOx was . 054 parts per million when the national standard is an annual
average not to exceed .05 parts per million. In Denver during 1977, 37 per-
cent of our nitrogen oxides (both NOx and NO2) were from automobiles.
Approximately 50% from large stationary sources; and approximately 10%
from space heating. In 1982, our px-edictions are for approximately the
same contributions toward our nitrogen oxide emissions, with a slight in-
crease in the percentage to be emitted from space heating.
The fact that our nitrogen oxides are emitted by three different large factors
differs from the situations we have with our other pollutants. In Denver,
93% of our carbon monoxide is contributed by automobiles, 85% of hydro-
carbons are contributed by automobiles and 75% of our Total Suspended
Particulates are contributed by automobiles. With the multiple contributors
of nitrogen oxides, the control strategies are more complicated than if the
greatest part of the pollutant came from one source.
In Colorado we are also facing a great deal of work and study to prepare
regulations to protect our visibility in Class I areas. As I indicated, we
in Colorado feel an enormous obligation to protect our mountain beauty.
The task will be especially difficult due to the fact that visibility-measure-
ment is not at all a fully-developed science.
Another very large problem facing us relates to the development of the oil
shale industry in northwestern Colorado. We are extremely conscious that
the entire nation is looking toward northwestern Colorado as a very valuable
source of oil shale and also coal, but the scale of some of the proposed oil
shale developments is absolutely overwhelming. Exxon, for example, has
recently proposed an eight-million barrel per day development in the next
several years. Most of our state officials and industry representatives
believe that a 1. 5 million barrel per day oil shale industry is much more
realistic. Even so, the population increases and the energy capability to
serve not only the industrial processes but the needs of the added people
themselves, pose enormous problems for northwestern Colorado. In
addition, the proposed MX missle system will also need approximately
180 megawatts of new power to serve not only the missile system itself, but
the added population in western Colorado and eastern Utah.
One of our other major endeavors in Colorado relates to automobile inspection
and maintenance. The Legislature finally did pass implementing Legislation
in May of 1980. The program will be operational in full beginning in January
of 1982. Many of us are concerned that our inspection and maintenance pro-
gram will have little if any effect on reducing visible pollutants. Although
most of us are most concerned about health-related pollutants, we believe
that many members of the public are most keenly aware of the visible effects
22
-------
which damage our mountain views. We will thus be feeling considerable
public pressure to control the visible pollutants as aggressively as we do
the invisible pollutants.
In other words, we have plenty of work to do in Colorado; and, given our
history and our recent activites, I have every confidence that we will be
able to handle our problems. With cooperation from industry, we will be
able to do even better. I am very glad that you have offered me the oppor-
tunity to speak to you, and I trust that your visit to Colorado will be pleasant.
Thank you very much.
23
-------
REGULATORY PRESSURES FOR INCREASED NOx CONTROL
By!
Ronald E. Wyzga, D.Sc.
Electric Power Research Institute
Palo Alto, California 94303
24
-------
REGULATORY PRESSURES FOR INCREASED NOx CONTROL
Electric utilities have had little concern with
nitrogen oxide emissions. Particulates and sulfur oxides have
received a great deal more attention. New and likely
regulations will change this situation soon and could cost the
electric utility industry as much as $11 billion (1).
I want to discuss the significance of impending and
likely NO regulations upon the electric utility industry
as well as the impact of other new laws upon the choice of
an NO control technology.
The Clean Air Act Amendments of 1977 request a short-
term ambient standard for N02. No standard has yet been
set, but current discussion centers upon a one-hour average
N02 ambient concentration of 0.125 - 0.50 ppm (2). Although
this limit will be more restrictive than current S02 standards,
small and medium-sized coal-fired power plants can probably
satisfy the N02 limit (1). If we assume a one-hour standard
of .25 ppm N02 and assume that all N0X is converted to N02,
then a 500 MW(e) power plant in flat terrain and complying
with new source performance standards can easily conform to
this standard if background levels of N02 are not already
high. (See Table 1.) The same is also true for 1000 MW(e)
plant. Significant background levels of N02 and uneven
terrain would make it more difficult to satisfy the regulation.
25
-------
Relaxation of the 100% N0x to N02 conversion rate would
make it easier to satisfy the regulation. A coal-fired
power plant 1500 MW(e) or larger could not, however, meet
a one-hour standard of .25 ppm even if it satisfied new
source performance standards. Additional NOx control
would be needed.
A second requirement of the 1977 Clean Air Act
Amendments is the Prevention of Significant Deterioration
(PSD) provision for N02« The format for these regulations
is not yet known, and several alternatives are under study.
If, however, an increment format similar to that established
for SC>2 and particulate matter is chosen, it will present
significant constraints upon the electric utility industry.
The allowable PSD increments for SO2 and particulate matter
are generally 25% of the ambient standards for Class II areas.
For Class I areas the increments are about 2 - 5% of the
ambient standards. If similar percentages were applied to
the NO2 standards, the impacts on power plant siting, size,
and control technology choice could be great. For example,
the N0x emission from a 500 MW(e) coal-fired power plant
satisfying new source performance standards would lead
to concentrations which would exceed the increment by 50
percent. (1) We assume flat terrain and total conversion
of N0X to NOj in this calculation. Uneven terrain would
lead to greater constraints, and relaxation of the total
26
-------
conversion assumption would reduce the constraints a bit.
The only way to satisfy the PSD regulations would be to
construct very small coal-fired power plants with capacities
less than about 300 MWe or to introduce controls which
would reduce NO emissions beyond new source performance
standards limits. The hypothesized Class I increments for
N02 would require that 500 MWe power plants complying with
new source performance standards be sited at least 100 km from
Class I areas. Again smaller power plants or greater NO
control could reduce the 100 km distance.
The Clean Air Act Amendments also provide for legislation
to improve and minimize visibility impairment. Visibility
regulations are imminent and could present significant
constraints for power plants sited near Class I visibility
areas. A relatively small amount of N02 in a power plant
can lead to perceptible brownish coloration of the plume.
In fact, preliminary results from the VISTA study under-
taken jointly by EPA and several private groups, indicate
that N02 is the most important contributor to plume coloration
in the Southwest. (3) Visibility regulations will then
certainly address NOx emissions. According to one set of
tentative criteria (4), a coal-fired power plant as small
as 750 MWe may require flue gas treatment of NO if the
27
-------
plant is sited in the visual range of the Class I visibility
area. (1> The visual range is 100 to 200 km in the West;
in the East it is from 15 to 50 km.(5)
There are two additional environmental concerns which
could lead to increased pressures for decreased NOx emissions.
The nitrogen oxides, along with hydrocarbons, are precursors to
ozone formation. Oxidant control in the past has relied heavily
upon hydrocarbon emissions control, but future controls may
also require additional NOx limitations. This is of
particular concern in those 16% of U.S. counties which
violate the oxidant standard (6). These counties, in general,
are the sites of the major urban areas in the U.S. Power
plants upwind of these areas might anticipate demands for
increased NO control. In some places, such as California,
demands have already been formulated.
The acid rain problem is receiving a lot of attention,
and there is some public clamor to control acid rain. Because
nitrogen oxides may be responsible for up to 30 - 40% of
the acidity of rainfall in the Northeast U.S. (7), any
efforts to control acid rain will probably involve some
control of nitrogen oxide emissions from stationary sources.
All five of the planned and potential regulations can
have tremendous impacts upon the electric utility industry,
particularly in its planning of new plants. The regulations
28
-------
will impact coal-burning power plants most, although synfuel
plants can also be significantly impacted. At a time when
nuclear power plants were more popular, the increased
restrictions and costs imposed by these regulations could
lead to a greater share for nuclear of new power plants.
Given the current mood, this is unlikely to happen. Alternative
coal types are not likely to influence N0x emissions
significantly, but shale oil and coal liquid could yield
greater NO emissions than fuel oil. (1) Hence these
A
regulations could influence fuel choice and the demand for
shale oil and synthetic fuels. The regulations could lead
to smaller power plants as they may be allowed with existing
and inexpensive controls and they may be sited more easily.
If, however, a PSD increment for NO2 is set as outlined above,
the maximum allowable power plant size, 300 MWe for a
coal-fired plant, may be too small for utility consideration.
Siting will also be impacted. New power plants can expect
to be sited further from Class I areas and probably further
from oxidant non-attainment areas.
Greater NO emissions control can allow power plants
to satisfy the above regulations without decreasing size
or siting flexibility. The potential PSD increments
could, however, even pose problems for larger power plants
using the most advanced available control technology.
29
-------
The increased costs of the control technologies will be
weighed against size and siting in planning new power
plants.
There is another factor, however. The most advanced
control technologies are new and largely untested. Before
these new technologies are chosen, we must make sure that
they do not introduce any new risks to the environment
and that they satisfy regulations in addition to those which
they are designed to help meet. Table 2 gives a partial
list of those regulations which require compliance of new
technologies with some minimal level of risk; moreover,
there are concerns with some of the substances which have
been mentioned in connection with the NO control technologies.
The Clean Air Act Amendments under section 112 (NESHAP),
for example, are concerned with airborne carcinogens
including polycyclic organic materials (POM). It is unclear
whether or not combustion modification leads to increased
POM emissions. The Clean Water Act Amendments are concerned
with ammonia and nitrosamines along with many other toxics.
It is unclear as to which if any quantities of these
substances may be released to the environment with selective
catalytic reduction. The Occupational Health and Safety
Act is concerned with the risk of occupational exposure
to vanadium, a catalyst in selective catalytic reduction.
30
-------
The risks associated with this use of vanadium have not
yet been estimated. The Hazardous Materials Transport
Act is concerned with the risk of transporting ammonia
as well as other substances. The risk of ammonia transport
associated with selective catalytic reduction is unknown.
Other legislative acts raise other concerns.
These examples illustrate the complexity of the current
regulatory environment and the necessity to consider a much
broader set of regulations than previously. This is
particularly true for the N0x control technologies. Relatively
little experience is available at the operational level.
As these technologies are tested for efficacy of NO
control, they must also be examined to insure that they
do not introduce any new risks to the environment and that
they can comply with the entire regulatory spectrum. The
failure to identify and correct any deficiencies during
the development of the technologies can only lead to greater
costs and problems. The uncertainty attached to any new
technology must be weighed by utilities in deciding how to
comply with the new N0x regulations. There is a price
attached to uncertainty. In this case one could imagine
technologies which may be obsolete before they are constructed
or the need to build one gismo on top of another a la Rube
Goldberg. Practices for the use of the technologies may have
to be altered as well. All of these will require more money,
lots of it. The result may be that utilities will give
31
-------
more weight to siting solutions and smaller power plants
in the near term than they would have given in a simple
regulatory environment.
32
-------
REFERENCES
1. Hayes S.R. et al (1980) NO.., Air Quality, and the
Electric Utility; A Guidance Manual, to be published
by EPRI, Palo Alto, CA.
2. U.S. Environmental Protection Agency (1978)
Proposed Short-Term National Ambient Air Quality
Standard for Nitrogen Dioxide, draft environmental
impact statement.
3. Personal communication, Dr. P. Bhardwaja, Technical
Director for Vista at Salt River Project, 1980.
4. Personal communication with D.A. Latimer, Systems
Applications, Inc., 1980.
5. Trijones, J. and D. Shapland (1979) Existing Visibility
Levels in the U.S., EPA-450/5-79-010. Technology
Service Corporation, Santa Monica, CA.
6. Garvey, D.B. et al (1979) Non-attainment of National
Ambient Air Quality Standards; Implications for Energy
Policy, Argonne National Laboratory, Argonne, IL.
7. Charles V. Cogbill & Gene E. Likens, (1974) , "Acid
Precipitation in the Northeast U.S,M Water Resource
Research 10(6), 1133-37.
33
-------
TABLE 1: Power Plant Abilities to Satisfy a 1-hour N02
Standard of 0.25 ppm
Standard of 0.25 ppm, Power-plant* abilities to satisfy
standard with a 1-hour NC>2
Power Plant Size Maximum Concentration
as % of Standard
500 MW(e) 40%
1000 MW(e) 80%
1500 MW(e) 120%
* coal-fired power plant satisfying New Source Performance
Standards located in flat terrain
34
-------
TABLE 2: Partial List of Regulations Requiring
Risk Assessment
o Clean Air Act Amendments
NESHAP (National Emissions Standards for
Hazardous Pollutants
o Clean Water Act
o Occupational Safety and Health Act
o Hazardous Materials Transport Act
o Resource Conservation and Recovery Act
o Toxic Substances Control Act
o Federal Railroad Safety Act
o Ports/Waterways Safety Act
o Safe Drinking Water Act
35
-------
DEVELOPMENT AND REVISION OF AIR QUALITY STANDARDS
WITH SPECIAL ATTENTION TO THE N02 STANDARD REVIEW
By:
Michael H. Jones
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
36
-------
ABSTRACT
This paper describes the process for review of National Ambient Air Quality
Standards. Special attention is given to the issues facing the Environmental
Protection Agency in assessing the need for and nature of possible modifications
to the UC>2 ambient air quality standards. The legal requirements for the
Clean Air Act Amendments of 1977 are discussed as they apply to this review
and to the decision process in making a standard choice. The paper describes
not only the importance of the scientific basis for selecting a standard but
also the role of the policymaker and the judicial process. Criteria document
development, the scientific review process, the preliminary staff position
paper and the public review process are all described. Finally, the critical
elements in the upcoming NO^ standard decision are identified and discussed.
37
-------
INTRODUCTION
The development of National Ambient Air Quality Standards (NAAQS) is
a step-by-step process followed by the Environmental Protection Agency (EPA)
that includes: (1) the assessment of scientific information, (2) generation
of a consensus within the scientific community on the veracity of this
assessment, (3) an exchange of views and information with the public sector
following proposal of a standard, and (4) promulgation and enforcement of a
final rule. EPA's charter for this review process is the Clean Air Act and
related amendments passed by the United States Congress. This paper reviews
the requirements of this legislation, how the Agency has Implemented these
requirements, and a brief status update of the NO 2 standard review.
BACKGROUND
Seven National Ambient Air Quality Standards have been established under
the provision of the Clean Air Act Amendments. Substances for which these
standards have been set are presented in Table 1 and include carbon monoxide
(CO), hydrocarbons (HC), lead (Pb), nitrogen dioxides (NO2), sulfur oxides
(sox)> particulate matter (TSP), and ozone (0^). All of these standards
with the exception of ozone and lead were originally set in 1971. The 0.12 ppm
ozone standard is the result of the 1978-79 review and revised the old 0.08 ppm
oxidant standard, while the lead standard was promulgated in October 1978 after
litigation brought against the Agency by the Natural Resources Defense Council.
Following promulgation of the original standards, a number of reviews
of the air quality criteria were conducted by the National Academy of
Sciences (NAS). For example, the NAS reviewed the basis for all of the
standards for Congress in 1974, and another NAS study was prepared on the
criteria for sulfur oxides for the Senate Committee on Public Works in 1975.
EPA also contracted with NAS to review the scientific basis for each of the
criteria pollutants, resulting in NAS reports completed in 1977 and 1978.
38
-------
The NAS reviews found that there was insufficient new scientific informa-
tion to justify changing the existing standards, although in several
instances revisions to ambient air quality criteria were recommended.
LEGISLATIVE REQUIREMENTS
The goal of the Clean Air Act is to protect the public health and
welfare and enhance the quality of the nation's air. Under the Act, the
Federal government is responsible for establishing, on a nationwide basis,
ambient air quality standards that are stringent enough to protect the
public health with an adequate margin of safety. In order to provide for
attainment of these standards, the States are responsible for specifying
emission limitations and other programs for individual sources through
State implementation plans (SIP's).
The first step in establishing an ambient air quality standard is a
finding by the Administrator of the Environmental Protection Agency that a
particular pollutant causes or contributes to air pollution which, in the
words of the Act, "may reasonably be anticipated to endanger the public
health or welfare." Within 12 months after the listing of a pollutant under
section 108(a) of the Clean Air Act, the Administrator must publish an air
quality criteria document which will form the scientific basis for the ambient
air quality standard. The criteria document must contain the "latest
scientific knowledge useful in Indicating the kind and extent of all
identifiable effects on public health or welfare."
Simultaneously with publication of the criteria document, the Adminis-
trator must propose primary and secondary national ambient air quality
standards, as appropriate. A primary standard is one that, in the Administra-
tor's judgment, is required to protect the public health with an adequate
margin of safety. Costs of attainment are not a germane consideration in
setting the primary standard, although such costs are to be considered in
the development of SIF's. A secondary standard is one that adequately
protects the public welfare. Public welfare is defined as including, but
not limited to, effects on solid, water, crops, vegetation, man-made materials,
animals, wildlife,weather, visibility, climate, damage to and deterioration
of property, and hazards to transportation, as well as effects on economic
values and on personal comfort and well-being. After providing a public
39
-------
comment period and hearing, the Administrator is required to promulgate final
standards; standards are usually promulgated within 6 months of his initial
proposal.
The 1977 amendments require that all existing criteria documents be
reviewed at 5-year intervals by a newly created Clean Air Scientific Advisory
Committee (CASAC) of EPA's Science Advisory Board (SAB). An independent
body made up of scientists and engineers with substantial scientific and
technical expertise, the SAB is chartered by the Administrator to provide
critical review of scientific matters before the Agency, as well as indepen-
dent advice. (The SAB's authority to comment on draft criteria documents
was statutorily established by the Environmental Research Development and
Demonstration Authorization Act of 1978.) In addition to the establishment
of the CASAC, section 109(d) of the 1977 Act further directed the Administrator
to complete reviews of all existing standards and criteria before the end of
1980 and at no longer than 5-year intervals thereafter, and to revise the
criteria and standards, as appropriate, based on those reviews.
Once an ambient standard is promulgated, responsibility under the
Clean Air Act shifts from the Federal government to the States. Within 9
months after promulgation, each State is required to prepare and submit a
State implementation plan (SIP) to EPA for approval. These SIP's must
contain emission limitations and must describe all other measures necessary
to attain the primary standard "as expeditiously as practicable" but not
later than 3 years after EPA approval and to attain the secondary standard
within a reasonable time.
CONSIDERATIONS IN ESTABLISHING AIR QUALITY STANDARDS
When the Agency undertakes either to establish or revise a standard,
questions often arise concerning what the Administrator must consider in
establishing a primary standard. Section 109(d) of the Clean Air Act requires
that national primary ambient air quality standards be set at a level allow-
ing for an adequate margin of safety, which the Administrator judges is
2
adequate to protect the public health. The statute and legislative history
make it very clear that the standards are to be solely health based, designed
40
-------
to protect the most sensitive group of citizens (but not necessarily the
most sensitive members of that group) against adverse health effects. The
task of deciding which health effects are adverse is a difficult one, and
to accomplish it the Administrator must exercise his judgment as allowed by
the Act in the investigation of the range of health effects of a pollutant
and in the consideration of the risks disclosed by the investigation. In
addition, the requirement of a margin of safety and the precautionary nature
of the Act indicate that the standards must protect against uncertain, as
well as certain, effects.
In the process of setting the primary standards, the Administrator must
typically deal with two different kinds of issues: (1) which effects on a
continuum of known effects should be regarded as adverse? and (2) what
degree of additional protection is required to protect against uncertain
harms—those not yet identified by research, or identified but not yet fully
understood?
Any attempt to determine that level of pollutant exposure at which the
effects are adverse to the sensitive population faces an immediate problem:
investigation of the effects typically reveals that while there are levels
above which "adverse effects" clearly exist, it is not generally possible
to identify sharp "thresholds" for such effects. Rather, expanding scientific
knowledge and better analytical techniques have made it clear that pollutant
effects typically exist as a continuum, ranging from clearly serious health
effects at high pollutant concentrations, to physiologically detectable
effects of uncertain significance, to effects which are too subtle to measure.
Even though Congress was aware that there is no sharp "breakpoint"
between a "no-effect" level and a level where the effect is clearly adverse,
it still required that standards be set and that the standards be set at the
point where there is no adverse effect for the sensitive population. The
fact that health effects exist on a continuum creates a difficulty in
identifying that level. In addition, the term "adverse" itself is difficult
to define in the abstract. At different points in the continuum are levels
of pollutants at which the effects will be conceded to be "adverse" by
any given person. That point will vary, however, since medical judgments
about what is "adverse" will vary with the information available to the
41
-------
person and his or her viewpoint. In a given case, determination of what
health effects are adverse may be as much an exercise of informed judgment
as a factual injury.
In recognition of this, Congress in section 109(b)(1) of the Clean Air
Act explicitly provided that the Administrator is to exercise his judgment in
setting the standard. Though relying heavily on scientific advisors for
technical evaluation of data and for those judgments that are essentially
scientific in nature, the Administrator alone is responsible for considering
risks and determining at what pollutant concentration the health effects on
the sensitive population should be regarded as adverse. Medical experts may
differ as to which particular health effects are adverse, but the statute
gives the Administrator the responsibility of making that judgment.
RECENT SUPPORT FOR EPA'S APPROACH TO STANDARD SETTING
On the 27th of June 1980, the United States Court of Appeals, upheld
the principal elements of EPA's standard-setting philosophy in its decision
3
on the lead standard. The court specifically supported EPA's contention
that: (1) costs cannot be considered in selecting primary or secondary
standards and, (2) the Administrator's authority and responsibility for
making reasoned judgments in protecting public health in the face of
incomplete or uncertain evidence. Two cites- from the court decision amply
verify this judgment.
Statutory Authority
"Furthermore, we agree with the Administrator that requiring EPA to
wait until it can conclusively demonstrate that a particular effect is
adverse to health before it acts is inconsistent with both the Act's
precautionary and preventive orientation and the nature of the Administra-
tor's statutory responsibilities. Congress provided that the Administrator
is to use his judgment in setting air quality standards precisely to permit
him to act in the face of uncertainty. As we read the statutory provisions
and the legislative history, Congress directed the Administrator to
err on the side of caution in making the necessary decisions. We see
no reason why this court should put a gloss on Congress' scheme by requiring
42
-------
the Administrator to show that there is a medical consensus that the
effects on which the lead standards were based are 'clearly harmful to
health'. All that is required by the statutory scheme is evidence in the
record which substantiates his conclusions about the health effects on which
the standards were based. Accordingly, we reject the Lead Industry Association's
(LIA) claim that the Administrator exceeded his statutory authority and turn
to LIA's challenge to the evidentiary basis for the Administrator's decisions."
Cost and Economics Role in Standard Setting
"The petitioners' first claim is that the Administrator exceeded his
authority under the statute by promulgating a primary air quality standard
for lead which is more stringent than is necessary to protect the public
health because it is designed to protect the public against 'subclinical'
effects which are not harmful to health. According to petitioners, Congress
only authorized the Administrator to set primary air quality standards that
are aimed at protecting the public against health effects which are known to
be clearly harmful. They argue that Congress so limited the Administrator's
authority because it was concerned that excessively stringent air quality
standards could cause massive economic dislocation.
In developing this argument, St. Joe contends that EPA erred by refusing
to consider the issues of economic and technological feasibility in setting
the air quality standards for lead. St. Joe's claim that the Administrator
should have considered these issues is based on the statutory provision
directing him to allow an 'adequate margin of safety' in setting primary air
quality standards. In St. Joe's view, the Administrator must consider the
economic impact of the proposed standard on Industry and the technological
feasibility of compliance by emission sources in determining the appropriate
allowance for a margin of safety. St. Joe argues that the Administrator
abused his discretion by refusing to consider these factors in determining
the appropriate margin of safety for the lead standards, and maintains
that the lead air quality standards will have a disastrous economic
impact on Industrial sources of lead emissions.
This argument is totally without merit. St. Joe is unable to point
to anything in either the language of the Act or its legislative history
that offers any support for its claim that Congress, by specifying that
43
-------
the Administrator is to allow an 'adequate margin of safety' in setting
primary air quality standards, thereby required the Administrator to
consider economic or technological feasibility. To the contrary, the
statute and its legislative history make clear that economic considerations
play no part in the promulgation of ambient air quality standards under
Section 109."
This court decision would appear to lay to rest arguments questioning
the role of economics in standard setting and the Administrator's responsibility
in making choices under uncertainty.
STANDARD DEVELOPMENT PROCESS
Criteria Document
Figure I illustrates the various steps in the standard development process.
The first step in the process is to review criteria, and develop a
revised criteria document where appropriate. Main responsibility for pro-
duction of the document rests with the Environmental Criteria and Assessment
Office (ECAO/RTP) in EPA's Office of Research and Development (ORD). The
first phase of the documentation process is to plan and initiate document
preparation procedures. This phase includes assembling an internal EPA
task force and recruiting outside experts as consultants to aid in writing
the document. Together, these groups develop a work plan and define a
schedule for production of the document.
The next step includes accumulating and analyzing literature and
writing initial rough drafts of document chapters. Hard copies of every
article cited are obtained and kept on file for public inspection at ECAO
facilities. The actual writing of the drafts is carried out by ECAO staff,
other EPA research scientists, or non-Agency consultants, depending upon
the avialability of authors with the required expertise. These activities
result in the production of an initial working draft of the document.
Following this phase, a workshop is held, where non-Agency experts meet
with the document preparation team, which includes authors of the draft
chapters, to provide preliminary peer review of the document contents and to
assist in its revision. Post-workshop revisions lead to the production of a
first external review draft of the document. This draft is circulated to
44
-------
the public and CASAC for review and comment. The document is reviewed
by CASAC at a public meeting.
Following the public review meeting, GCAO staff members undertake
indepth cataloging of public and CASAC comments on the first external
review draft. All comments from CASAC, the public, and other reviewers
are passed on to the appropriate authors and are given consideration
in revising the document. Each comment and its disposition is considered
and entered into a docket, which is available for public inspection.
Consideration of comments and appropriate revision of the document text
result in a second external review draft.
The revised and reprinted draft is normally submitted to the public
and the CASAC again for external review, and an effort is made to achieve
final closure on the document with the CASAC. If no substantive crltcisms
are received as a result of this cycle of review or remain outstanding
after this cycle, then the CASAC/SAB indicates, in a written report to
the Administrator, that such is the case, confirming the CASAC's evaluation
of the document as being of appropriate quality for use as the scientific
basis for the related air standard.
STAFF PAPER
Once the criteria document has been reviewed by the public and the
CASAC and the document is nearing its final form, the Agency staff prepares
a paper, which evaluates the key studies in the criteria document and
identifies critical elements to be considered in the review of the standard.
The staff paper identifies those studies that the staff believes should be
used in making the best scientific judgment on the level at which adverse
effects signal a danger to public health in the sensitive population. In
addition, the paper provides a discussion of the uncertainties in the
medical evidence and of other factors that the staff believes should be
considered in selecting an adequate margin of safety and a final standard
level. In addition, the paper evaluates studies that the staff believes
should be used in making the necessary scientific judgments on the level at
which adverse effects signal a danger to public welfare. The paper does not
present a judgment on what concentration level should be established for
45
-------
the standard. The paper does help bridge the gap between the science con-
tained in the criteria documents and the judgment required of the Administrator
in setting ambient standards.
The staff paper is reviewed externally by the public and the CASAC. A
public meeting is held with the CASAC to receive their comments and the
comments of the public. Once the paper has been reviewed by the CASAC, the
scientific judgments made in the paper form the basis for the staff's
recommendation to the Administrator on any revisions to the standard. Our
4
initial experience with the review of the CO staff paper was extremely
good. The CASAC members were very positive in their comments, and they
found the paper to be an excellent vehicle for conveying the staff's view
of how studies presented in the criteria document should be used in setting
standards.
Assessing and interpreting the scientific evidence is a very complex
undertaking. For example, the CO criteria document contained over 200
references on the human health effects of CO. Reviewing these studies,
determining which are the most relevant to standard setting, and finally
interpreting the scientific evidence from the relevant studies are very
difficult and challenging tasks which the Agency must undertake each time
a standard is reviewed. This is even more difficult because there is often
considerable disagreement in the scientific community over how the studies
should be interpreted. The preparation of the staff paper and its review
by the public and the scientific community are our way of ensuring that the
staff's interpretation of the scientific evidence is sound and that the
Administrator has available to him a properly interpreted data base for his
decision making on air standards.
REGULATION DEVELOPMENT
The first general principle of the regulation development process is
the extensive and continuous participation by various EPA offices. Participatory
decision making continues to be important at EPA because systematic review by
offices other than the office with primary responsibility provides several
types of valuable input. Scientists and engineers check data and analyses;
lawyers check procedures, clarity, and consistency with the law; and other
46
-------
program managers evaluate how proposed regulations would affect their
programs. This process starts when the lead office, which has the
responsibility for the standard, invites Assistant Administrators, the
General Counsel, Regional Offices, and Staff Offices to send representatives
to participate in a work group in developing a regulation.
The work group advises and assists the lead office in preparing a
proposed regulation. The initial review of the regulation is by the
Steering Committee. This committee is a continuing group representing the
six Assistant Administrators' staff. Following Steering Committee reviews,
proposed regulations are reviewed by all Assistant Administrators, General
Counsel, and chief Staff Office directors. When consensus is not reached at
a particular level, the disagreement is spelled out, and the matter is taken
to a higher level for review. When consensus is reached on major issues at
lower management levels, the lead office identifies for senior management
the nature of the issue and the consensus that has been reached. As a result,
final decisions remain with publicly responsible appointed officials at the
top of the Agency.
The Agency also places a high priority on public participation in our
standard review process. EPA has provided for public and scientific review
of our criteria documents and staff papers. Ample public review is provided
for during rulemaking under section 307(d) of the Clean Air Act, as added by
the 1977 Amendments, including establishment of a public docket, provision for
a public hearing, and an opportunity to submit written comments. The final
regulation includes the Agency's response to the public comments.
The Agency process has worked extremely well in practice and was used as
the model for regulatory reform as presented in Executive Order 12044. The
process ensures both outside public review and top Agency management review
during the standard development process. The result has been an open and
objective decision-making process that gives consideration to opposings
viewpoints.
47
-------
N02 standard review
At this stage of the N02 standard review process, EPA has completed
a draft criteria document that covers concerns for both a short and a
long-term air quality standard. The document is currently undergoing
review by the CASAC and is expected to receive a favorable endorsement
from that group. A meeting will be held with the CASAC group to discuss
not only the veracity of the criteria document but also EPA's preliminary
ideas on how the scientific evidence will be used in making a decision on
NC>2 standards.
To provide insight into how this process may progress, Tables 2 and 3
are included as a representation of the critical studies expected to play
a major role in selecting the standard. As can be seen from Table 2,
clinical studies report a variety of effects at concentrations ranging from
0.1 ppm to 2.5 ppm. These effects include subtle responses such as elevated
reaction to a bronchial constrictor agent at 0.1 ppm to significant and
measurable pulmonary function decrements in the range 0.7 to 2.0 ppm. Other
responses such as symptomatic effects and slight pulmonary function impairment
have been observed at intermediate concentrations. A decision which has yet to
be made by EPA, is just what level of response and associated concentration,
should be considered as an indicator of an adverse health effect. EPA is
soliciting counsel on this issue from the CASAC and from the public and other
interested parties.
A second series of studies which will bear on the standard decision are
the so-called "gas stove studies." These studies seem to indicate that young
children suffer an elevation in the frequency of respiratory illness at NO^
concentrations of about 0.4 to 0.6 ppm. Unfortunately, a key piece of
information is missing from the gas stove studies, in that the exact air
quality distributions are unknown. It is not clear whether the 0.4 - 0.6
peak concentrations are for very short (minutes) periods of time, or for
longer periods (hours). It is also uncertain whether the observed response
is due to a single peak concentration, to several peak concentrations, or to
the long-term chronic exposure over a several-month period. Some evidence does
48
-------
exist, however, from animal studies, that it is the repeated short-term
peak concentrations that are most responsible for the reported insult.
A third area of major concern are the animal studies reporting a
decreased resistance to disease at NC>2 concentrations of 0.5 ppm and
above. These studies are, in turn, supported by other animal work showing
less pronounced effects at considerably lower concentrations. These effects,
such as destruction of lung tissue are considered by some investigators,
precursors to the elevated disease levels shown at higher concentrations.
Just how EPA will use these studies in arriving at a final standard
decision, will not be decided until after discussions with the CASAC and
after there has been an opportunity to receive public input on this important
issue.
49
-------
REFERENCES
1. The Clean Air Act as Amended, August 1977 (42 U.S.C. 7414).
2. A Legislative History of the Clean Air Act Amendments of 1970,
93rd Congress, 2nd Session (Comm. Print 1974), Volumes 1 and 2.
3. U.S. Court of Appeals for the District of Columbia Circuit.
Lead Industries Inc. vs. Environmental Protection Agency.
Motion on Abeyance filed June 27, 1980.
4. U.S. EPA Preliminary Assessment of Adverse Health Effects from
Carbon Monoxide and Implications for Possible Modifications of
the Standard (Draft), June 1979. (Staff paper presented at the
meeting of the Clean Air Scientific Advisory Committee, June 14-16,
1979.)
50
-------
OAQPS (OANR)
ECAO (ORD)
OAQPS (OANR)
SCIENTIFIC
COMMUNITY
ECAO (ORD)
OAQPS (OANR)
SCIENTIFIC
RESEARCH
(ORD) - OFFICE OF RESEARCH AND DEVELOPMENT
(ECAO) - ENVIRONMENTAL CRITERIA AND ASSESSMENT OFFICE
(OANR) - OFFICE OF AIR, NOISE, AND RADIATION
(OAQPS) - OFFICE OF AIR QUALITY PLANNING AMD STANDARDS
FIGURE 1. National Ambient Air Quality Standards
Standard Sotting Process
AGENCY REVIEW
PROMULGATION
PROPOSAL
CRITERIA
DOCUMENT
REGULATORY IMPACT
ANALYSIS
AGENCY
REVIEW
ADMINISTRATOR
DECISION
PUBLIC MEETINGS
AND COMMENTS
REGULATORY
DECISION
PACKAGE
ADMINISTRATOR
DL'CISION
REGULATORY DECISION
PACKAGE REFLECTING
PUBLIC COMMENTS
PUBLIC AND
SCIENTIFIC PEER
REVIEW
PUBLIC AND
SCIENTIFIC PEER
REVIEW
STAFF PAPER
INTERPRETING KEY
STUDIES IN
CRITERIA DOCUMENT
-------
TABLE 1. NATIONAL AMBIENT AIR QUALITY STANDARDS
Pollutant
Primary Standards
Averaging Time
Secondary Standards
Carbon monoxide
10 g/m3
40 g/nr
8-hour*
1-hour
Same as primary
Hydrocarbons
(Non-methane)
160 yg/m3
3-houra
(6 to 9 a.m.)
Same as primary
Lead
1.5 yg/m3
Quarterly average
Same as primary
Nitrogen oxides
100 yg/m3
Annual
(arithmetic mean)
Same as primary
Particulate Matter (TSP)
75 yg/m3
260 yg/m
Annual (geometric mean)
24-houra
60 yg/m3 b
150 yg/m
Ozone
235 yg/m3
l-hourc
Same as primary
Sulfur oxides
80 yg/m3
365 yg/m
Annual (arithmetic mean)
24-hour
3-hour
1300 yg/m3
aNot to be exceeded more than once per year.
^Guide to achieving the 24-hour standard.
cThe standard is attained when-the expected number of days per calendar year with maximum hourly average
concentrations above 235 yg/m is equal to or less than 1.
-------
TABLE II. KEY STUDIES
Compilation of Effects Reported In Selected Human Studies Examining Nitrogen Dioxide Exposures
N02
Concentration, Exposure Study
ppm Durations Population
Reported Effect(s)*
References
0.1
0.5
0.5 to 5.0
0.6
0.7 to 2.0
1.0
1.0 and 2.5
1.0 to 2.0
1 hr.
2 hrs.
approx.
15 min.
2 hrs.
10 mins.
2 hrs.
2 hrs.
2% hrs.
20 asthmatics
10 healthy
adults
7 chronic
bronchitics
13 asthmatics
63 chronic
bronchitlcs
15 healthy
exercising
adults
10 healthy
adults
16 healthy
adults
8 healthy
adults
10 healthy
adults
Effect of bronchoconstrlctlon Orehek, 1976
enhanced after exposure to N0?
1n 13 of 20 subjects. Neither
effect observed in 7 of 20
subjects. A bronchocon-
strictor (carbachol) was used.
1 healthy and 1 bronchltic
subject reported slight nasal
discharge. 7 asthmatics re-
ported mild symptomatic ef-
fects. Bronchitlcs and as-
thmatics showed no statisti-
cally significant changes for
all pulmonary functions tested
when analyzed as separate
groups, however small, but
statistically significant
changes in quasistatlc com-
pliance were found when
analyzed as a single group.
Significant increase in
airway resistance at or
above 1.6 ppm
No physiologically significant
changes in cardiovascular,
metabolic, or pulmonary function
after 15, 30, or 60 minutes of
exercise.
Increased Inspiratory and ex-
piratory flow resistance of
approximately 50% and 10X of
control values measured 10
mins. after exposure.
No statistically significant
changes 1n pulmonary function
tests with exception of small
changes 1n forced vital capacity
(1.5% mean decrease; p < .05).
Respiratory symptoms slightly
Increased after exposure to NO.,
but change not statistically
significant compared to controls
Kerr, et al., 1979
Von N1ed1ng et al.
1971
Follnsbee et al.,
1978
Suzuki and
Ishlkawa, 1965
Hackney, 1978
Increase 1n airway resistance
at 2.5 ppm but not at 1.0 ppm
Bell and Ulmer,
1976
Alternating exercise and rest Posln, et. al., 1978
produced significant decrease
for hemoglobin, hematocrit, and
erythrocyte acetycho1inesterase.
53
-------
TABLE III. COMPILATION OF REPORTED EFFECTS ASSOCIATED WITH EXPOSURE TO NITROGEN DIOXIDE
IN THE HOME IN COMMUNITY STUDIES INVOLVING GAS STOVES
Concentration
Study Population
Reported Effects
Referencefs)
Frequent peaks 0.4-
0.6 (gas), maximum
peak 1.0 (gas).
95th percentile
24 hr avg 1n activity
room
0.02-0.06 (gas)
0.01-0.05 (elec)
8,120 children, ages 6-10,
6 different cities, data
also collected on history
of illness before age 2
Significant association
between history of serious
respiratory Illness before
age 2 and use .of gas stoves.
Small but statistically signi-
ficant decreases In pulmonary
function (FEV, and FVC) 1n
children from gas stove homes.
Speizer et al.,
1980
NO2 not measured at
time of study
2,554 children from homes
using gas to cook compared
to 3,204 children from
homes using electricity,
ages 6-11
Bronchitis, day or night
cough, morning cough, cold
going to chest, wheeze, and
asthma Increased In homes
with gas stoves
Melia, et al.
1977
NO- not measured
in same homes studied
4,827 children, ages 5-10
Higher incidence of
respiratory symptoms
and disease associated with
gas stoves
Melia, et al.,
1979
Kitchens:
.005-0.317 (gas)
.006-0.188 (elec)
Bedrooms:
0.004-0.169 (gas)
0.003-0.037 (elec)
808, ages 6-7
Higher Incidence of
respiratory Illness 1n
gas-stove homes
Florey, et al.
1979 Companion
papers to
Melia et al.,
1979; Goldstein
et al.. 1979
Sample of Households
24 hr avg
0.005-0.11 (gas)
0-0.06 (elec)
0.015-0.05 (outdoors)
128 children, ages 0-5
346 children, ages 6-10
421 children, ages 11-15
Ho significant difference in
reported respiratory Illness
betweep-homes with gas and
electric stoves in children
from birth to 12 years old
Mitchell et
al., 1974
See also Kellrr
et al.. 1979
Peak hourly 0.25-0.50 Housewives cooking with
(gas),max hr 1.0 (gas) gas stoves compared to
those cooking with
electric stoves.
No Increase in respiratory
Illness
U.S. EPA, 1976
24 hr avg
0.005-0.11 (gas)
0-0.06 (elec)
0.015-0.05 (outdoors)
Housewives cooking with
gas stoves, compared to
those cooking with electric
stoves. 146 households
No evidence that cooking with
gas 1s associated with In-
crease in respiratory
disease
Keller, et al.
1979.
See above for
monitoring data
Members of 441 households
No significant difference
In reported respiratory
Illness among adults 1n
gas vs electric cooking
homes
Mitchell, et
al., 1974
See also
Keller et al.
1979
^Exposures in gas stove homes were to NOg plus other gas combustion products.
54
-------
ACID RAIN ISSUES
By:
Ralph A. Luken
U.S. Environmental Protection Agency
Office of Policy Analysis
Washington, DC 20460
55
-------
ABSTRACT
The U.S. Environmental Protection Agency (EPA) is concerned about the
effects of acid rain because the acidity of precipitation fallinq on the
U.S., Canada, and Scandinavia has been increasing for the past two decades.
An annual average precipitation of pH of 4.0 to 4.5 is not uncommon in the
eastern U.S., southeastern Canada, and western Europe.
Acid rain also has become more widespread in the past twenty years.
Once confined to urban and industrial areas, the effects of acid rain are
now being experienced in places as remote from industry as northern
Minnesota and Florida.
Although all the consequences of acid deposition are not well under-
stood, a growing body of evidence suggests that acid rain is responsible
for substantial adverse environmental effects. These include the acidifi-
cation of lakes, rivers and groundwaters, injury to aquatic species,
acidification and demineralization of soils, reduction of forest produc-
tivity, damage to crops, and deterioration of buildings and man-made
materials. In addition, the effects of acid rain on metallic elements in
soil, aquatic ecosystems, and drinking water systems may affect human
health adversely.
Current EPA authority generally focuses on control of the groundlevel
concentrations of the precursor pollutants of acid rain. Control of
these concentrations will not necessarily impact the level of acid rain
which is more a function of the total precursors emitted into the air in
a region over an extended period of time.
In order to reduce SO2 and NO* emissions, EPA in cooperation with the
Department of Energy, is evaluating alternative emissions reductions
strategies. These strategies focus primarily on utilijy and industrial
boilers.
(Paper not submitted for Proceedings)
56
-------
STATE OF CALIFORNIA PERSPECTIVE
NOx CONTROL FOR STATIONARY SOURCES
By:
Alan Goodley
California Air Resources Board
Sacramento, California 95814
57
-------
ABSTRACT
The California NOx control program is directed toward the achievement and
maintenance of air quality standards not only for nitrogen dioxide, but also
for ozone, total suspended particulate and visibility. In addition to
stringent controls on mobile sources, controls on existing stationary sources
and best available control technology (BACT) on new sources are needed in
non-attainment areas. In these non-attainment areas, the state is encouraging
local districts to adopt controls on refinery boilers and heaters, industrial
boilers, gas turbines, stationary I.C. engines, glass plants and cement
plants, in addition to existing controls on power plants. The state considers
selective catalytic reduction (SCR) to be BACT for most natural gas and oil
fired combustion sources, and that (SCR) will be BACT for coal-fired power
plants. We also believe that combustion modification techniques can be
improved so that SCR may be unnessary on some sources.
58
-------
If this meeting had been held in Los Angeles instead of the
beautiful City of Denver, most of you who live outside of Los
Angeles would probably have wondered to themselves whether or not
it would be smoggy in Los Angeles during the meeting. Furthermore,
those of you who work at reducing air pollution would probably have
the feeling that you are doing something to reduce the smog
problem.
Los Angeles and other California cities have severe air pollution
problems. I want to talk to you today about that air quality
problem and what the California State Government is doing about
it.
The federal annual average air quality standard for nitrogen
dioxide of 100 micrograms per cubic meter or 0.053 part per
million has been violated for many years in the South Coast Air
Basin or as it is otherwise known the Los Angeles Metropolitan
Air Quality Control Region. The State of California one hour
standard for nitrogen dioxide of 0.25 part per million is also
violated in the South Coast Air Basin. In Kern County, in
California, ambient concentrations of nitrogen dioxide have also
59
-------
been steadily increasing and almost approaching the air quality
standards for nitrogen dioxide as a result of the installation
of many hundreds of steam generators used to produce heavy oil
by steam injection into the reservoir.
As most of you probably know, oxides of nitrogen also participate
in the photochemical process which produces oxidants, among which
is ozone. The ozone standard is widely violated in many areas in
California, particularly in the South Coast Air Basin. While it
is generally considered that control of emissions of hydrocarbons
are necessary to achieve the ozone standard, it can also be shown
that control of emissions of oxides of nitrogen can result in
reduction in ozone concentrations under some circumstances and
particularly in rural areas.
Oxides of nitrogen are also converted, in. Dart, to nitrates in
atmosphere and form one of the constituents of total suspended
particulate. This is generally known as secondary particulate.
In the South Coast Air Basin, and in most urban areas, conversion
of sulfur oxides, oxides of nitrogen and hydrocarbons to particulate
accounts for a major part of the total suspended particulate. The
federal and state annual average and 24-hour standards for total
suspended particulate are also widely violated in many areas
particularly in the South Coast Air Basin and other urban centers.
60
-------
Achievement of the standards for TSP could not be made unless the
precursors of the secondary particulate portion of the total TSP
were controlled. For this reason, we believe that NOx control
is necessary to achieve the TSP standards, particularly in urban
areas.
The secondary particulate is very effective in reducing visibility
since the particles are close to the light scattering size range.
Therefore, NOx control is one of the essential st^ps that must
be taken in order to improve visibility.
In California, the program to control NOx is separated into two
elements: mobile source control and stationary source control.
The California Air Resources Board has the primary responsibility
for controlling emissions from mobile sources and the local air
pollution control districts have the primary responsibility for
controlling emissions from stationary sources. However, the
Air Resources Board is required by law to ensure that the local
air pollution control districts adopt and enforce measures needed
to achieve the state and federal air quality standards. If the
local air pollution control districts do not adopt such regulations,
the Air Resources Board is empowered to adopt the necessary regula-
tions for the local districts.
61
-------
The Clean Air Act provides that the State of California can receive
a waiver from the EPA vehicular emissions standards to adopt more
stringent standards. The Air Resources Board has adopted more
stringent standards for NOx emissions than has EPA. However, our
ability to reduce NOx emissions further from vehicular sources
has just about reached its limit, given the state of the art.
To achieve the ambient air quality standards, it will be necessary
to further reduce emissions from existing stationary sources and
to require offsets for emissions from new stationary sources.
All of the air pollution control districts in which the large coastal,
ocean water cooled power plants are located have adopted regulations
to limit emissions of oxides of nitrogen from power plants to 225
ppm when burning oil and 125 ppm when burning natural gas. There
is some variation from these levels, particularly for smaller power
plants. In addition, in the South Coast Air Quality Management
District emissions from large industrial boilers are also limited
to 225 ppm when burning oil or coal and 125 ppm when burning natural
gas. Of particular importance to utilities in the South Coast Air
Basin, a regulation has been adopted which requires further reduction
in power plant NOx emissions.
This rule, which is presently stayed and will be reconsidered at a
hearing at Los Angeles next month, introduces the new concept of
setting limits on total system-wide emissions in the air basin,
62
-------
rather than setting limits on emissions from individual units. This
concept, which is similar to EPA's bubble concept, allows a utility
to reduce emissions at the lowest cost. The rule, which is probably
the most stringent in the United States, requires a 90 percent
reduction in total system-wide emissions from levels that would have
occurred if all of the units had been burning oil. This reduction
must be achieved by 1988 or 1990, depending on the option chosen.
It also requires the two large utilities. Southern California Edison
and the Los Angeles Department of Water and Power, to demonstrate
by the end of 1982, on the equivalent of 100 megawatts or more of
capacity, equipment designed to achieve a ninety percent reduction
in NOx emissions. To comply with that requirement, Southern
California Edison plans to start installing selective catalyic
reduction equipment soon to treat one half of the flue gas
stream on a 215 megawatt unit at its Huntington Beach plant.
A paper to be delivered at this meeting will discuss that install-
ation.
The rule contains four options for compliance. The first two
options have two stages of reduction. The first option requires
about a 50 percent reduction from oil burning emission levels
by the end of 1983, and a 90 percent reduction by 1990. The
second option requires about a 40 percent reduction from oil
63
-------
burning emission levels by the end of 1983, and a 90 percent
reduction by 1988. The third option, which was basically proposed
by the Los Angeles Department of Water and Power, has a single
stage of reduction, 90 percent by 1990.
The fourth option, which was basically proposed by the Southern
California Edison Company requires an evenly stepped reduction
in annual average emissions each year until 90 percent is achieved
in 1990 plus a 75 percent reduction in peak emissions. The last
two options require installation of controls as early as feasible.
All of these options allow the reduction in emissions by any method,
including reduced burning of natural gas and fuel oil. For example,
if a nuclear power plant were brought on line, it would be considered!
as replacing oil and gas burning and therefore, would be a way of
reducing NOx emissions. However, it is anticipated that selective
catalytic reduction will be required on some units to meet the final
requirements of the rule. We believe that selective catalytic
reduction is a proven method of controlling NOx.
In Kern County, where I previously mentioned the increasing
concentrations of nitrogen dioxide, a rule has been adopted which
limits emissions of oxides of nitrogen from oil field steam gener-
ators. That rule has several innovative features. Firstly, it
has three stages or levels of control. The first stage of 0.3
pound of NOx per million Btu of heat input must be met by July 1,
64
-------
1982. The second stage of 0.25 pound of NOx per million Btu of
heat input is only applicable if the ambient concentration of
nitrogen dioxide reaches 0.20 ppm on a one-hour average or 0.04 5
parts per million on an annual average. The third stage of 0.14
pound of NOx per million Btu of heat input is only applicable if
the ambient concentration of nitrogen dioxide reaches 0.25 parts
per million on a one-hour average or 0.053 parts per million on
an annual average. If the second and third stages are not triggered
by air quality changes, emissions reductions made to meet the first
stage of the rule can be banked and used as offsets. Furthermore,
emissions may be averaged over the central or the western areas
of the county such that the total emissions from all existing steam
generators shall be no more than if each steam generator were in
exact compliance with the rule. It is expected that the first and
second stages of the rule can be achieved by the use of combustion
modifications. The oil companies and the burner manufacturers have
been taking aggressive steps to develop means to comply with this
rule. In particular, North American Manufacturing, the supplier
of most of the burners used in the steam generators in Kern County,
has developed a combustion modification which results in reduc-
tion of NOx to well below the requirements of the first and second
stages of the rule. The North American combustion modification
package is being tested on three separate steam generators, one
65
-------
each at three large oil producers. The cost of the control package
has been reported to be less than $35,000. At the current oil
prices, the saving in the cost of fuel in two years would be more
than the cost of the equipment. Also, it is our information that
more than 100 units are on order. If the third stage of the rule
is triggered by an air quality change, it is anticipated that unless
further reductions are achieved with combustion modifications,
Thermal DeNOx may be required on some steam generators in order to
comply with the emissions limit of 0.14 pound per million Btu of
heat input.
Concern has been expressed that the use of low NOx burners and
Thermal DeNOx in tandem may result in problems of: 1) the temper-
ature in the transition section of the boiler changing because
of the reduced excess air requirements of the low NOx burner and
2) the emitting of a blue plume because of the formation of
ammonium sulfate in the scrubber. To investigate these problems,
Getty Oil Company is conducting a year-long experiment on a 50
million Btu per hour scrubber-equipped steam generator which will
be equipped with a low NOx burner and Thermal DeNOx.
To effect further NOx reductions required to meet the air quality
standards, the Air Resources Board is working with local air
pollution control districts for measures to control emissions
from refinery boilers and heaters, industrial boilers, utility
66
-------
gas turbines, internal combustion engines, glass plants and cement
kilns. For refinery heaters we are looking at a limit of about
0.0 6 pound of NOx per million Btu of heat input, which is a reduction
of about 50 percent. We expect that the limit for existing refinery
and industrial boilers will be about 0.1 pound of NOx per million
Btu of heat input. We believe that such levels can be achieved by
the use of combustion modification techniques. For utility gas
turbines, we believe limits of 0.18 to 0.28 yg/Joule (25 to 40
ppm) could be achieved by the use of water injection but in the
long range we anticipate that catalytic combustion will be
developed so that NOx can be controlled to very low levels. For
internal combustion engines, we have proposed a limit of 0.75
grams per horsepower hour, a limit which would probably require
SCR or other catalytic means. Our investigations indicate that
NOx can be reduced from glass plants by modification of combustion
additional electric boost, new furnace designs and flue gas
treatment including Thermal DeNOx, SCR and potentially wet scrubbing.
It is anticipated that either singularly or in combination, these
control techniques will enable the Glass Industry to control NOx
emissions by 60 to 70 percent from uncontrolled levels. Work on
development of measures to control NOx emissions from cement kilns
has just begun with a workshop being held on October 8 in Los
Angeles.
67
-------
Major new sources in non-attainment areas are required to use Best
Available Control Technology to control NOx and in addition to
offset any emissions by reducing emissions from other existing
facilities. In our opinion, SCR is well proven for natural gap
and oil fired boilers and heaters.
Except for an 830 million Btu per hour heat input boiler operated
by Kerr McGee, there are no coal-fired power plants in California.
To meet future requirements, Pacific Gas & Electric Company has
proposed the construction of two 800 megawatt coal-fired units in
Northern California. In addition, Southern California Edison and
others have proposed the construction of three 500 megawatt coal-
fired units in Southern California. We are pressing for the
installation of selective catalytic reduction on these units. We
believe that by the time these units go into operation in the late
eighties, SCR will have been proven on coal-fired power plants in
Japan.
As you can see, we have a very aggressive NOx emission control program
in California, probably the most aggressive in the nation. Many of
you may feel that we are expecting too much, but history has shown
us that industry can rise to meet the challenge of stringent regul-
ations. We are deeply appreciative to those of you here who have
designed and developed ways to reduce emissions of oxides of nitrogen
and other pollutants and those of you in industry who have led the
way to get these methods or devices proven in practical operations.
We still need your help. There is a lot to be done. Thank you for
your attention. 68
-------
FOSSIL STEAM GENERATOR NOx CONTROL UPDATE
By:
Joseph A. Barsin
The Babcock & Wilcox Company
Barberton, Ohio 44203
69
-------
ABSTRACT
Since the Second NO Control Seminar of 1978, much additional information has
A
been collected concerning actual NO emissions from fossil fuel power plants
A
equipped with B&W Dual Register Burners/Compartmented Windbox furnace systems
for a wide range of both bituminous and subbituminous coals. These field
results from actual units fixing coal, oil and/or gas have demonstrated NO^
reductions of up to 60 percent compared to units without N0X control. This
presentation will show our controlled N0^ emission level experience and the
present status of B&W's advanced N0V control systems to meet the more restrictive
A
NO^ emission levels expected in the future.
ACKNOWLEDGEMENTS
The author appreciates the assistance of Ed Campobenedetto and Gayle Hixson
for their extensive work obtaining and reducing all of the NO^, data collected
by our Results Engineering Section on the many units tested to date. In
addition, I wish to thank Babcock-Hitachl in general and, specifically, Messrs.
Kubota, Mimura and Takeyama of that organization for their assistance.
70
-------
The 1971 federal New Source Performance Standards limited NO^ emissions for
most fossil fuels for the first time. Prior to the federal controls, local
air quality districts had established emission limitations in polluted areas.
For example, Babcock & Wilcox instituted a NO reduction program in 1957 at
A.
the request of utilities firing gas and oil fuels in the south coast air
quality district. The program led to successful application of two-stage combustion
and a patent granted to B&W in 1959 for the process. This early work was
followed by the selective use of gas recirculation to the burner to reduce
peak flame temperatures and further reduce N0X emissions. These approaches
were successfully retrofitted to the larger California units in the late
1960s and early 1970s, resulting in up to 60 percent N0X emission reduction.
In 1970, extensive work commenced on the development and application of a
combustion system to limit N0V emissions while firing pulverized coal. Previous
to the New Source Performance Standards, our mission as designers was to
maximize turbulence in the combustion zone, which would optimize carbon utilization
and, thereby, reduce the furnace residence time required to obtain acceptable
carbon utilization. Such an approach resulted in extremely high input burner
zones, small furnaces with high heat release rates, and the application of our
turbulent cell burner. This optimized combustion system successfully obtained
the goal of maximum carbon utilization and, unfortunately, generated high
levels of nitrogen oxide emissions. The kinetics involving the air/nitrogen
to NO transformations were understood well enough from our gas and oil experience
A
to provide direction in developing a new combustion system that would limit free
oxygen available in the flame zone and reduce peak flame temperatures.
Our thoretical combustion model indicated what we wanted to achieve in a burner
design to inhibit N0^ formation and at the same time continue to maximize carbon
utilization. We postulated that a fuel rich jet mixing with secondary air at a
controlled rate would reduce the peak flame temperature and control the oxygen
availability. This model was utilized to develop the Dual Register Burner. The
hardware development of that concept indicated that a spreader device of some
type was required to maintain homogenous particle distribution and avoid pitfalls
similar to those encountered in operation without impellers. Also required was
some means of balancing coal line pressure drop variations caused by different
-------
lengths of coal pipes being fed by the same pulverizer.
The mixing device selected initially was the venturi. While subjected to
erosion, the venturi was not subjected to high temperatures and therefore
would not coke or fall off. The overall primary air/fuel flow to each burner
was maintained by a permanent, calibrated orifice located in each coal line
between the pulverizer and the burners. The orifices were sized to equalize
the calculated line pressure drops for the various coal pipe lengths from each
mill.
The remaining combustion air (secondary air) was introduced through two
concentric air zones which surround the coal nozzle. The air flow to each air
zone was independently controlled through an outer air zone register and inner
air sleeve.
In addition to a mixing device (venturi diffuse) to obtain proper particle
distribution, some means had to be provided on the air side to control the
flame front location. Spin vanes were added to the inner air zone to control
axial/tangential secondary air velocity. The optimum vane position was found
to be a function of the fuel being fired.
These two additions to the model resulted in the "Dual Register Pulverized-
Coal Burner" (Figure 1).
By controlling the mixing of coal and air in a Dual Register Burner, the
combustion process can be initiated at the burner throat and the zone of
completion can be varied in the furnace chamber. This method of delayed
combustion reduces combustion intensity and acts to reduce peak flame temperatures
at each burner. Thus, the peak temperature in the furnace is lowered, minimizing
the thermal conversion of combustion air nitrogen to N0X< In addition, a
larger percentage of the furnace zone water-cooled surface is utilized during
the combustion process to further lower the peak temperature In the flame.
Finally, through controlled fuel and air mixing, the oxygen availability is
minimized during the process, thereby reducing fuel nitrogen conversion.
72
-------
The Dual Register Burner lowers NO by delaying combustion, not by staging.
Previous work on pulverized-coal firing has shown that two-stage combustion is
an effective method for NO reduction.*' However, in reviewing overall unit
performance, the Dual Register Burner has the following benefits over staging
techniques:
1. The furnace is maintained in an oxidizing environment to minimize slagging
and reduce the potential for furnace wall corrosion when burning high
sulfur bituminous coals.
2. More complete carbon utilization occurs through better air/coal mixing in
the furnace.
3. Lower oxygen levels are required when total combustion air is admitted
through the burners rather than above the turner zone.
RETROFIT
A retrofit system for the E. C. Gascon plant, Southern Electric Generating
Company, demonstrated N0X reductions of up to 50 percent, unstaged, compared
to levels obtained with the high turbulence circular burners which were replaced.
The correlation of heat available per square foot of adjusted burner zone
surface, which had been developed and utilized for commercial commitments on
gas and oil fuels, has been utilized for coal with good results. In fact,
the prediction curves generated by our N0X experiments on the retrofit at
the E. C. Gaston unit provide the basis upon which all N0^ emission guarantees
were granted. Initial retrofit data obtained in 1972, and retested in 1974
and 1976, provided the only projection of expected N0X emissions with the new
burner until units equipped with the new system actually went into commercial
operation. The first of these New Source Performance Standards designs was
declared commercial late in 1975, the second and third in 1976, and at present
we have 26 operating pulverized-coal-fired units that were sold to meet the
New Source Performance Standards. Data obtained from 24 units tested correlate
extremely well with the projections generated by the Dual Register Burner
retrofit at the Gaston plant. Presently, we have 1800 burner years of experience
with the Dual Register Burner system. Modifications have been made to improve
primary air/coal mixing and decrease pressure drop in the primary air system,
improve the operability of the spin vanes, and change the manner In which
-------
inner air is introduced from a swirl type assembly to a streamline sleeve type
assembly. These functional improvements have been field tested both in test
tunnels and in actual operating units and indicate a further gain in stability
and N0X reduction. Presently, 14,237 MW of New Source Performance Standard
coal units supplied by Babcock & Wilcox are in service. Most of these units
have been fitted with a complete combustion system for low N0V, including the
Dual Register Burner, the Compartmented Windbox and larger furnaces (Figure 2).
PRESENT PRACTICE
The 1979 New Source Performance Standards promulgated on June 11, 1980, further
reduced NO emissions requirements for bituminous coal to 0.6 lbs of NO per
A X
million Btu input and for subbituminous coal to 0.5 lbs of NO^ per million Btu
input. In addition, most of the synthetic fuels coming from coal are limited
to 0.5 lbs NOv per million Btu input. These new reduced levels and the proposed
A
research goals for 1985 implementation pose new challenges for the designers,
manufacturers, and suppliers of low NO combustion systems. Figure 3 indicates
A
the NO^. emissions measured on all of our bituminous-fired steam generators
tested which were equipped with the low NO^ combustion system and indicates
how their emissions relate to the New Source Performance Standards of 1979.
The majority of units are meeting the new 0.6 level even though they were
designed to meet the old 0.7 level. However, ndesigns must be developed to
provide some operating margin between the regulated limit of NO^ emissions and
the actual expected level of N0V emissions and insure that some operational
flexibility is available. Figure 4 indicates our experience with units equipped
to meet New Source Performance Standards while firing subbituminous coal and
compares those levels to the new regulations. In all cases except one, units
designed to meet the 0.7 former standard are meeting or exceeding the new
standard of 0.5. The one exception, operating at high levels of excess air,
does exceed the design standard of 0.7. We anticipate a further tightening of
the subbituminous New Source Performance Standards and, once again, are developing
systems to further reduce NO^ emissions on these fuels.
74
-------
FUTURE WORK
Previous efforts successfully resulted in NO reductions of 50 percent
A
from uncontrolled emission levels. Future development (Figure 5) currently
under way is expected to result in 85 percent reductions in NO^ emissions
from uncontrolled levels. The most developed of these systems, the Low
N0X Combustor (LNCS), conceived by B&W and funded jointly by EPRI and B&W,
is a first step in the second generation combustion modification approach
to N0X reduction. The LNCS is a deep staging approach with stoichiometry
held at the 65 to 75 percent range in the flame zone. The industry has
been and will continue to be concerned with applying two-stage concepts to
pulverized-coal containing high iron and greater than 2 percent sulfur
2
levels. Concern within B&W is based upon many years of witnessing uncontrolled
furnace corrosion resulting from localized reducing atmospheres. Our
position is that coals with high iron and sulfur are not suitable for
classical two-stage combustion In the generally accepted method where a
portion of the secondary furnace is subjected to reducing conditions and,
thereby, potential corrosion. The LNCS approach is to isolate the reducing
zone in small, controlled primary furnaces that could be either separately
cooled to maintain low metal temperatures, made of a more corrosion resistant
material, or designed to be replaceable. During the Second EPRI N0V
1
Control Technology Seminar in 1978, we reported on results of the LNCS 4 million
Btu Model. Now, the results of the 50 million Btu LNCS prototype are available.
Model and prototype data correlate, and the feasibility study, now being
finalized, will be submitted to EPRI at the end of October. One result
reported in the feasibility analysis is that the total surface area required
for heat transfer on the small individual primary furnaces, when added
together, exceeded 30 percent of the total furnace surface area. Thus,
the reducing zone isolation approach taken initially looses validity and
an alternative approach has been substituted. The venturi furnace (Figure 6)
is the present application of the Low N0X Combustor concept. Advantages
of that embodiment are that the components (such as ducts, windboxes,
etc.), are all standard components, and the circulation system is standard.
We can continue with known designs, such as the Dual Register Burner
system, simply modified for adaptation into the LNCS. The next step in
this development is to locate a field retrofit and obtain the associated
external funding to cover the costs.
75
-------
Our next step in low NO combustion modification development work would be
A
the coupling of the "planetary burner concept," advanced by Babcock-Hitachi K.K.
under CRIEPI funding, within the B&W LNCS. The overall stoichiometry of
the combined concept is limited to approximately 70 percent in the primary
furnace, but individual burner level stoichiometry is varied to achieve
4
the planetary concept.
The lowest rows of burners would operate at a stoichiometry of 0.8 to
0.85; the highest row of fuel input nozzles would operate at less than
0.50. The flames of the relatively fuel-lean lower levels pass through
the extremely fuel-ruch reducing flame. Laboratory-scale demonstrations have
indicated that N0^ formed in the lowest zone is decomposed as it passes through
the reducing zone, with radical products such as NH, CH present reduced to
^2' anc* ^2^' A-b°ve primary combustion furnace, the secondary air
ports increase stoichiometry to the 117 to 120 percent range required for
total combustion of the fuel. This approach is similar to work reported by
M. Heap, L. Muzio, and J. Beer, on char and fuel rich flame reduction."'
Our emission goal for joining these two concepts is 110 ppm on coal. That
combined concept has been tested recently on a full size utility unit located
in New Mexico and initial data indicates promising reductions. Once again,
the advantage of this type of combustion modification is that all of the
components, circulation, and air duct configurations are standard and well-
known to us as boiler designer and manufacturer and to our users. The area
of greatest concern to us as designers is the mixing requirements at the
secondary air introduction point above the primary combustion furnace. Our
extensive experience in gas tempering port design, and the subsequent mixing
in the furnace, used extensively to control furnace exit gas temperatures
provides an excellent data base for us to utilize for the secondary air/flue
gas mixing problem.
When only seven units had been tested in the field, we had excellent fuel
nitrogen correlation with the resulting NO^ emissions and reported it in
1978. However, the more units we tested, the more varied were the nitrogens
in those fuels and the more it became obvious that we were not able to
correlate NO emissions with nitrogen in the fuel to any great extent.
A
76
-------
To further reduce NO emissions with combustion modifications, we must
better understand the fuel-nitrogen transformation. B&W has funded a program at
its Alliance Research Center to further investigate various nitrogen level
coals in an attempt to understand the parameters that influence percent
conversion and continue to follow the work of Dave Pershing at the University
of Utah (Figure 7).
In the event that New Source Performance Standards are applied to existing
units in localized non attainment areas, our successful retrofit on the
E. C. Gaston plant indicates that it is possible to retrofit a combustion
modification. Combustion modifications for NO reduction continue to be
X
both fuel and site specific with each case requiring study. What modification
was reasonable at the Gaston plant with the Dual Register Burner system
might not be reasonable at another plant with a different fuel. As an
alternative approach to reduce NO^ resulting from combustion modifications,
B&W is under contract to the EPA to develop and apply their Distributed
Mixing Burner concept to actual burner hardware which can be retrofitted
to a U.S. utility unit."*
CYCLONE FURNACES
Cyclone units that have gone into service recently, but were not subjected
to the New Source Performance Standards because none had been promulgated
for North Dakota lignites, have achieved N0V emission levels lower than
those presently established by the New Source Performance Standards for
Northern lignites. One new cyclone unit firing bituminous coal entered
service during 1979 under the exemption granted to units firing at least
25 percent mine waste along with their base coal. The unit has been in
operation for approximately 15 months. NO levels have not been measured
because the client is complying with the requirement for using 25 percent
mine waste and therefore, is exempt from the NSPS. One other 450 MW
cyclone unit is scheduled for initial service early in 1981, firing a
North Dakota lignite. Retrofits to existing units firing typical cyclone
low fusion coal are not practical. The two-stage approach has been field
demonstrated both at the Board of Public Utilities, Kansas City, and at
Basin Electric's Leland Olds Unit #2 and showed significant NO reduction.
X
-------
In the North Dakota lignite case, Basin Electric required extensive amounts
of supplementary fuel to maintain flame temperatures high enough
to insure tapping of the slag. The Board of Public Utilities experiment
was successful while the Utility fired natural gas with N0X reductions of
50 percent obtained under staged conditions. However, the base fuel for
that Unit and the majority of all our cyclone furnaces utilize a fuel
containing sulfur above 3 percent and iron above 18 percent in the ash.
The combination of these elements in a molten slag pool, under reducing
conditions, will result in immediate catastrophic loss of tube metal,
subsequent to failures, and molten slag tapping through the floor of the
cyclone and the steam generator. The application of two-stage techniques
to cyclones firing typical cyclone suitable coal is, of course, possible
and NO reductions would result; but the effects in an extremely short
A
time would be catastrophic and we do not recommend the applications of two
staging to cyclone furnaces (Figure 8).
The number of B&W New Source Performance Standards units firing oil in service
in the United States is 17 and represent 6343 MW. Only one unit is presently
under construction and none have been sold for domestic siting since 1974.
The approach that has been utilized since 1959 involves classic two
staging, gas recirculation through the windbox to reduce flame temperature.
and BOOS (burners out of service) to optimize the staging and reduce the
excess air levels required to maintain a clean smokstack. Initially,
we investigated the effect of atomizer (mixer) design upon N0X generation
and, as expected, the atomizer producing the largest mean particle sizes produces
the lowest NO . The design approach used for the Dual Register Coal
X
Burner indicated that reduced peak flame temperatures and reduced turbulence
are important criteria to incorporate in any combustion system to reduce
N0X (Figure 9).
The parallel flow burner was developed for oil and gas firing on New
Source Performance Standards units because it did not depend upon turbulence
to enhance the mixing and obtain acceptable burnout. The philosophy
behind its design is streamline development and slow mixing coupled with
better atomization. The independence of swirl requirements in this
design make it possible to operate at extremely low excess airs over the
78
-------
load range than had been possible with a swirl register assembly. There
are approximately 2,000 burner years' experience with this approach,
and when coupled with either gas recirculation or two staging, proves
to be adequate to meet the New Source Performance Standards for both oil
and gas.
In 1975, the Dual Register Oil Burner was developed and retrofitted to the
Mandalay Unit 1 of the Southern California Edison system. Prior to the
retrofit, utilizing two-stage combustion, it was possible to reach 340 ppm
NO^ emissions on 0.3 percent ^ at 2 percent O2. Following the Dual
Register Oil Burner retrofit and utilizing two-stage combustion, N0V could
A
be reduced to 210 ppm. It appeared possible to obtain up to a 35 percent
reduction resulting simply from the combustion system utilized. The data also
indicated no reduction at all at times and, therefore, the value of the retrofit
and data are questioned. Additional retrofits were made at five Florida
Power Corporation plants. Florida Power Corporation's concern centered
around dust emissions and meeting the Performance Standards in Florida for
0.1 lbs of dust per million Btu input. The retrofitted burners met that
level. We were permitted to test for NO emissions in 1980 and found
X.
levels of 200 ppm. No baseline NO levels are available. This level was
A
obtained with full stoichiometric air to the burners (no staging, BOOS,
or gas recirculation). Fuel nitrogens averaged 0.3 percent during the
test series.
The first steam generator initially designed for the Dual Register Oil
Burner is owned by Hawaiian Electric, designated as Kahe Unit 6, rated
for 140 MW and currently in the prestart-up, stretchout period. Kahe
Unit 5 is a duplicate unit which utilizes N0V ports and circular (high
turbulence) burners to meet an old state NO emission standard of 300 ppm,
A
in effect at the time of its construction. Kahe Unit 6 is required to
meet the federal New Source Performance Standards, and has been supplied
with the Dual Register Oil Burner and no NO ports. We expect to better
A
the NSPS for oil on this unit (Figure 10).
79
-------
Babcock-Hitachi has developed the Dual Register Oil Burner one step further
by injecting gas recirculation, undiluted, into the primary flame zone.
They refer to this device as the Primary Gas Dual Register Oil and Gas
Burner. This combustor, when coupled with staging and gas recirculation
in the secondary air, has resulted in retrofit, NO emission levels as
X
great as 85 percent below uncontrolled levels. Application of this combustor,
coupled with a primary combustion furnace (LNCS) and operating in the
planetary mode similar to what is presently under test for pulverized
coal, will result in even lower commercially obtainable NO emissions than
A
presently possible (Figure 11).
Research and Development plans for the 1980s will concentrate in three
major NO areas. The first area is that of the fuel nitrogen correlations
which govern the NO conversions on both solid fuels and liquid fuels. The
second major area is to continue our successful efforts in combustion
modification, retrofitting the LNCS and demonstrating total unit performance,
and retrofitting the Distributed Mixing Burner and demonstrating total
unit performance. The third area concerns Synthetic Fuels and includes
SRC I and II, coal/oil mixtures, coal/water mixtures, chars, and other
liquifaction products from coals. We have gained extensive experience with
the SRC I and II fuels under EPRI and DOE funding for both the field
retrofit portion and laboratory characterizations. In addition to the N0V
emissions, the burning, fouling, slagging and handling characteristics of
these fuels must be classified to allow us to design combustion systems to
handle them.
SUMMARY
Results from the 24 pulverized coal units tested to date show that the
Dual Register Burner/Compartmented Windbox furnace system is an effective
tool for NO emission control. NO levels 40 to 60 percent lower than
X A
those achievable with the high turbulance circular burner system are
obtained through limited turbulence combustion. At the same time, carbon
utilization has been maintained at levels comparable to those obtained
80
-------
utilizing a high turbulence circular burner. Actual data collected for
both fuel classes demonstrate NO emission levels to be within the 1978
X
NO^ limits established by the EPA New Source Performance Standards.
Additional work in the area of combustion modifications, fuel nitrogen
correlations and new fuel investigations must be continued to meet the
anticipated tighter NO emission limitations expected over the next six
A
years.
In the short term, through 1985, it is anticipated that we will have
applied our Dual Register Burner/Compartmented Windbox furnace system to
150 utility steam generators with an approximated capacity of 67,500 MW.
Typically, these units have generated half the NO they would have if the
A
New Source Performance Standards of 1971 had not Implemented emission
controls. We are prepared with proven combustion modification retrofits to
apply to units in non attainment areas on a base by base basis and plan to
continue working with both EPRI and EPA to develop our lower NO concepts
for commercial applications from combustion modifications in the late 1980s.
81
-------
REFERENCES
1. Barsin, J. A. Pulverized Coal Firing NO^ Control. 2nd EPRI
N0V Control Technology Seminar. Denver, Colorado. November 8 & 9, 1978.
A
2. Manny, E. H. and P. S. Natanson. Fireside Corrison and NO^ Emission
Tests on Coal-Fired Utility Boilers. Exxon Research and Engineering
Company. In: Proceedings of the Joint Symposium on Stationary N0y
Control. October 1980.
3. Johnson, S. A. and T. M. Sommer. Commercial Evaluation of a Low N0y
Combustion System as Applied to Coal-Fired Utility Boilers. The Babcock
and Wilcox Company. In: Proceedings of the Joint Symposium on
Stationary NO Control. October 1980.
X
4. Babcock-Hitachi Internal Information Exchanges. 1979 and 1980.
5. Campobenedetto, E. J. Field Evaluation of Low NO Coal Burners
for Utility Boilers. The Babcock & Wilcox Company. In: Proceedings
of the Joint Symposium on Stationary NO Control. October 1980.
X
6. Song, Y. H., J. M. Beer and A. R. Sarofin. Fate of Fuel Nitrogen During
Pyrolysis and Oxidation. M.I.T. In: Proceedings of the Second
Stationary Source Symposium. July 1977.
7. Chen, S.L., M. P. Heap, D. W. Pershing, R. K. Nihart and D. P. Rees.
Fate of Coal Nitrogen During Combustion. In: Proceedings of the
Joint Symposium on Stationary Combustion N0X Control. October 1980.
82
-------
8
Compartmented
windbox
Typical burner
Burner secondary air
control dampers
Burner secondary
air foils
Furnace
observation doors
Dual Register Burner/Compartmented Windbox system
Figure 1
-------
Dual Register Burner
Figure 2
84
-------
CO
cn
lb N02/106 Btu
l.Or
Circular
burner
EPA NOx
emission limit
Dual
register
burner
240 250 250 333 460 472 550 630 650
Unit capacity, MW
Comparison of NOx emissions from circular burners
and Dual Register Burners with bituminous coal
Figure 3
-------
lb N02/106 Btu
1.0 r
0.8
oo
cn
0.6
0.4
0.2
Circular
burner
EPA NOx
emission limit
Dual
register
burner
90 330 470 470 550 550 550 575 580 600 650 675 700 700 700
Unit capacity, MW
Comparison of NOx emissions from circular burners
and Dual Register Burners with subbituminous coal
Figure 4
-------
Relative
NOx scale
00
¦*>1
Percent
Circular Dual
burner
LNCS
Planetary
and
LNCS
B&W low NOx (coal) development
Figure 5
-------
V
\
Secondary
Furnace
0 = 1.18
Primary
Furnace
£=0.6-0.8
Low NOx combustion system concept
Figure 6
88
-------
500
NOx, ppm
at 3% 02
400
300
200
Fuel N = 1.0 lb/106 Btu /
/ Fuel N = 0.7 lb/10'
200 300 400 500 600
HA/Sc, KBtu/hr-ft2
Fuel N2 correlation with NOx
Figure 7
89
-------
Ng/J NOx (ppm at 3% 02)
l£>
o
Pre NSPS firing
high sulfur
subbit. coal
0.8 lb N0x/106 Btu NSPS
for wet bottom units
firing northern lignites
450 450
MW MW
Unit
Cyclone furnace NOx levels
Figure 8
-------
Y-JET
T-JET
RETURN FLOW
BURNER: CIRCULAR
FUEL: SRC FUEL OIL
LOAD: FULL
J L
2 3
% 02 @ BURNER
Influence of fuel atomizer design on NOx emissions
Figure 9
91
-------
Final
Ignition and
Stabilizing Air
Combustion
Dual register oil burner
Figure 10
92
-------
Percent
100
75
Relative
NOx scale
50
VO
Co
25
Circular 2
burner stage
register
burner
2 Dual
stage register
gas 2 stage
recirculation gas
recirculation
thru
secondary
air
Dual register
2 stage
gas recirculation
in secondary air
and directly in
burner
(BHK-PG)
B&W low NOx (oil) development
Figure 11
-------
CURRENT DEVELOPMENTS IN LOW NOx FIRING SYSTEMS
By:
Tomozuchi Kawamura
Mitsubishi Heavy Industries, Ltd.
Tokyo, Japan
Donald J. Frey
Combustion Engineering, Inc.
Windsor, Connecticut, United States of America
94
-------
ABSTRACT
Low NOx firing systems for natural gas and oil were developed for
horizontal and tangential firing. The oil and natural gas "PM" firing
system uses fuel rich and fuel lean regions in combination with flue
gas recirculation to achieve low NOx emissions, the former region being
produced by a diffusion flame, while the latter is produced by a premixed
flame. The pulverized coal "SGR" and "LNCFS" tangential firing systems
achieve low NOx emissions by delaying mixing of the main combustion air
with the fuel. '
95
-------
INTRODUCTION
The reduction of NOx emissions from steam generators has become a
paramount issue in most industrialized nations, with the strictest govern-
mental regulations being in Japan and the USA (Table I). The local emission
requirements in both countries are often more stringent than the national
codes and vary considerably, depending in part on industrialization density
and climatic conditions.
Although tangential firing has the lowest proven NOx emissions,
Combustion Engineering, Inc. (C-E) and Mitsubishi Heavy Industries, Ltd.
(MHI) have been deeply involved in NOx control studies for over a decade.
Since the first oil embargo in 1973, C-E efforts have been directed primarily
at controlling NOx formation in coal combustion and eventually led to the
development of the "Low NOx Concentric Firing" (LNCF) concept. This system
has been successfully demonstrated in both the laboratory (1) and the field
(2) and is scheduled for rigorous testing in a large utility boiler, under
contract with EPA (3). The results of these studies will be released following
completion of additional testing in the utility units in mid-1981.
MHI first developed NOx control technology for oil and gas firing and
the principles of this technology were proven with an impressive array of
both new and retrofit applications. The continued price spiraling of fuel
oil has altered the fuel market structure in Japan and the shift to coal as
a fuel for steam generation has been increasing in recent years. Fortunately,
the NOx control technology developed by MHI for gas and oil was found to
be applicable to coal and the first commercial applications of these techniques
are being evaluated currently.
C-E has obtained a license to apply the MHI NOx control techniques on
steam generating equipment in the USA and Canada. The license covers two
basic methods of reducing NOx emissions: control of NOx formation during
combustion and removal of NOx from the flue gas following the completion of
combustion. The latter system is described in a companion paper, "The
Development of a Catalytic Reduction System for Coal-Fired Steam Generators" (4).
96
-------
This paper will discuss the MHI method of controlling NOx generation
during combustion and will describe the "PM" burner as originally developed
for oil and gas, the "SGR" burner for coal, and finally, the PM burner for
coal.
OIL AND GAS FIRING PM BURNER
PRINCIPLE OF OIL AND GAS FIRING PM BURNER
Recognizing that the "diffusion flame combustion system," employed in
almost all existing oil-fired boilers, has a limitation on NOx production
performance, MHI produced the PM burner, based on the "premix flame
combustion theory" generally used in designing rockets. The marked reduction
of NOx emission demonstrated by this burner has been recognized by both
industry and academia and the operational results on actual boilers has been
A
excellent.
Here is the principle upon which this PM burner was developed. Generally
speaking, there are two types of flames, diffusion flame and premixed
flame (Fig. 1). The former is the flame that is produced when the in-
jected fuel diffuses and burns while mixing with surrounding air. With
this type of flame, NOx production decreases as air supply decreases and
increases with increasing air. A premixed flame is one produced when the
fuel is evenly mixed with air prior to ignition. With this type of flame,
NOx production decreases with excessive air supply as well as with inadequate
air supply. Combining the characteristics of premixing with "off-set"
combustion, MHI developed the "off-set premixed flame theory." The PM
burner is the product of a study based on this theory.
The concept of the PM burner is shown in Fig. 2. Combustion occurs
in two different zones of the fuel-air mixture: in the fuel-rich mixture
(NOx value at cp and in the fuel-lean mixture (NOx value at C£) with the
same total excess air ratio "X" as that for conventional burners. Thus,
the average NOx value, point C (the mean value of C^ and C2) is markedly
lower than that obtainable with conventional firing techniques. This is
*The 1977 prize of the Japan Society of Mechanical Engineers for excellent
technical achievements was awarded to MHI because of this burner.
-------
the case with gaseous fuels, i.e., they are burned under premixed con-
ditions in both fuel-rich and fuel-lean mixtures, so the overall NOx
value falls at point C.
With liquified fuels, however, burning occurs under diffused flame
conditions in the fuel-rich mixture in order to maintain a stable flame.
Thus, the overall NOx value is shown as Point C', the mean value of C^'
and C^. The characteristically stable ignition associated with this PM
burner provides the additional benefit of more flexible operation than
conventional burners over a wide range of burner loads.
OIL-FIRED PM BURNER FOR TANGENTIAL FIRING
The corner assembly of a tangential firing system consists of a number
of fuel nozzles stacked one upon another, the number depending on the
total heat input to the unit. A typical fuel cell for the PM burner is
shown in Fig. 3.
A specially designed atomizer produces two oil spray patterns; an
inner conical spray for fuel-rich combustion and an outer "umbrella"
partial spray for fuel-lean combustion. In order to control the oxygen
atmosphere in the fuel-rich combustion zone, separate gas recirculation
(SGR) is injected through nozzles on either side of the oil compartment.
The outer (fuel-lean) oil is introduced through widely spaced atomizer
ports, which permit passage of the SGR.
Thus, the inner spray ignites at the diffuser and burning occurs
simultaneously with the mixing of fuel and air. The process is one of
diffused combustion and corresponds to point C', in Fig. 2. Ignition
of the outer spray is delayed and takes place after mixing with air
from the auxiliary compartment. The process is one of premix combustion
and corresponds to point in Fig. 2. The average NOx value is then
at point C'.
Figure 4 shows the effects of NOx production of fuel nitrogen
content, overfire air (OFA), flue gas premixed with total combustion air
(GM), flue gas injection through SGR compartments (SGR), and total excess
air. The effect of SGR on NOx production with various fuel oils is
shown in Fig. 5.
98
-------
OIL FIRED ROPM BURNER FOR HORIZONTAL FIRING
Figure 6 illustrates a typical structure of the oil fired ROPM
burner assembly for horizontal-fired applications. The oil fired ROPM
burner is composed of a fuel-rich oil spray, an air-rich oil spray,
and intervening SGR.
Only one oil gun is provided at the center of the inside air passage.
As was the case with the PM burner, the fuel oil is sprayed from the
tip of the oil gun at two different spray angles forming two concentric
hollow cones, providing an inner fuel-rich combustion zone and an
outer fuel-lean zone. The NOx level vs. fuel N and NOx vs. SGR ratio
of the oil fired ROPM burner are shown in Figs. 7 and 8.
GAS FIRED PM BURNER
Figure 9 illustrates a typical gas-fired cell of the PM burner
assembly installed in a tangentially fired windbox. The cell consists of
one fuel-rich nozzle and two air-rich nozzles.
The SGR nozzles are not installed since both the fuel-rich and
fuel-lean nozzles produce premixed and almost straight flames and there
is very little interference between these flames. If the boiler employs
gas recirculation for furnace outlet gas temperature (FOT) control,
the GR can be admitted between fuel cells, rather than through the
furnace bottom, for more effective NOx control. The NOx level vs.
GM ratio of the gas fired PM burner is shown in Fig. 10.
GAS FIRED ROPM BURNER
Figure 11 illustrates a typical structure of the gas fired ROPM
burner assembly for horizontal firing, consisting of one fuel-rich gas
burner and one fuel-lean burner. The air passage is radially divided
into two sections; the fuel gas and air supplied through the central
or internal passage produces the fuel-rich flame and the fuel gas and
air supplied through the outside circular passage produces the fuel-
lean flame.
An oil gun is shown in the center of the burner, but this is of
a conventional design and is used only for auxiliary purposes, not for
low NOx firing of oil. The NOx level vs. GM ratio of the gas fired
ROPM burner is shown in Fig. 12.
99
-------
COMMERCIAL APPLICATION OF OIL & GAS BURNERS
R.ecognizing the benefits and practical features of the PM burner,
Tokyo Electric Power Company, Inc. (TEPCO) adapted the PM burner to its
Anegasaki No. 5 unit (tangential gas-fired 600-MW unit) in 1977. The
operation was excellent and the unit achieves very low NOx levels over
the entire operating range of the boiler.
Table II is a list of operating converted boilers and several new
contracts for both gas and oil fired PM burners.
As of this writing, a total of 25 boilers representing all kinds
of firing—gas and oil, tangential and horizontal—have been put into
commercial operation with PM burners. Selected data on four of these
boilers is shown in Table III. The following facts were obtained from
these installations:
(1) The theory of "Off-set Premix" has been proven in commercial
installations and the operation of PM burners lends itself to
industry practices.
(2) Low NOx emission levels are obtainable over a wide range of
boiler loads.
(3) Combustion is extremely stable and the flames are easily
detectable over the entire range of boiler loads.
Figure 13 shows the lowest NOx emission levels obtained in four actual
boilers (same plants as for Table II). The chart compares PM burner
performance with conventional burners.
COAL FIRED SGR BURNER FOR TANGENTIAL APPLICATIONS
STRUCTURES AND FUNCTION
Figure 14 shows a conventional coal tilting tangential firing cell,
which consists of a coal compartment and two auxiliary air compartments.
Coal and primary air are introduced through the center of the coal
nozzle and some secondary air (termed "fuel air") passes through the
annulus around the coal nozzle. The remainder of the secondary air
(termed "auxiliary air") is introduced through separate compartments
above and below the coal. Although the coal and air are admitted in
100
-------
parallel streams, secondary air is continually introduced into the coal
stream, both before and after ignition, and there is no clearly defined
primary combustion zone.
The conventional burner has been redesigned to create separate
primary and secondary combustion zones and has been named the "SGR
burner" (Fig. 15).
The SGR burner has three features distinguishing it from the con-
ventional burner:
1) Gas recirculation is introduced through SGR compartments
above and below the coal nozzle.
9
2) The auxiliary air compartments are farther from the coal
compartment.
3) The coal nozzle has a divergent tip, which acts as a flame
holder for maintaining ignition close to the nozzle.
The first two features maintain a reducing atmosphere at the nozzle
during primary combustion by minimizing the infusion of secondary air
into the coal stream. The third feature assures that devolatization
and improved combustion of the coal occurs within this fuel-rich
primary combustion zone.
TEST RESULTS WITH SGR BURNER
MHI has two coal-fired test furnaces at its Nagasaki research
laboratory. The smaller furnace is water cooled, of circular wall,
steel construction with refractory lining and rated at 0.5 TPH. The
larger furnace is similar in construction and rated at 4 TPH.
Two SGR test burners were constructed and tested; the first with
a capacity of 525 kg/h was tested in the small furnace and the other,
rated at 3000 kg/h, was tested in the larger furnace. Both were of a
design similar to that shown in Fig. 15.
The test results are shown in Figs. 16 through 21. Analysis of
coals indicated in these figures is shown in Table IV.
(1) Effect of excess air on NOx is shown in Figs. 16 and 17.
As with conventional burners, decreasing excess air is an
effective method of reducing NOx.
(2) Effect of OFA on NOx is shown in Figs. 18 and 19. Again,
its effect is characteristic of most other burner designs.
101
-------
(3) Combined effect of OFA and SGR on NOx is shown in Figs. 20 and
21. The concept of the SGR ports was to impede the ingress
of secondary auxiliary air into the primary combustion zone
in which devolatilization and combustion occurs. These tests
indicate that the effectiveness of SGR varies from coal to
coal. In those cases where it is ineffective, minimum SGR
flow is introduced for cooling these ports without introducing
air, which would increase NOx.
CONFIRMATION OF BURNER SCALE-UP PERFORMANCE
MH1 has had considerable experience relating test furnace performance
to actual field performance with converted burners. This experience
indicates that if the following relationship is maintained between
the burner firing rate (Q) and the burner height (H) , the NOx
emissions from the burners will be the same:
N
H=AQ where A is a constant
Figure 22 shows the results of such an investigation and compares
the NOx production of a small burner in this test facility with a
full sized burner in the Takasago No. 1 unit, firing the same coal.
The burner design for both coals was conventional tangential type.
COMMERCIAL APPLICATION OF COAL-FIRED SGR BURNER
The first commercial application of the SGR burner is Matsushima
1 and 2 of Electric Power Development Co. (EPDC). These are duplicate
500-MW boilers and the first large coal-fired units ordered after a
decade of oil and gas utilization.
Since Matsushima 1 has just started its trial operation, we
do not yet have data on the performance of its SGR burners. We plan
to have this information within a few months.
Figure 23 shows the relationship between burner capacity (Q)
and burner height (H) for the same NOx production. This is based
on the formula discussed previously and shows control data points
taken from the test furnaces and the Takasago unit. From this relation-
ship, it is expected that the 1830 mm burner height for Matsushima
will achieve the same NOx emissions as demonstrated for that coal in
both small and large test furnaces.
An analysis of the coals tested in the two test furnaces thus far
are shown in Table IV.
102
-------
COAL FIRED PH BURNER
During the study of the SGR coal burner, which is the first design
series in the low NOx coal burner program, MHI collected data pertaining
to the relation between the primary air/coal ratio and NOx formation
for the purpose of determining possible further improvements in low
NOx burners.
Data from the test furnaces is plotted in Fig. 24 and one can
see that when the primary air/coal ratio is in the range from 1-3,
the NOx value decreases as this ratio is reduced with every kind of
coal tested, although the slopes of the curves vary.
On the other hand, C-E had acquired data on the relationship
between the primary air/coal ratio and NOx formation at ratios ranging
from 3 to 8, which is higher than the usual ratios. Some of this data
is shown in Fig. 25, including that of a Japanese coal and an Australian
coal which had been sent to C-E for testing.
Figure 25 presents a quite different trend from that of Fig. 24,
in that the NOx value decreases as the primary air increases when the
primary air/coal ratio ranges between 2 and 7 or 8. The test results
in these two cases are apparently contradictory, but may be explained
as follows.
By means of the explanation of the principle of the PM burner,
the principle of the SGR burner will become clear as well.
PRINCIPLE OF NOx REDUCTION
Generally, the pulverized coal combustion zone is theoretically
divided into two sub-zones as shown in Fig. 26; one is the primary
combustion zone where pulverized coal carried by primary air burns
while mixing with that primary air, and the other is the secondary
combustion zone where the remainder of the coal burns while mixing
diffusing secondary air injected from the auxiliary air compartments.
However, with conventional burners the secondary air starts to mix
with the primary air/coal stream as soon as they are injected into the
durnace and distinct primary and secondary combustion zones do not exist.
103
-------
Thus, volatile matter burn^ with high excess air instead of burning
only with the primary air in the primary combustion zone.
Because of this, the combustion process and NOx reducing theory
described here can not be applied to conventional burners. In other
words, MHI's Low NOx burners are so designed that they can establish
distinctly separate primary and secondary combustion zones.
The quantity of coal that burns in the primary combustion zone
is proportional to the ratio of primary air to coal and the quantity
of coal that burns in the secondary combustion zone is the remainder
of the coal. Thus, the amount of NOx generated from the combustion
of pulverized coal can be expressed as a sum of that generated in the
primary and that in the secondary combustion zones, as shown in Fig. 27.
(1) The formation of NOx in the primary combustion zone
({N0x}p) in Fig. 27.
The primary air/coal ratio of 3-4 approximately corresponds to
the air quantity theoretically required for combustion of volatile
matter in coal. In the region where this ratio is within the range
of 3-4, primary air is almost all consumed by the combustion of volatile
matter. At lower ratios, oxygen availability is decreased, the
combustion of volatile matter is retarded, and the conversion of N
in volatile matter to NOx is reduced. Moreover, as a result of in-
complete combustion, the quantity of unburned gas from the volatile
matter entering the secondary combustion zone is increase.
A primary air/coal ratio of 7-8 corresponds to the air quantity
theoretically required for complete combustion of the coal. Therefore,
in the region where the primary air/coal ratio is between 3-4 and
7-8, Oj is insufficient for the complete combustion of fixed carbon
but the conversion of N in volatile matter to NOx becomes more active
as the ratio increases. However, the reducing action on NOx formation
of substances such as HC, NH^, HCN etc., produced from the fixed
carbon, also becomes more active, with the net result that the NOx
valued at the outlet of the primary combustion zone eventually drops
as the primary air/coal ratio increases.
104
-------
Primary air/coal rations greater than 7-8 result in excess air
combustion, negating the NOx reducing action and resulting in a steep
rise of NOx value.
Thus, the NOx formed in the primary combustion zone has the
characteristics as expressed by the solid lines in the chart of (NOx)
P
in Fig. 27.
(2) The formation of NOx in the secondary combustion zone
({NOxIs in Fig. 27):
In the region where the primary air/coal ratio is less than 3-4, a
large quantity of unburned gas from the primary combustion zone burns
with the char As the secondary air diffuses and mixes with them in
the secondary combustion zone. The quantity of unburned gas increases
as the primary air/coal ratio is throttled further, and causes relative
retardation of diffusion and mixing of secondary air and a consequent
drop of the conversion rate of N in unburned gas and char to NOx. Thus,
the formation of NOx varies little with the change of the primary air/
coal ratio, as shown by the heavy solid line in the chart.
In the region where the primary air/coal ratio is between 3-4
and 7-8, the percent of coal that burns in the primary combustion zone
increases as the primary air/coal ratio increases. Consequently,
the percent of the coal that burns in the secondary combustion zone
decreases and the formation of NOx in this zone decreases.
The sum of (NOx)p from the primary combustion zone and (NOx)s
from the secondary combustion zone is the total of the NOx produced
from the combustion of pulverized coal. As shown in the bottom chart
in Fig. 27, NOx increases in the region where the primary air/coal
ratio is between 0 and 3-4 and decreases in the region where that
ratio is between 3-4 and 7-8 and increases again when that ratio
exceeds 7-8.
PRINCIPLE OF THE COAL FIRED PM BURNER
It is a tested conclusion that with coal fired burners, the NOx
value is high in the region where the primary air/coal ratio is 3-4
and falls as the ratio deviates in either direction, increasing or
decreasing. This is a similar trend in NOx formation to that which
occurs with the premixed oil and gas fired burners when the air ratio
105
-------
is approximately 1 and falls as the ratio deviates from this point
in either direction.
In a direct-fired system, the pulverizer is directly coupled with
the firing system. The primary air/coal ratio is dictated by pulverized
drying and conveying requirements, and generally falls in the range
1.5-2.0. Pulverizers provided with spare capacity, can increase this
ratio to 3.0 or thereabout. This is undesirable in a normal firing
system since NOx increases as this ratio increases, as shown in Fig. 27.
Fortunately the relation between the NOx value and the primary air/
coal ratio in combustion of coal is similar to that between the NOx
value and the air ratio in premixed combustion of oil and gas. This
suggests the possibility of using the principle of fuel-rich and
fuel-lean with coal, as well, and this forms the basic principle of
the PM burner.
Referring again to the bottom chart in Fig. 27, the circled
points illustrate the low NOx burner. The coal-air mixutre from
the pulverizer can be divided into two streams; one with a ratio of
(fuel-rich) and the other C^ (fuel-lean). If these two streams
are permitted to fire independently, the NOx emission level for the
whole burner assembly will be the level corresponding to the point
Cq which is considerably lower than the point C'Q which represents
the NOx emission level obtainable if the entire coal/air stream burned
as a single flame in separate primary and secondary combustion zones,
as with the SGR burner. The structure of the coal fired PM burner
is shown in Fig. 28. The concentrating effect of coal to the outside
radius of the delivery pipe is used to make the separation between con-
centrated and dilute coal streams. Figure 29 shows comparison of NOx
emission of conventional burner, SGR burner and PM burner.
CONCLUSION
The PM burner for oil and gas, based on "Offset Premix Flame
Theory," successfully achieves dramatically lower NOx emission levels
than is possible with conventional burners. The most prominent feature
of the oil and gas PM burner is that it can greatly reduce NOx emission
while maintaining stable combustion. Moreover, it can reduce
106
-------
particulate matter emission as well and requires no additional effort
in operation and maintenance of the equipment. Neither catalyst nor
ammonia injection is needed. Thus, the advantages of the PM burner
may be summarized by stating that it can effect a marked reduction
in NOx emission economically, not only in new boiler applications
but in most retrofit applications, as well.
A number of oil and gas PM burners have already been successfully
applied to various types of boilers such as supercritical pressure, forced
and natural circulation, pressurized, and balanced draft, indoor and
outdoor types. The fact that all these burners have been operated
successfully and to the customer's satisfaction is proof that this
burner design is applicable to any type of utility boiler.
In the case of coal-firing, the SGR burner provides an effective
means of reducing NOx, but field confirmation of its performance must
await testing to the Matsushima until later this year.
We believe that the more recently tested coal-fired PM burner
will replace the SGR burner on future new boiler applications since
its potential for NOx reduction is much superior. The current status
of the low NOx burners, then, is shown in Fig. 30.
107
-------
REFERENCES
1. EPA Contract No. 68-02-1885
2. C-E Studies at Richmond Power & Light Company, Whitewater Unit No. 2
3. EPA Contract No. 68-02-3655.
4. Sengoku, T., Todo, Y., Yokoyama, N., and Howell, B. M., "The
Development of A Catalytic NOx Reduction System for Coal-Fired
Steam Generators,"Presented at the EPRI-EPA Joint Symposium on
Stationary Combustion NOx Control, Denver, Colorado, October 6-9,
1980; Published as Combustion Engineering publication TIS-6710,
Windsor, CT: Combustion Engineering, Inc., 1980
108
-------
OlffUttOft KAMI
r ¦™'-i
-
Fig. 1: Diffusion flame and premixed flame
IF AIR SUPPLY IS REOUCEO
BE LOW THIS LEVEL.
UNBURNEO COMBUSTIBLE
WILL BE FORMED
PREMIXED COMBUSTION
OtFFUSEO COMBUSTION
a-
< A
THEORETICAL
AIR RATIO
TOTAL OPERATING
EXCESS AIR (X)
1.0
FUEL • RICH
Fig. 2: Concept of the PM burner
109
-------
WINOBOX DAMPER
FUEL OIL
OIFFUSER CONE
WINOIOX
AUXILIARY COMPARTMENT
S6R COMPARTMENT
SCR
\ FUEL COMPARTMENT
\ AUXILIARY
Fig. 3: PM burner cell for firing oil in tangential firing system
LEGEND
Of A w
BMW
MR (X)
AIR
TE*»..{#C)
EX. 02(ft)
0
•
1
•
ZM
2
n
e
0
W
2f0
. 2
¦
is
B-10
21
2U
1-1.2
1M -
.o-—
• 1.1 U I)
FUCL NITROft EN, WT K
Fig. 4: Oil fired PM burner for tangential firing—fuel N vs. NOx emission
110
-------
160 —
LEGEND
FUEL
OFA(H)
GM (H)
AIR
TEMP. (#C)
e,02(*>
O
HEAVY OIL
(N-0.1K)
0
260
2.0
~
CRUDE OIL
(N - 0.03*}
0
0
250
2.0
0
HEAVY OIL '
(N - 0.03%) | 0
0
290
2.1
1 L 1
$0fl RATIO, *
Fig. 5: Oil fired PM burner for tangential firing—SGR ratio vs. NOx emission
(test furnace)
\ FLMtt ICANNtft
Structure of oil fired ROPM burner for horizontal firing
111
-------
290
IN
99
BO
TO
10
M
40
y* »
e
m
LEGEND
OFA(S)
IGR(X)
AIR TEMP. (°C>
e.o2<*>
~
0
0
2M
1.1
a ,
•
11
299
1.S
¦
If
20
2M
2.0
F-OCL N (WT X)
Fig. 7: Oil fired ROPM burner for horizontal firing—fuel' N vs. NOx emission
(test furnace)
LEGENO
FUEL
OFA (X)
GM(%>
AIR TEMP. (°C)
E„02 (W
0
HEAVY OIL
(N • 0.1 %)
3
3
MO
IS
~
CRUDE OIL
(N-0J3X)
3
3
210
16
SGH RATIO,*
Fig. 8: Oil fired ROPM burner for horizontal firing—SGR ratio vs. NOx emission
(test furnace)
112
-------
WIN080X DAMPER
x GAS FOR FUEL RICH
f-Jr COMBUSTION
(C)
GAS FOR f DEI - LEAN
COMBUSTION
m
FUEL RICH ^C3
GR tOniONAl)
FUEL GAS NOZZLE
Fig. 9: PM burner cell for firing gas in tangential firing system
Fig. 10: Gas fired PM burner for tangential firing—GM ratio vs. NOx emission
(test furnace)
113
-------
full II** FIAMI
fUll IMMFIAJH
. I f-
v'3l 44
' if.l!
fUlt 6*1 FOR full I
:3&=
-* ^ oil BUN
L#1/"'
L-if ISWITOR
if
J t'H]-.
Fig. 11: Structure of gaB fired ROPM burner for horizontal firing
2M
m
Ml
LEGEND
OM <%)
O
1
•
IB
AJH TEMP. (*C)
E.O,*}
m
2J
m
U
fuei-ih
ft* RATIO (M
Fig. 12: Gas fired ROPM burner for horizontal firing—GM ratio vs. NOx emission
114
-------
THERE IS NO OPERATIONAL RESULTS
OF FRONT FIRING CONVENTIONAL
GAS BURNER
STEP 3 « OFA ~ GM
49%
2IK a
PLANT 0
LNG
2ISMW
PLANT I
HEAVY OIL
IMHMf
Fig. 13: Operational results of PM burner in actual boiler (comparison of minimum
NOx before and after modification)
AUXILIARY AIR
FUEL AIR
Fig. 14: Configuration of conventional coal tangential firing nozzle
115
-------
AUXILIARY AIR
ex
3
SGR
FUEL AIR
COAL
___FUEL AIR
AUXILIARY AIR
Fig. 15: Configuration of SGR burner for coal tangential firing nozzle
116
-------
100
z
M
<
>
ae
z
e
«
at
<
LU
Z
cc
UNBURNED CARBON
500
O 80. AFRICAN COAL
No. 4
• JAPANESE COAL
No. I
400
I
IN
O
300
MS KG/H
o*
s
LOAD
GM(SCR)
0FA
PRIMARY
AIR/COAL
200
2.1
100
EXCESS 0,.%
Fig. 16: NOx with excess SGR test burner
117
-------
30
*
x'
tn
<
>
£ 20
600
i
fs, 300
o
£
o
* 200
¦Ah
9—---0—-—6-
~>r V-
co
I
I ! ) UNBURNEDCARBON
( 2 3 4 5 6 7 1 9
O JAPANESE COAL No. I
• AUSTRALIAN COAL No. 7
-f"!
LOAD
GMSGR
OFA
PRIMARY
AIR/COAL - 2.8
3000 KG/H
0%
MIN.3%
• ; ~,<$
I i
i
i
1 2 3 4 S 6 7 I «
EXCESS 0j,*
Fig. 17: NOx reduction with excess O —SGR: 3000 kg/h capacity
test burner•
118
-------
VO
_ M. AFRICAN COAL
O— ¦*.«
. CHMEK COAL
¦a.f
JAM* E* COAL
Hail
OFAXttRM
Fig. 18: NOx reduction with OFA—SGR
* 39
X
3
-J
s 2f
2 i«
r-"
CO
-o
o-
,
—.
-o
" —(
-o-
>¦—\
| UNBU
RNEO CARBON
200
100
- JAPANESE COAL
o " Na. S
# --• AUSTRALIAN COAL
No. 7
m
M
o
LOAD
CM (S6R)
PRIMARY
AIR/COAL
KG/H
£
M
e
M
4%
2.1
OFA % (SGR-0%)
Fig. 19: NOx reduction with OFA—SGR
-------
1
*
s
c
a
¦ - 0
CARBON
t_. -
t
I 8
ro
o
O — SO. AFRICAN COAL >¦ 4
O —CNMfS COALNc.1
COALIN.I
IMKfi/H
a n
S8RKI0FA-1IX)
Fig. 20: NOx reduction with SGR of 0FA=18%—SGR
* *
(
—
—
>
m*
Ik
S "
I (
CO
•o-
s
icr
o
IUMIHWED
J
CAMON
n___ JAPANESE COAL
W Nil
MM
— •
— AUSTRALIAN
Nt.7
L0A0
6M
C»°2
- 3M8KG/H
• K
* 4%
- OFA
PRIMARY
AIR/COAL
- m
- 2.1
1M(
• —
i ii a )i 4i
MR * fOFA - 2fK)
Fig. 21: NOx reduction with SGR at OFA-20% SGR
-------
NO, FORMATION FROM
TAKA1AGOII
(•UM)fR MtlGHT • 1100 ¦»>
N0a FORMATION FROM
TEST FURNACl
(lURNiH HII6MT • 700nm)
Fig. 22: Relationship between burner size and burner capacity for the same NOx
formation (conventional and tangential burner
k: CONSTANT CHAftACTlRWTIC
TO BIMNf RS
Tin IURNIR MR)
T»tT IIHHItR (COAVUTKMUl ¦MWIR)
IVMHR GAMSITT. • (TAB
Fig. 23: Relation between burner size and burner capacity for the same NOx
formation (conventional burner)
MR IURNIN COM RATIIM KS/N 1,0t OPI MM
O «0 AFRICAN COAIN*. 1
A AMTRAIIAM COAl Nr )
M AUffRAUAM CflilN*. I
| Q AUnRAllAM COAlNt. 4
@ 10 AFRICAN COAt *• t
9 JAAAMU COAlNc'l
MUMMY AM/COAL. Kl/M
Fig. 24: NOx primary air-coal (fuel rich side)
121
-------
AMTMUA**
Fig. 25: Effect of primary stage stoichiometry 011 NOx (fuel lean side C-E)
'ftlMAAV CQMIUSllOH/ONI If COttOARV COMIUttlON *0*f
I
I
OMC0N0ARV C0MIUICON 20NE
Ol • tOtlW Of]
0 COMUtTIOfc HATE
Fig. 26: Combustion model of pulverized coal firing
122
-------
NITROGEN COMPOUNOS FROM CHAR
NITROGEN COMPOUNDS
FROM V M
NITROGEN COMPOUNOS POTENTIALLY
CONVERTIBLE TO NO. IN PRIMARY
COMBUSTION ZONE
REDUCTION OF
CONVERSION RATE
Of NO, IV ACTIVE
MATTER FROM CHAR
NITROGEN COMPOUNOS
CONVERTIBLE TO NO. IN
SECONDARY ZONI
REOUCTION OF
CONVERSION RATE
mini
(NO J SOU
STOICHIOMETRIC RATIO
FOR V.M
STOICHIOMETRIC RATIO
FOR CHAR
PRIMARY AIR/COAl RATIO
Fig. 27: NOx formation characteristics with primary air/coal ratio
and conceptional figure of PM coal burner
123
-------
OIL
AUX2
SGR ~
FUEl
RICH
Oil
COAL
Fig. 28: Structure of coal-fired PM cell in tangential firing system
CONVENTIONAL FLAM!
PRIMARY AIR/COAL
LOW NO, M FLAME
I
s"
raMARV AIR/COAL RATIO
Fig. 29: Comparative NOx formation characteristics of coal-fired PM and
conventional flames
124
-------
FUEL FIRING SYSTEM N0X TARGET VALUE
FIRST APPLIED UNIT
COAL-TANGENTIAL
SGR BURNER SYSTEM
NOx 150 PPM
COAL FIRING PM
BURNER SYSTEM
NOx 100 PPM
FUTURE
PROJECTS
COMBINED CIRCULATION ¦ 500MW
OCT.. 1980
i—TANGENTIAL
OIL
^HORIZONTAL
CONTROLLED CIRCULATION • 265 MW
JULY, 1977
CONTROLLED CIRCULATION - 350 MW
MAR.. 1979
OIL FIRING ROPM BURNER SYSTEM
NO. 80 PPM
OIL FIRING PM BURNER SYSTEM
NO_ 70 PPM
GAS
TANGENTIAL
HORIZONTAL
GAS FIRING PM BURNER SYSTEM
COMBINED CIRCULATION - 680 MW
NOx 20 PPM
APR., 1977
GAS FIRING ROPM BURNER SYSTEM
CONTROLLED CIRCULATION - 265 MW
NOx 30 PPM
FEB., 1978
NOTE: (1) 1 ' ALREADY IN COMMERCIAL OPERATION
|l |j NOW UNDER CONSTRUCTION OF MANUFACTURING
(2) KIND OF FUEL COAL: AVERAGE JAPANESE BITUMINOUS COAL
OIL: HEAVY OIL (N 0.1K)
GAS: NATURAL GAS
(3) N0X PPM BASIS: 4% 02 FOR OIL, 6X 02 FOR GAS, 6S02 FOR COAL (DRY VOL)
Fig. 30: Development of PM burner system by MHI
125
-------
TABLE I
COMPARISON OF NATIONAL REGULATIONS ON NO EMISSION RATES IN JAPAN AND UNITED STATES
A
Gas Quantity
AHowable NOx Emission Rates for New Units
Rate Emitted
Japan (1977)
USA (1979)
Fuel
®10>NmVh
©MHSCF/h
® ppm at 3% 02
lb/10* Btu
ppm at 3% Oz
lb/10* Btu
Gas
>500
40-500
10- 40
< 10
>18,662
1493-18,662
373- 1493
< 373
68
113
146
169
0.08
0.14
0.18
0.21
161
0.2
OH
> 500
10-500
< 10
> 18,662
373-18,662
< 373
138
159
191
0.18
0.21
0.25
230
0.3
Coal
No stipulation
480
0.67
©360/0432
©0.5/©0.6
Notes:
©Normal cubic meters per hour (0#C and 760 mm Hg).
©Standard cubic feet per hour (60°F and 14.7 psia).
©Values are corrected to 3% 02 base for convenience in comparing with USA standards. Japanese values based on 4% 02 for oil,
5% 02 for gas, and 6% 02 for coal.
©Sub-bituminous coal and solids derived from coal.
©Most other coals.
-------
TABLE II
LIST OF UTILITY BOILERS EQUIPPED WITH MHI PM BURNER SYSTEM
Unit name
Unit
cap. (MW)
Boiler
type
Firing
Draft
Fuel/Atom izati on
Start of
operation
1
S
*£
c
3
I
z
Anegasaki #5, Tokyo Electric
Tobata Kyodo #4, Tobata Joint Elec.
Shin-Kokura #3, Kyushu Electric
Shin-Kokura #4, Kyushu Electric
Sodegaura #4, Tokyo Electric
600
375
600
600
1000
SC
CC
SC
SC
SC
T
T
T
T
T
P
B
P
P
P
LNG, LPG/-
LNG, BFG/-
LNG/-
LNG/-
LNG/-
APR., '77
FEB., '78
SEP., '78
JUN., '79
JUN., '79
a.
i
-------
TABLE III
PRINCIPAL DATA OF BOILERS EQUIPPED WITH VARIOUS TYPE OF PM BURNERS
Unit
ttam —
A
B
C
D
Burner type
Oil-PM
Oil-ROPM
Gas-PM
Gas-ROPM
Unit output (MW)
265
350
600
265
Fuel
Heavy oil
Heavy & crude oil
LNG
LNG & Naphtha
Firing system
Tangential firing
Horizontal firing
Tangential firing
Horizontal firing
Draft system
Balanced draft
Pressurized draft
Pressurized draft
Pressurized draft
Type of boiler
Controlled circulation,
twin furnace
Controlled circulation,
divided furnace
Supercritical
combined circulation,
divided furnace
Controlled circulation,
divided furnace
No. of burners
3 elevations*8 comers
=24
3 elevations*6 rows
=18
5 elevations*8 comers
=40
3 elevations*6 rows
=18
Capacity of one burner
(kcal/h)
32.86*10®
52.02*10®
38.93*10®
40.02*10®
Type of fuel
atomization
Mechanical atomization
—return system
Steam atomization
—
Steam atomization
-------
TABLE IV
ANALYSIS OF COALS TESTED
NO.
COAL
PROXIMATE ANALYSIS
ULTIMATE ANALYSIS
HHV
S.M.F.
kcal/kg
V.M.
%
F.C.
%
ASH
%
F.C./V.M.
N
DRY/DAF
%
0
DRY %
O/N
1
SOUTH AFRICAN
25.3
55.0
17.0
2.17
1.6 (1.93)
10.8
6.75
6420
2
CHINESE
28.1
57.8
10.7
2.06
0.8 (0.9)
11.7
14.63
7020
3
JAPANESE
33.8
37.0
26.8
1.09
0.9 (1.23)
7.9
8.78
5850
4
SOUTH AFRICAN
31.3
51.6
13.2
1.65
1.7 (1.96)
13.3
7.82
6540
5
CHINESE
29.8
48.8
18.4
1.64
1.03 (1.26)
12.0
11.65
6330
6
JAPANESE
43.9
37.7
12.7
0.88
1.1 (1.26)
14.5
13.18
6370
7
AUSTRALIAN
30.7
54.2
11.9
1.77
1.6 (1.82)
9.1
5.69
6890
8
JAPANESE
46.4
37.3
10.7
0.80
1.1 (1.23)
14.1
12.82
6510
-------
AN EVALUATION OF NOx EMISSIONS
FROM COAL-FIRED STEAM GENERATORS
By:
R. A. Lisauskas and J. J. Marshall
Riley Stoker Corporation
Worcester, Massachusetts 01613
130
-------
ABSTRACT
The design evolution of the Riley coal-fired TURBO Furnace and
Directional Flame Burner is reviewed. Burner aerodynamics are characterized
and the effectiveness of burner adjustments and staged combustion in
reducing NO^ emissions in this unique firing system are discussed. Field
test emissions data are presented and analyzed with respect to burner
operating variables. A decrease in NO^ emissions is observed as mixing of
fuel and air in the near burner zone is delayed. Further development of
directional flame and controlled mixing burners for coal-firing applications
is also discussed.
131
-------
INTRODUCTION
In a continuing effort to characterize the production of NO^ in
coal-fired Riley TURBO Furnaces, the Riley Stoker Corporation is obtaining
field data on operating units as part of the company's long-term NO^
Program. Since our report at the Second EPRI NO^ Control Technology
Seminar, the additional field data collected has been incorporated into the
data base from which a regression analysis is performed to aid in the
prediction of NO emissions from the TURBO Furnace. This model for
12
predicting NO^ was discussed in some detail in previous conferences '
and is only briefly mentioned here.
The additional data collected in the past 2 years includes
information on dry bottom TURBO Furnaces with overfired air. A great effort
has been placed on trying to quantify the effects of variation and
adjustments to the specific types of burner hardware unique to this firing
system. This includes components such as velocity control dampers,
adjustable secondary air vanes, and coal stream spreaders. Although the
Riley TURBO Furnace and its firing system have been detailed in the open
literature, a summary review is made here to reacquaint the reader.
The focal point of much of the latest effort has been to characterize
the burner aerodynamics which have the greatest effect on controlling the
evolution of fuel-bound nitrogen into NO. The aerodynamics of concern are
in the near burner zone, described in our previous paper as the primary zone
of the overall furnace model. Although the thermal fixation of nitrogen in
air contributes significantly to the NO emission, the conversion of
fuel-bound nitrogen is the prime contributor to the overall emission of NO
in the range of values evaluated here.
132
-------
TURBO FURNACE DESIGN
Furnace Design Features
The unique configuration of the Riley TURBO Furnace was first used in
the late 1940's. At that time, the difficulty concerned the combustion of a
petroleum byproduct known as fluidized coke. While this fuel had a high
calorific value, it was slow in burning and needed a longer furnace
residence time. Bending the front and rear walls inward, but not to the
extent of a roof- or arch-fired furnace, provided a cost-effective means of
furnishing a longer path for the combustion process compared to a straight
wall unit front-, rear- or opposed-fired. As illustrated in Figure 1, the
fuel and air admitted through the burners on both walls was inclined to the
center of the furnace. Combustion took place as the mixture curled down and
then up through the lower furnace.
The venturi shape formed by this tubing arrangement and the diffused
flame pattern provided by the Directional Flame Burner produced an even heat
distribution across the unit while the walls of the lower and upper furnace
remained comparatively free of ash or slag buildup.
Second generation Riley TURBO Furnaces were built primarily for
natural gas and residual oil firing as was dictated by the energy climate at
that time. The furnace had a flat floor formed by the water wall tubes.
With the ability of the burner to vary the direction of the air with respect
to the fuel flow, it brought the performance of these widely varying fuels
more in line with each other by bringing the furnace exit gas temperature on
each fuel closer together than it would be on a straight wall unit. Coal
was also burned in these flat bottom or slag tap type furnaces, but this was
usually limited to comparatively high slagging characteristic coals. A
typical coal-fired wet bottom TURBO Furnace of this type is shown in
Figure 2. For other types of coal, Riley still used straight wall-fired
units at that time.
During the mid- to late-1960's, many boilers were being built to fire
oil and gas. About this same time, Riley began to intensify its research in
the area of NO emissions. This work occurred into the early 1970's and
* 3 4 5 6
has been well documented. ' ' ' The work consistently showed that the
133
-------
baseline emission of NO^ was always lower in the TURBO Furnace than the
straight wall-fired units. To meet the modern needs of the industry and the
switch of the energy mix back to a coal base, the third generation Riley
TURBO Furnace was developed. This modern concept includes a dry hopper
bottom with a water-impounded ash receiver below the hopper throat. The
critical furnace design dimensions are shown in Figure 3.
A list of some of the Dry Bottom Riley TURBO Furnaces supplied for
coal firing is shown in Table I. These units were designed to fire a wide
range of coal types ranging from eastern bituminous to western
subbituminous. The typical analysis for a few of these coals in shown in
Table II. Figures 4 through 9 illustrate the variation of the sectional
side elevations for some of the units in Table I. As can be seen in these
figures, there are significant differences in each unit design and its
equipment for this one type of boiler. These differences are the results of
many factors. Some of the more important factors are the coal
characteristics such as higher heating value, slagging tendency,
grindability, and customer preferences.
Burner Design Features
The Directional Flame Burner (Figure 10) is used exclusively in the
Riley TURBO Furnace. This is a unique diffusion-type burner. These are
slow mix, nonswirl burners. The secondary air and primary air coal jets are
essentially parallel. Adjustments can be made to the directional secondary
air vanes to deflect the secondary air into or away from the primary
stream. A velocity control damper is provided to allow for varying the
secondary air velocity. An adjunct overfire air or staged combustion port
is dampered to adjust burner zone stoichiometry. With the combination of
these three airflow control systems in the burner, the degree of mixing in
the primary flame zone can be controlled over a wide range. A major effort
is being made to characterize the effect of these three design features on
N0x production in coal-fired TURBO Furnaces.
Although not adjustable on commercial burners, some modification has
been made for test purposes to utilize an adjustable coal spreader. This
allows variations of the axial relationship between the primary and
134
-------
secondary stream while also significantly changing the interaction point of
the flames from the front and rear wall burners.
All of these burner adjustment features not only impact NO^
evolution, but can have an important major effect on overall furnace
performance. Care must be taken when making burner adjustments that this
performance is not adversely compromised. This would include situations
such as high CO, excessive carbon loss in the ash, reduced unit efficiency,
poor or loss of superheater or reheater temperature control, extreme
reducing atmospheres that could cause corrosion or excessive fouling, or a
host of other difficulties.
With this in mind, our long-term NO^ control program includes a
potential problem analysis in our decision process to ensure that a viable
N0x control scheme is formulated that will incorporate safe and reliable
boiler operation with the most efficient utilization of present fuel
resources.
FIELD TEST RESULTS
Burner Aerodynamics
As discussed earlier, one of the methods of controlling the fuel and
air mixing history in the Directional Flame Burner is through adjustment of
the secondary air vanes located above and below the coal nozzle (as shown in
Figure 10). These directional air vanes can be tilted up or down in a
number of different positions to change the rate of fuel/air mixing. A
description of these various positions is given in Figure 11.
Burner exit geometries for four selected vane positions are
illustrated in Figure 12 for the fuel nozzle slot. These sketches help to
provide a conceptual understanding of the influence of burner vane position
on near field fuel/air mixing.
The burner input to the furnace, as shown in Figure 12, can be
divided into three streams: the coal plus primary air stream^ and secondary air
streams above and below the burner nozzle. The slowest mixing occurs for
air vane setting characterized as Number 2. Secondary air above the burner
nozzle at this setting is introduced essentially parallel with the primary
coal air stream while the air entering below the nozzle is directed away
135
-------
from the primary stream. The degree of mixing between these primary and
secondary streams increases progressively for burner vane positions
numbered 2, 9, 10, and 7, respectively. Burner vane setting Number 7
produces the earliest intersection of the upper secondary air stream with
the primary coal-air stream. In all of the cases shown in Figure 12, the
coal spreader is tilted up with respect to the burner centerline. The same
order of burner air vane ranking, however, would also exist for a coal
spreader designed with a 0° upward tilt.
The results of tests on three dry bottom TURBO Furnaces to determine
the effect of burner vane position on NO^ is shown in Figure 13. The
results for each unit represent operation at maximum continuous rating and
are evaluated for similar firing conditions with the only change being the
adjustment of the burner directional air vane position. One of the units is
designed without overfire air and is operated on subbituminous coal. The
other two units are designed for different levels of staged combustion and
each fires bituminous coal.
The mixing concepts described earlier can be used to provide some
insight on the trend in the variation of NO emissions with burner air
x
vane setting. NO^ levels are observed to increase with the degree of
mixing. Highest emission levels occur at air vane position Number 7, while
lowest emission levels occur at position Number 2.
The effect of a limited range of coal spreader angles on NO^
emissions is shown in Figure 14. N0x emissions appear to be the most
sensitive to the degree of coal spreader tilt at the higher NO^ producing
(i.e., earliest mixing) air vane settings.
Staged Combustion
Field data analyzed to date shows that staging the combustion process
by the use of overfire air always has the effect of decreasing N0x. The
percentage reduction can vary a great deal, from as low as 8 percent to as
high as 24 percent. All reductions were evaluated under similar firing
conditions with the only mechanical change being the shift of overfire air
dampers from shut to full open. The large variations in the NO^
percentage reductions are the result of numerous design and operating
factors.
136
-------
The effectiveness of overfire air as a NO^ control technique is
dependent, to a degree, on what other control methods are applied. If the
NO^ emissions, for example, have already been reduced by some other means
with the burner aerodynamics, such as directional vane or coal spreader
setting, the effects of overfire air are not as great. This is basically
verifying what has been known for some time. NO^ controls are not
additive and this is even more prevalent when the baseline NO^ levels are
already low.
FUTURE WORK
The current effort of collecting and analyzing field data on
emissions and furnace performance will continue into the forseeable future.
Many of the units listed in Table I not already online, will be coming
online in the next 12 months. The information collected on these new TURBO
Furnaces will be included into the data base for this type of boiler. This
will aid in refining the regression analysis used to predict emissions and
furnace performance. Test series on specific units will be designed to give
further quantifications to the effects of the design and adjustment of
the various burner hardware.
In the laboratory, the field program will be supplemented by numerous
development projects. These will include cold flow model studies into the
aerodynamic characteristics of the burners, further development of a lower
furnace model, and firing tests in a research furnace. The present Riley
research furnace, capable of firing 300 MMBtu/hr on gas or oil, is in the
process of being retrofitted to fire solid fuels to a nominal rate of
100 MMBtu/hr. This conversion includes equipment for sizing, storing, and
metering the fuel, as well as an air preheater for proper combustion, and a
tail-end cleanup system to ensure local emission regulations are adhered to.
Development will be for the directional flame and circular controlled
mixing types of burners. Circular burners are being developed in a full
range of sizes for application in both utility and industrial units,
primarily for retrofit and upgrading purposes.
SUMMARY
The dry bottom Riley TURBO Furnace offers a unique furnace and burner
combination that allows flexibility in limiting N0x release while
137
-------
maintaining safe and efficient operation consistent with economic use of our
indigenous natural resources. The field data has shown that adjustments to
the various burner hardware, such as overfire air ports and directional air
vanes, can significantly reduce the NC>x emissions. The improved
combustion techniques being developed will assist in further reducing
pollutant emissions to meet the new environmental regulation that will be
more stringent in the future for the full range of boiler sizes in the
utility and industrial market.
138
-------
REFERENCES
1. Rawdori, A. H. , Lisauskas, R. A. and Zone, F. J., "Design and Operation
of Coal-Fired TURBO Furnaces for N0X Control," Proceedings: Second
N0X Control Technology Seminar, EPRI FP-1109-SR, July 1979.
2. Lisauskas, R. A., "Design and Operation of Coal-Fired TURBO Furnaces for
N0X Control," Presented to the Committee on Power Generation
Association of Edison Illuminating Companies, April 19, 1979.
3. Rawdon, A. H. and Sadowski, R., "An Experimental Correlation of Oxides
of Nitrogen From Power Boilers Based on Field Data," Presented at the
93rd Winter Annual Meeting of the American Society of Mechanical
Engineers, New York, November 27-30, 1972.
4. Rawdon, A. H. and Johnson, S. A., "Application of N0X Control
Technology to Power Boilers," Proceedings of the American Power
Conference, Vol. 35, pp. 828-837, 1973.
5. Rawdon, A. H. and Johnson, S. A., "Control of NOx Emissions from Power
Boilers," Presented at the Annual Meeting of the Institute of Fuel
(Australian Membership), Adelaide, Australia, November 1974.
6. Hunt, P. J., "Boiler Design for Reduced Emission of Pollutants,"
Presented at the Industrial Fuel Conference, Purdue University, West
Lafayette, Indiana, October 3, 1974.
139
-------
Figure 1. Configuration and flow patterns of the TURBO Furnac
140
-------
11
n n-—-
€:
Figure 2. Typical coal-fired wet bottom TURBO Furnace.
141
-------
B
B
B
B
B
A
B
B
B
B
B
A
PLAN VIEW
SIDE VIEW
Figure 3. Dry bottom TURBO Furnace critical design parameters,
142
-------
t
—1
Figure 4. Southwest Power Station City Utilities of Springfield
Springfield, Missouri.
Figure 5. Interstate Power Company Lansing Power Station Lansing, Iowa.
143
\
-------
[I ••aJj-XT:-
2i
Figure 6. Santee Cooper South Carolina Public Service Authority,
Georgetown Steam Electric Station, Unit No. 2, Georgetown,
South Carolina.
rajt,-
Figure 7. South Mississippi Electric Power Association, Purvis Plant
Units No. 1 and 2 Hattiesburg, Mississippi. '
144
-------
Figure 8. Alabama Electric Cooperative, Inc. Tombigbee Plant, Units No. 2
and 3, (near) Jackson, Alabama.
'Jl —
& • -;'W ¦
Figure 9. Arizona Electric Power Cooperative, Inc. Apache Station,
Units No. 2 and 3, (near) Cochise, Arizona
145
-------
STAGED COMBUSTION PORT
UPPER VANE CONTROL
BURNER HEAD
GAS-ELECTRIC
IGNITOR
LOWER —¦
VANE
CONTROL
VELOCITY
CONTROL
DAMPER
COAL
SPREADER
COAL
SPREADER
DIRECTIONAL VANES
VELOCITY
CONTROL
DAMPER
SECONDARY AIR
DAMPER
Figure 10. Riley Directional Flame Burner with staged combustion ports
for coal firing.
146
-------
NUMBER OF
VANE SETTING
REAR FRONT
1
j/ (UPPER VANES) *
S (LOWER VANES) jf
2
/ \
\ /
3
//
//
4
(ALTERNATING BURNERS)^*
5
< >
< >
6
7
//
\\
8
9
\\
//
10
—
Figure 11. Burner air vane settings.
147
\
-------
WATER WALL
UPPER VANES;
BURNER
*"
COAL SPREADER
BURNER NOZZLE
^NOZZLE SUPPORT LUG
LOWER VANES
AIR CHAMBER ASSEMBLY
"NUMBER •" TILTING VANE SETTING
WATER WALL
UPPER VANES
BURNER
COAL SPREADER
BURNER NOZZLE
NOZZLE SUPPORT LUG
LOWER VANES
AIR CHAMBER ASSEMBLY
"NUMBER 2" TILTING VANE SETTING
WATER WALL
UPPER VANES,
BURNER
COAL SPREADER
BURNER NOZZLE
'NOZZLE SUPPORT LUQ
LOWER VANES
AIR CHAMBER ASSEMBLY
C WATER WALL
UPPER VANES
BURNER
COAL SPREADER
BURNER NOZZLE ^
IOZZLE SUPPORT LUG
LOWER VANES
¦AIR CHAMBER ASSEMBLY
NUMBER 10" TILTING VANE SETTING "NUMBER 7" TILTING VANE SETTING
Figure 12. Directional flame burner exit geometries.
-------
0.8-
£ 0.7 H
0Q
I
Z 0.6
v>
I0-5
(A
s
UJ
0.4
0.3
MIXING INCREASES
T
UNIT "A"
STAGED
BIT.
UNIT "B"
UNSTAGED
SUB. BIT.
UNIT "C"
STAGED BIT.
-r
t
T
No. 2 No. 9 No. 10 No. 7
SECONDARY AIR DIRECTIONAL VANE SETTING
Figure 13. Effect of burner air vane position on NO^ emission.
3
»-
CD
|
o
z
o
z
0.8'
0.7-
0.6
S 0.5
0.4
0.3-1—V
~ NO.7 DIRECTIONAL VANE SETTING
A NO. 10 DIRECTIONAL VANE SETTING
O NO. 9 DIRECTIONAL VANE SETTING
O NO. 2 DIRECTIONAL VANE SETTING
I i r i i
30 35 40 45 50 55
—i—
60
COAL SPREADER ANGLE DEGREES ¦ UPWARD
Figure 14. Effect of coal spreader angle on NO^ emission.
149
-------
TABLE I. RILEY DRY BOTTOM TURBO FURNACE UNITS
Scheduled
Customer Startup Year
Union Carbide 1963
City of Springfield 1976
Emery Industries 1976
Interstate Power Company 1977
Cleveland Cliffs Iron Company 1978
South Mississippi Electric Power Association (two units) 1978
Santee Cooper Public Service Authority 1977
Delmarva Power and Light Company 1980
Dairyland Power Cooperative 1979
Salt River Project 1978
Arizona Electric Power Cooperative, Inc. 1978
City of Kansas City 1980
Alabama Electric Cooperative, Inc. 1978
Wisconsin Electric Power Company 1980
Wisconsin Electric Power Company 1982
Arizona Electric Cooperative, Inc. 1978
Alabama Electric Cooperative, Inc. 1979
Salt River Project 1978
Salt River Project 1984
Cleveland Cliffs Iron Company 1979
Cajun Electric Power Cooperative, Inc. 1979
Cajun Electric Power Cooperative, Inc. 1980
Hoosier Energy Division 1981
Hoosier Energy Division 1982
Santee Cooper Public Service Authority 1980
Santee Cooper Public Service Authority 1981
Central Illinois Light Company 1983
150
-------
TABLE II. COAL ANALYSIS
Coal Rank
Properties
Bituminous
Bituminous
Subb ituminous
% C
66.32
66.80
47.81
% H
4.62
4.67
3.43
% N
1.41
1.40
1.05
% 0
7.07
4.01
11.13
% S
2.07
3.54
0.42
% Ash
12.71
12.88
6.16
% Moisture
5.8
6.7
30.0
HHV Btu/lb
11,832
12,064
8,253
Ash fusion temperature
(I.D.-Oxy)
2,550
2,180
2,120
Slagging index
Low
High/severe
High/severe
151
-------
NO EMISSIONS CHARACTERISTICS OF ARCH-FIRED FURNACES
x
By:
T. W. Sonnichsen
KVB, Inc.
Irvine, California 92714
J. E. Cichanowicz
Electric Power Research Institute
Palo Alto, California 94303
152
-------
ABSTRACT
Field tests have been conducted on three subbituminous pulverized-coal
arch-fired utility boiler configurations. The objective of these tests was to
determine as-found N0x emission levels and the influence of combustion modifi-
cations on these emissions . These configurations are unique in that the coal
is introduced downward from the arch into the furnace with the bulk of the
combustion air added through the front wall perpendicular to the flame jet.
Staged combustion conditions are thereby generated which have been shown to be
conducive to low N0X emissions.
Corrected NO levels ranged from 200 ppm to 350 ppm. The lowest emissions
were emitted from the largest (275 MW) boiler. Variations in excess air, air
flow injection distribution between burner and front wall and burner stoi-
chiometry were shown to reduce NO emissions by 5 to 35 percent. A discussion
of these results is presented.
153
-------
INTRODUCTION
Promulgation of New Source Performance Standards (NSPS) by the U.S.
Environmental Protection Agency (EPA) for coal-fired utility boilers has
created significant interest in developing suitable low-NOx combustion
systems. Several programs under boiler manufacturer, EPA and EPRI sponsorship
(1-5) have been underway directed at developing new burner/boiler design
concepts. All of these designs in one manner or another use a staged combus-
tion process which has been shown to be the most practical means of achieving
reduced NOx emissions.
An alternative to these new burner/boiler configurations is the
arch-fired furnace. Boilers of this design are characterized by (A) burners
situated in the furnace arch, oriented vertically and firing down into the
furnace and (B) a significant portion (up to 80 percent) of the combustion air
supplied through the furnace frontwall along the flame front perpendicular to
the flame jet. A long U-shaped flame is thereby generated with inherent
staged combustion conditions. This method of firing has acquired several
descriptions including "arch-fired," "vertically-fired," "down-fired," "u-
flame," and "down-shot." The terms "arch-fired" and "vertically-fired" have
been used interchangeably during this study. The arch-fired design should not
be confused with other firing configurations which, although fired from the
top of the furnace downward, introduce all combustion air at the burner.
These boilers, therefore, do not have the inherent staged combustion
conditions of the arch-fired design and consequently do not have corresponding
low NOx emission characteristics.
Arch-fired boilers have been used in utility applications for over sixty
years. Boilers of this general configuration were the first to successfully
154
-------
use pulverized coal combustion. Development of subsequent designs contributed
significantly to modern design concepts (7). Modern use of arch-firing in the
United States has been limited to three subbituminous coal-fired boiler con-
figurations used by the Wisconsin Electric Power Company (WEPCo) and several
applications to anthracite coal-fired boilers (8).
This paper presents a discussion of arch-firing as a low-NOx design
approach for subbituminous coal combustion in utility boilers. The boilers
considered in this study are the three WEPCo configurations. Descriptions of
these boiler designs are presented together with results of field tests con-
ducted to characterize NO emissions. Comparisons are made between the
emission characteristics of the arch-fired boilers. A discussion is presented
identifying in a speculative manner the significant combustion parameters
conducive to low-NOx.
This study was conducted as part of an EPRI sponsored program to
investigate the use of arch-firing as an alternative utility boiler design.
The results of this study are documented in an EPRI report, "NO Emissions from
Pulverized Coal Vertically-Fired Boilers" (9). An economic evaluation of the
use of arch-firing has been conducted by Foster Wheeler Energy Corporation
( 10).
BOILER DESCRIPTIONS
The three boiler configurations are unique having been custom designed by
engineers at WEPCo. The smallest boilers, with a capacity of 80 MW, are
located at the Port Washington Generating Station. The two other designs are
at the Oak Creek Generating Station. The 125-MW boilers are located in the
North Plant while the 275-MW boilers are in the South Plant. Nine boilers
were included in the test program: four of the 80-MW class (Boilers 2, 3, 4,
and 5), three of the 125-MW class (Boilers 1, 2, and 4), and two of the 275-MW
size (Boilers 5 and 6).
A summary of the design parameters for the three boiler configurations is
presented in Table I. This information is provided as reference for inter-
ested utility and manufacturing engineers. A brief description of the furnace
155
-------
design factors follows. More comprehensive descriptions of these boilers can
be found in References 8 and 9.
The general arrangement of the three boiler types are illustrated in
Figure 1. The boilers share several common design features as well as having
individual characteristics• In all three, the burners are located in the
furnace arch, are oriented vertically and fire down into the furnace. The
pulverized coal for these boilers is prepared and temporarily stored above the
burners in bins* Feeders control the flow of pulverized coal into the primary
air stream that transports the coal through the burner and into the furnace.
Ignition occurs approximately three to five feet from the burner outlet,
producing a flame jet that penetrates to the bottom of the furnace. Combus-
tion air is added around the burner (termed secondary air) and from the front
wall perpendicular to the flame jet. Front wall air is introduced into the
furnace through slits between the front wall watertubes.
The final common design feature of these units is the incorporation of
radiant superheat and reheat heat transfer in the furnace. These elements
were included in the original design to improve low-load steam temperatures
and overall operating efficiency. In essence, however, this design results in
heat transfer characteristics in the furnace that differ from those of conven-
tional waterwall furnaces. Close control of the combustion conditions in the
furnace is also required in order to maintain proper steam conditions. Exces-
sive steam tube temperatures are controlled primarily by overall boiler excess
air. In all three designs, the furnace front wall and division walls under
the arch (where present) consist of watertubes. The furnace side and back
walls consist of steam tubes.
The primary differences between the WEPCo boilers (other than capacity)
involve (a) the burner design and combustion air distribution, generating
differing near burner combustion conditions, (b) the arrangement of the
burners and (c) the presence of division walls.
The design of the burners used on the 80 MW and 125 MW boilers is simple,
consisting of tapered pipes with a four pronged helix inside near the burner
tip. The helix is used to impart swirl to the pulverized coal/primary air
stream. The burners used on the 275 MW boilers, on the other hand, are much
156
-------
more complex. A diagram of the burners used on the 275 MW boilers is
presented in Figure 2. The pulverized coal is transported from the feeder to
the burner by scavenging air. The primary air is spiraled around the coal
stream generating a hollow cylinder of pulverized coal injected into the
furnace. The balance of the combustion air is supplied both at the burner and
from the furnace front wall. The additional air supplied at the burner passes
through the two concentric cones shown in Figure 2. The inner cone converges
to generate high velocity air that assists the flame jet in penetrating to the
bottom of the furnace. The outer cone diverges to produce low velocity
combustion air in the immediate vicinity of the burner tip. Both cones have
non-adjustable vanes to impart swirl to the incoming air. Modifications have
been made to block approximately 50 percent of the outer cone inlet area.
There is no available hardware to control secondary air flow to individual
burners in a cell.
The arch-fired designs also differ in the number and arrangement of the
burners and the presence of division walls. These factors are illustrated in
Figure 3. The 80 MW boilers are fired through 20 burners arranged in a single
line evenly spaced across the length of the arch. There are no division walls
used on this design. The 125 MW are fired through 16 burners also arranged in
a single line but separated into four groups of four. The presence of water-
tube division walls underneath the arch between these burner groups form four
furnace "cells." The sixteen burners on the 275 MW design have also been
separated into four groups of four. The arrangement of these burners, as
shown in Figure 3, is significantly different. The burners in each group are
configured in a "trapezoidal" pattern forming two rows of burners across the
arch. The outer row of burners (farthest from the furnace front wall) are
therefore somewhat shielded from the front wall air by the inner row of
burners. The division walls on the 275 MW boilers extend across the furnace
and up into the initial convective passes. The furnace is thereby effectively
separated into four sub-furnaces operating somewhat independently.
157
-------
NO EMISSIONS MEASUREMENT RESULTS
Presented in this section are the results of the NO emission testing
conducted on the three arch-fired boiler designs. The test program on each
boiler consisted of two parts: (1) a baseline characterization to determine
as-found emission levels and the influence on NO levels of excess air, and
(2) combustion modification tests directed at determining the effect of
alterations in the inherent staged combustion conditions on NO emissions. The
available flexibility of combustion air distribution along the flame jet and
burner stoichiometry through burners-out-of-service operation were used as the
two methods of implementing staged combustion. The following discussions
follow this format, describing first the results of the baseline tests and
then the combustion modification tests.
All tests conducted during this program used the coal customarily fired
in these boilers. The primary coal sources for these boilers are located in
Southern Illinois/Western Kentucky. The 125 MW boilers also periodically fi^
western coal from Wyoming. A summary of analyses for as-fired pulverized co^i
samples collected during the program are presented in Table II.
BASELINE NO EMISSION CHARACTERIZATION
The full load NO emissions for the three boiler types are presented in
Figure 4. With the exception of one of the 80 MW boilers, the emissions data
form three groups corresponding to the three boiler types. Note that the No
levels decrease with increasing boiler capacity. For all three designs, the
NO levels increased with increased excess air.
The spread in the emission levels for three of the four 80 MW boilers
(Port Washington Boilers 2, 4 and 5) was introduced by variations in the
158
-------
excess air level. NO emissions are shown to increase by approximately 30 ppm
for each one percent increase in flue gas 02« The NO levels for the remaining
unit (Boiler 3), however, were significantly lower, by 80 ppm at comparable 02
levels. This difference between the emission levels from this boiler and the
other three 80 MW units was partially attributed to the quantity of primary
air flow. While the data presented in Figure 3 for Boilers 2, 4, and 5 repre-
sented operation at primary air pressures of 15 to 16 inches of water, the NO
emissions data for Boiler 3 were obtained while operating at 13 inches of
water. To further investigate this, NO emissions were measured over a range
of primary air pressure on Boilers 2 and 3. While these data indicated a
strong dependence of NO on primary air pressure (10 to 15 ppm per one inch H2O
increase in primary air pressure in this range), the NO emissions from
Boiler 3 at comparable primary air pressures were still 50 ppm below those of
the remaining three boilers. This difference has not been completely
resolved.
Also presented in Figure 4 are the emission levels for the three 125 MW
boilers tested (Oak Creek Boilers 1, 2 and 4). Boilers 1 and 4 operated at
nearly full capacity, while Boiler 2 had a maximum load of 92 MW due to
convective superheat tube metal temperature limitations. Emissions for
Boilers 1 and 4 are shown to be comparable. Operation of Boiler 2 at the
lower load required higher excess air and consequently produced higher NO
levels. The limited data presented in Figure 4 indicate a sensitivity of
approximately 20 ppm NO decrease with each 1 percent decrease in 02» This was
confirmed by the more extensive test series conducted at lower loads.
The composite NO emission data for loads in excess of 225 MW for both
275 MW boilers are also included in Figure 4* As shown, corrected NO
emissions were between 190 and 205 ppm, with the highest emissions correspon-
ding to the peak load of 265 MW. Failure to achieve full load (275 MW) and
limitations in excess air flexibility were due to excessive steam tube element
temperatures, increased flexibility in excess air was available at 200 MW
operation. Results of tests conducted at this reduced load are shown in
Figure 5. Furnace exhaust 02 ranged from 6 to 10 percent, with corresponding
NO levels increasing from 175 to 250 ppm, an increase of 20 ppm NO per
159
\
-------
1 percent 02» These data suggest the possibility of additional NO reductions
of 30 to 50 ppm with reductions in excess air to levels corresponding to
4 percent 02—levels that had been achieved previously on individual 80 mw and
125 MW boilers. Operation at this excess air level was not possible on the
275 MW boilers due to excessive tube metal temperatures.
As mentioned previously, the 275 MW boilers are divided into four cells
by division walls, effectively producing four independent sub-furnaces. Some
variation was noted between the NO emission levels of the four furnace cells
comprising each of these boilers. NO emissions differed by as much as 25 ppm
at comparable excess air levels. The reasons for this could include manage-
able differences (heat input, combustion air distribution, coal feed
distribution) and several other nonmanageable conditions (furnace slagging and
subtle differences in burner hardware). The causes of the differences between
the cells were investigated but never fully identified.
COMBUSTION MODIFICATION TESTS
The initial combustion modification tests conducted on the arch-fired
boilers were directed at investigating the impact on NO emissions of changes
in the front wall air addition patterns along the flame front. These tests
were made possible by the compartmentalized front wall windboxes, especially
on the 80 MW and 125 MW boiler configurations which are equipped with three
levels of secondary air injection along the flame path. Distribution of
combustion air between the secondary to front wall air on the 275 MW boilers
was less straightforward due to the more complex burner design.
A second series of combustion modification tests were conducted on each
boiler type involving burners-out-of-service operation to increase the fuel
richness in the initial flame region. On the 80 MW and 275 MW boilers, lack
of suitable hardware precluded the removal of individual burners from
service. Instead, coal flow through individual feeders was stopped resulting
in the associated pair of burners being removed from service. Shutoff dampers
were available on the 125 MW boilers allowing individual burners to be removed
from service.
160
-------
NO emissions on the 80 MW boilers were shown to be significantly
influenced by the distribution of front wall air addition. The measured
distribution of air flow from primary (P), secondary (S) and the three front
wall levels (F1, F2, F3) and resulting furnace NO emissions are presented in
Figure 6. As shown in Figure 6, decreasing the proportion of combustion air
flow through the lowest air levels (farthest from the burner) by adjusting
sidewall air position increased NO emissions. NO emissions at base condition
(all sidewall dampers wide open) increased from 275 ppm to 325 ppm with the
lower dampers half closed and to 340 ppm with the lower dampers closed. As
noted in Figure 6, the proportion of air flow to the upper two levels
increased as did the proportion of secondary air introduced around the burner,
thus maintaining constant total air flow. Primary air flow remained constant.
Similarly, NO emissions decreased as the front wall air was biased to the
lower levels. Operation with the top level of dampers closed on Boiler 3
reduced NO by approximately 15 ppm from the full open position. Repeating
these tests on Boiler 2 decreased NO levels by 45 ppm ( 8% reduction) by
stopping the air flow to the top level.
Operation with two of the ten feeders out of service (4 of the 20
burners) on Boiler 2 reduced NO by approximately 50 ppm. Operation at this
condition with all dampers full open and 4 percent Oj at 55 MW achieved NO
emissions of 200 ppm, the lowest levels reached during testing on the 80 MW
class boilers.
Typical results of the front wall air distribution tests conducted on the
125 MW boilers is shown in Figure 7. NO emissions were shown to be relatively
insensitive to front wall air distribution. Measured air distribution
patterns are presented in Figure 7. If anything, NO emissions increased
slightly as air was biased to the lower levels, that is, closing the top level
dampers. It should be noted that numerous combinations of full and partially
closed dampers were tried on all three 125 MW class boilers with results
consistent with those presented in Figure 7.
A series of tests were conducted on Boiler 4 (125 MW), with two and four
burners out of service, to investigate the impact on NO emissions of gross
flame stoichiometry in the near burner region. Results from these tests
161
-------
indicated a relatively small impact oil NO emission levels. NO emissions
decreased by up to 30 ppm depending on the pattern of burners out of service.
Combustion modification tests conducted on the 275 MW boilers were
limited to (A) changes in the combustion air distribution between the
secondary and front wall air and (B) biasing the coal feed between the inboard
and outboard burner pairs. Operation with a feeder out of service was not
attempted as this would have resulted in severe disruptions in tube element
temperatures. Similarly, operation with individual burners out of service
could not be applied since there were no coal shutoff dampers to individual
burners. Other procedures that could potentially have been employed to alter
the combustion process, such as (A) varying the secondary air flow between the
four burners in a cell or (B) changing the primary or secondary air swirl,
were also not possible due to hardware limitations.
The combustion modification tests discussed in this section were
conducted on cells 2 and 3 only. Tests showed that these interior cells
(controlling the radiant superheat areas of the furnace) had somewhat more
flexibility and were amenable to the modifications imposed without causing
severe boiler upset. While the results are limited to one cell, the trends in
the emissions are taken to be characteristic of the entire boiler.
The results of the series of tests of redistributing the secondary and
tertiary air flows are presented in Figure 8. These data represent operation
at a reduced load of 200 MW with consistent cell exhaust O2 concentrations of
6 percent. Biasing air flow to the front wall decreased NO emissions by 10 to
25 ppm. Similar adjustments to the combustion air distribution in Cell 3
resulted in a somewhat larger NO reduction. The differences in NO reductions
between the cells were investigated but not resolved.
A brief series of tests were conducted on cell 2 to determine the effects
of varied coal feed distribution on NO emissions. From a base condition,
feeder speeds were varied to provide a biasing of the coal first to the
outboard feeder (delivering coal to the outer burner pair) and then to the
inboard feeder (delivering coal to the burner pair next to the front wall).
Biasing the coal away from the front wall decreased NO by ~10 ppm, while
biasing towards the front wall increased NO by a similiar amount.
162
-------
FLY ASH CARBON CONTENT MEASUREMENTS
Fly ash samples were extracted from the flue gases upstream of the air
preheater and analyzed for carbon content using ASTM procedures (11).
Excessive fly ash carbon content is, of course, an indication of incomplete
combustion and could be a factor limiting the application of staged
combustion.
The results of the data on fly ash carbon concentration obtained during
the test program are presented in Figure 9 as a function of furnace exhaust
NO. Although a significant degree of scatter is present, it is evident that
carbon content generally increased as the NO decreases. This was especially
true for the 275 MW class boilers. It has been speculated that the high
carbon content from these boilers is due to the multiple burner rows with the
outboard flame jet somewhat screened from the front wall air addition.
Additional investigation into causes of the trend shown in Figure 9 will be
necessary in order to propose solutions and the impact on NO emissions. It
should be noted that the data presented in Figure 9 represent as-fired
conditions with no attempts to optimize the potential tradeoffs of NO and high
carbon content.
163
-------
DISCUSSION
It is interesting to note from the results presented in the previous
section that the N0X emission levels from the three arch-fired boiler designs
decreased as the size and complexity of the combustion process increased. The
lowest emissions were measured on the 275 MW design that had the largest
proportion of combustion air introduced at the burner and therefore would have
been expected to have less inherent staged conditions and correspondingly
higher NC>x levels. Two explanations have been proposed to account for this;
(A) the heat absorbing characteristics of the furnaces and (B) fundamental
differences in the combustion processes between the boiler types. The purpose
of this section is to discuss in a speculative manner the results of this test
program in terms of these hypotheses.
FURNACE HEAT ABSORPTION
It has been demonstrated that reduced flame temperatures result in lower
N0X emissions (12). Large furnace volumes and/or furnace division walls
providing increased heat absorbing surfaces could be expected to result in
lower N0X emissions. As discussed previously, there are significant
differences between the furnaces configurations of the arch-fired furnaces,
especially with regard to the division walls on the 275 MW design. This has
been proposed as a possible explanation of the reduced N0X emission from these
boilers.
To test this hypothesis, estimates have been made of the volumes and
available heat absorbing surfaces for three regions in each arch-fired design
corresponding to: (A) the entire furnace, (B) the lower furnace below the
plane of the arch and (C) the region directly below the arch.
164
-------
A summary of the estimated volumes and surface areas of these regions is
included in Table III together with heat absorption on a per unit volume and
area basis. As shown in Table III by the consistent heat absorption, the
three arch-fired boiler designs show a nearly proportional increase in volume
and area as the size of the boiler increases. It would appear from these data
that the hypothesis of the lower NO emissions on the 275 MW boilers due to
size and heat absorption does not appear to be valid.
This analysis, however, does not take into account the location of the
heat absorbing surfaces with respect to burner arrangement, flame jet mixing
patterns and the medium (water or steam) transported through the furnace and
division wall tubes. Each of these factors could significantly influence heat
removal patterns in the flames and resulting N0X formation.
DIFFERENCES IN THE COMBUSTION PROCESSES
The influence of differences in the inherent staged combustion processes
on outlet N0X emissions is much more complex. While the application of staged
combustion has repeatedly been demonstrated as a practical means of achieving
low N0X emissions, the level of understanding of the fundamental nature of
this process is generally low (12). Staged combustion has been conceptualized
as reducing N0X emissions by reducing the formation rate of both thermal and
fuel N0X» In addition, the destruction of NO to other products has been
observed in the intermediate regions of one staged combustion process (13).
It may be that competing or counterbalancing reactions occur within the flame
structure that significantly affect final NO levels.
The complexity of understanding the N0X formation/reduction processes in
the arch-fired furnace design are compounded by the aerodynamic mixing pat-
terns characteristic of these configurations. T3ie inherent staged combustion
process can be conceptualized as taking place in four steps: (A) the near
burner region, (B) the intermediate stage, (C) lower furnace combustion and
(D) burnout. These zones are identified in Figure 10 for the arch-fired
design.
As discussed previously, significant differences exist between the burner
designs of the 80 MW and 125 MW boilers, and that of the 275 MW configuration
165
-------
that are reflected in the near burner mixing patterns identified as region A
in Figure 10. The simpler burner arrangement produces a coal/air jet entering
a region of relatively quiescent furnace gases. This situation is conducive
to forming a well mixed jet in which the coal rich internal core is broken
down relatively rapidly. The more complex burner on the 275 MW design
produces a jet with a fuel rich central core surrounded by a concentric ring
of air moving at comparable velocities. The resulting low levels of shear
between these two streams could result in the fuel rich central core
maintaining its integrity (and its fuel rich conditions) until well into the
furnace. The longer residence times at fuel rich conditions could result in a
decreased proportion of fuel nitrogen conversion to NOx as opposed to the 80
MW and 125 MW designs.
Conceptually, the intermediate region (region B in Figure 10) would
correspond to the region of entrainment of the front wall combustion air into
the flame jet. Calculations have shown, however, that buoyancy forces in this
region are important and tend to drive the relatively cold front wall air down
the wall delaying mixing with the flame jet until into the lower furnace
regions. This flow pattern shown in Figure 10 is also enhanced by front wall
flow deflectors on the 125 and 275 MW arrangements. It can be speculated that
oxygen deficient flame conditions exist within this intermediate region
containing intermediate combustion products (H, CO, NH3, HCN) that have been
shown to reduce N0x to N2. This would be especially true of the outer row of
burners in the 275 MW design that are partially shielded from penetration of
the front wall air by the inner burner row. The inherent staged combustion
process of the arch fired design, especially the 275 MW furnaces, may be
extremely conducive to N0X reduction.
The lower furnace region identified as region C in Figure 10 is also of
interest. Results of earlier tests conducted on arch-fired designs
demonstrated the lower regions to have the highest furnace temperatures
(14). In this region mixing between the frontwall air and flame jet are still
incomplete so that furnace geometry could play a significant role in NOx
formation by controlling local flame stoichiometry. It has been
conceptualized that the application of arch-firing to boilers in the 600 MW
166
-------
class will require an opposed fired design. The opposed flame interaction in
this region could significantly alter N0X formation and/or reduction
patterns. Also, the possibility of N0x formation from nitrogen contained in
the char in the upper furnace burnout zone (region D) cannot be discounted.
Again, the level of understanding of these processes is not complete enough to
formulate any definitive answers.
CONCLUSIONS
. Arch-fired utility boilers with front wall air addition that have
been in service for up to 40 years have demonstrated NOx emission
levels comparable to specially designed low NOx combustion
systems«
, It has been speculated that these low emission levels are
attributable to combustion conditions similar to that found in
the more sophisticated new designs.
Assessment of the arch-fired configurations as a viable
alternative to the modern low NOx combustion systems will require
economic and reliability comparisons which are now in progress.
167
-------
REFERENCES
1. Barsin, J. A., "Pulverized Coal Firing NO Control," Proceedings: Second
N0X Control Technology Seminar, EPRI FP-1^09-SR, July 1979.
2. Vatsky, J., "Experience in Reducing N0X Emissions on Operating Steam
Boilers," Proceedings: Second N0x Control Technology Seminar, EPRI
FP-1109-SR/ July 1979.
3. Brown, Richard A., "Alternate Fuels and Low N0X Tangential Burner
Development Program," Proceedings of the Third Stationary Source
Combustion Symposium, Volume II, Advanced Processes and Special Topics,
EPA-600/7-79-050b, February 1979.
4. Zallen, D. M. et al., "The Generalization of Low Emission Coal Burner
Technology," Proceedings of the Third Stationary Source Combustion
Symposium, Volume II, Advanced Processes and Special Topics,
EPA-600/7-79-050b, February 1979.
5. Johnson, S. A. et al., "An Advanced Low-N0x Concept for Pulverized Coal
Combustion," Proceedings: Second NO Control Technology Seminar, EPRI
FP-1109-SR, July 1979.
6. Dornbrook, F. L,, "Developments in Burning Pulverized Coal," Paper
presented at the semi-annual meeting of the ASME, Milwaukee, Wisconsin,
June 1948.
7. Orning, A. A., "The Combustion of Pulverized Coal," Chemistry of Coal
Utilization, Volume II (John Wiley and Sons: New York), 1945.
8. Cichanowicz, J. E. et. al., "N0X Emissions Characteristics of Down-Fired,
Sequential Air Addition Furnaces," Proceedings: Second NOx Control
Technology Seminar, EPRI FP-1109-SR, July 1979.
9. Sonnichsen, T. W., "NO Emissions from Pulverized Coal Vertically-Fired
Boilers," EPRI, Final report in preparation, Contract RP-1339-1.
168
-------
10. Vatski, J., "Economic and Engineering Analysis of Arch-Fired Furnaces,"
EPRI Report in preparation, Contract RP-1339-2.
11. Laboratory Sampling and Analysis of Coal and Coke, ASTM D271-70.
12. Thompson, R. E. et al., "Assessment of N0X Control Technology for Coal
Fired Utility Boilers," Environmental Control Implications of Generating
Electric Power from Coal, Appendix D, Argonne National Laboratory,
ANL/ECT-3, December 1977.
13. Johnson, S.A. et al., "The Primary Combustion Furnace System—An Advanced
Low-NO^ Concept for Pulverized Coal Combustion," Second EPRI N0x Control
Technology Seminar, Denver, CO, November 1978.
14. Tenney, E. J., "Practical and Theoretical Aspects of Firing Low Grade
Bituminous Coal in Pulverized Form," Proc. 3rd Intern. Conf. Bituminous
Coal, 2,370-399, 1931.
169
-------
Port Washington
(80 MW)
Oak Creek
South Plant
(275 MW)
Oak Creek
North Plant
(125 MW)
1. WEPCo arch-fired boilers
170
-------
Scavenger Air and
£ Pulverized Coal
Secondary Air
4 Primary Air
«¦
Secondary Air
Figure 2. Oak Creek South Plant 275 MW Burner Configuration
171
-------
Front Wall
ooooooooooo ooooooooo
Arch
80 MW
Front Wall
oooo oooo oooo oooo
I
1
Arch
125 MW
Front Wall
o o
o o
o o
o o
o o
o o
o o
Arch
275 MW
Figure 3. Arch arrangement of WEPCo furnaces - not to scale
172
-------
500
400
300
>*
B
Q
(N
O
*
Zl 200
I
100
—I 1 1
o Port Washington - 70 MW
Oak Creek North Plant - 125 MW
^ Oak Creek South Plant - 225 MW+
©--<5D
A-
FURNACE EXHAUST «
Figure 4. Full load baseline NO emissions# WEPCo arch-fired boilers.
The numbers in the symbols denote the boiler numbers for
each station.
173
-------
275
Q.
P
~
*
Figure 5. Composite HO Bnissions as a Function of Exeats Air,
Oak Creek Boiler 6, 200 Mw
174
-------
400 _
300 —
200 _
100 —
P S F1 F2F3
P S F1F2F3
Lower Dampers
50% closed
P S F1 F2 F3
Lower Dampers
closed
Figure 6. Impact of front wall air distribution on NO emissions,
Port Washington Boiler 3 (80 MW), 55 MW, 6.5% O.
(P-Primary Air Flow, S-Secondary Air Flow, Fl-Front
Wall Upper Level Air Flow, F2-Front Wall Mid Level
Air Flow, F3-Front Wall Lower Level Air Flow)
175
-------
Base Top Dampers Mid Level Lower Dampers
closed Dampers-closed closed
Figure 7. Impact of front wall air distribution on NO emissions
Oak Creek Boiler (125 MW), 107 MW, 4.0* 02
(P-Primary Air Flow, Fl-Front Wall Uppper Level Air Flow,
F2-Front Wall Mid Level Air Flow, F3-Front Wall Lower
Level Air Flow)
176
-------
Cell 2
Cell 3
200 _
150 —
d*>
m
s
a
£
A
O
R
100 —
50 _
P S F
P S F
P S F
P S F
Figure 8. Impact of combustion air distribution on NO emissions
Oak Creek Boiler 6 (275 MW), 200 MW, 7% 0^
(P-Primary Air Flow, S-Secondary Air Flow, F-Front
Wall Air Flow)
177
-------
T
T
T
o Port. Washington
Oak Creek North Plant
A Oak Creek South Plant
&
&
175
O
4$
o
° ^
O o
o v
O o
o
o o
o
200
J
225 250 275
NO, ppm (3* O^)
300
325
o\
i
350
Figure 9. Fly Ash Carbon Content as a Function on NO Bnissions,
WEPCo Arch-Fired Boilers.
178
-------
///
Figure 10. Schematic of Inherent Staged
Combustion Process
A Near Burner Region
B Intermediate Stage
C Lower Furnace Region
D Burnout Zone
179
-------
TABLE I. ARCH-FIRED BOILER OPERATING PARAMETERS
(AT PEAK LOAD DURING TEST CONDITIONS)
Port
Washington
Oak Creek
North Plant
Oak Creek
South Plant
Load, MW
70.0
125.0
265.0
Steam flow
520.0
888.0
1780.0
103 lb/hr
Heat input
694.0
1185.0
2293.0
106 Btu/hr
Coal flow
57.8
98.8
191.0
103 lb/hr
Total air flow
19.9
32.6
72.0
106 ft3/hr
Primary air flow
1.9
4.4
7.2
106 ft3/hr
Secondary air flow
16.3
28.2
28.2
106 ft3/hr
Tertiary air flow
1.7
—
36.6
106 ft3/hr
Furnace volume
74.0
111.0
171.6
103 ft3
Tube surface area
10.8
17.2
38.3
103 ft2
Number of burners
20
16
16
Burner diameter, in.
6.0
4.5
6.0
180
-------
TABLE II. AS-FIRED COAL ANALYSES
(Percent by Weight)
proximate (Wet)
Moisture
Ash
Volatile
Fixed Carbon
Btu/lb
Ultimate (Dry)
Carbon
Hydrogen
Nitrogen
Chlorine
Sulfur
Ash
Oxygen
Port Washington Oak Creek North and South Plants
Midwestern Midwestern Western
1.40
15.38
32.37
50.85
11,985
67.99
4.34
1.30
0.20
3.01
15.60
7.56
1.82
14.29
35.04
48.85
11,925
68.73
4.69
1.46
0.30
1.72
14.55
8.55
2.81
20.70
37.29
39.20
10,318
60.55
4.55
1.27
0.06
1.04
21.30
11.23
181
-------
TABLE III. ARCH-FIRED HEAT RELEASE DATA
80 MW 125 MW 275 MW
Total heat input, Q 0.69 1.19 2.29
109 Btu/hr
Total furnace volume, V_ 74.0 111.0 171.6
103 ft3 F
Q/Vf 9.3 10.7 13.3
Total furnace area, A,, 10.8 17.2 38.3
103 ft2
Q/Ap 63.9 69.2 59.8
Lower Furnace Volume, VLF 36.0 58.5 97.3
103 ft3
Q/Vlf 19.2 20.3 23.5
Lower furnace Area, ATT, 6.1 10.5 19.5
"5 O Lit
103 ft2
Q/A^ 113.1 113.3 117.4
Below Arch Volume, VDA 12.5 25.0 46.6
•a 1 iSA
103 ft3
QAba 55.2 47.6 49.1
Below Arch Area, ABA 2.8 6.5 10.0
103 ft2
Q/AgA 246.4 183.1 229.0
182
-------
RELATIONSHIP BETWEEN NO AND
FINE PARTICLE EMISSIONS
By:
M. W. McElroy and R. C. Carr
Air Quality Control Program
Coal Combustion Systems Division
Electric Power Research Institute
Palo Alto, California 94303
183
-------
ABSTRACT
Data from EPRI-sponsored field test programs at pulverized coal-fired
utility power plants indicate that boiler combustion conditions producing low
N0X emissions also tend to suppress the generation of fine, submicron
particulate matter. Specifically, the mass of fine particles measured at the
outlet of boilers in the 0.1-micrometer diameter region are reduced by up to
one or more orders of magnitude when low N0x emissions are observed. These
observations are consistent with the present theories of volatilization/
condensation processes believed to be responsible for particle generation in
the fine particle size region.
The significance of this discovery is that particulate collectors
(electrostatic precipitators and fabric filter baghouses) generally exhibit a
minima in collection efficiency at this particle size region. Furthermore,
these particles: (1) can contribute to visibility problems due to particle
growth within the plume, and (2) have been implicated as bad actors from a
health effects standpoint due to their possible enrichment in trace elements
and unfavorable transport properties. It now appears that these deficiencies
inherent to particulate control devices may, in part, be overcome by the
application of N0X combustion control.
184
-------
ACKNOWLEDGEMENTS
The data presented in this paper are derived from EPRI test programs
conducted by Meteorology. Research, Inc. (MRI) under the direction of
Dr. David Ensor. Special credit is given to Mr. Gregory Markowsky of MRI for
the development of crucial fine particle data reduction methods.
185
-------
RELATIONSHIP BETWEEN NOx AND FINE PARTICLE EMISSIONS
INTRODUCTION
Over the past several years EPRI has supported a number of field test
programs at pulverized coal-fired utility boilers to evaluate state-of-the-art
particulate control devices, namely electrostatic precipitators and fabric
filter baghouses. A major objective of this continuing effort is to provide
the utility industry with basic information on particulate collection per-
formance with emphasis on particle size dependent collection efficiency, trace
element emissions and stack opacity. Operational and economic data are also
obtained to provide the necessary information for a complete assessment of
control technology options.
The field test approach used in these evaluations involves monitoring of
the boiler and combustion processes (i.e., the source of particulate matter)
during the particulate control device testing. As a consequence of this total
systems approach to field testing, and the emphasis on fine particle collec-
tion efficiency measurements, a conspicuous dependence emerged between N0X
emissions and the generation of fine submicrometer size particles in the
boiler. Specifically, it appears that suppressed fine particle generation is
associated with lower NO emissions. This observed dependence between N0„ and
X X
fine particles is the subject of this paper. Before presenting these results
and their ramifications, fundamental characteristics of particulate matter
emissions from coal fired boilers are first briefly discussed.
CHARACTERISTICS OF PARTICULATE MATTER EMISSIONS
The distribution of particulate matter (fly ash) mass over the particle
size range at the outlet of a 520 MW opposed wall fired boiler^^ is shown in
Figure 1. These data are especially significant since they represent the
first time that a bimodal particle size distribution was measured in the
field. The bimodal distribution consists of (1) a fine particle mode (aerosol
186
-------
spike) at approximately 0.1 micrometer diameter and (2) a large particle mode
which actually reaches a peak and then tails off to the larger particle
sizes. It should be emphasized that these data (and data contained in subse-
quent figures) are based on measurements taken before the particulate control
device and therefore represent boiler outlet conditions. The fine particle
mode was measured with an extractive Thermo-Systems electrical aerosol size
(mobility) analyzer. The large particle mode was measured in situ with a
conventional cascade impactor. Prior to discovery of the aerosol spike, it
was very common practice to artificially extend size distribution curves to
zero at the low end of the cascade impactor data.
The format of Figure 1 is well known to those involved in the particle
measurement sciences but may be a little confusing to those not familiar with
reduced particle size distribution data. For purposes of interpretation this
differential mass plot can be viewed as a mass histogram where the area under
the curve is equal to total mass. Thus, the contribution of particulate mass
within a specific size range to the total mass can be easily visualized by
comparing the area under the curve in the region of interest to the total
area. Clearly, the aerosol spike represents a very small fraction of the
total mass of particulate matter leaving the boiler. In this case, the mass
of the spike is roughly one percent of the total mass, or 99 percent of the
total number of particles. This may seem insignificant until one realizes
that conventional particulate control devices (electrostatic precipitators and
baghouses) typically exhibit reduced collection efficiency in this particle
size range. At this particular boiler, which was equipped with a large
electrostatic precipitator, the particles in this size range represented about
20% of the stack emissions on a mass basis.
It was apparent from these early tests that the fine particulate matter
can represent a significant fraction of total stack emissions. The importance
of this from an environmental standpoint stems from the fact that these small
particles (1) can grow once in the atmosphere due to agglomeration and contri-
bute to plume opacity and ambient visibility impairment; (2) serve as
condensation sites and catalyze atmospheric chemical reactions and (3) are
highly respirable. Furthermore, there has been much concern that these
particles might be enriched in trace elements.
187
-------
In retrospect, the discovery of the aerosol spike should not have been
totally unexpected. Flagan at Cal Tech^ independently predicted the
presence of the aerosol spike at about the same time that the above measure-
ments were made, based on a theoretical model of particle formation during
coal combustion. Figure 2 is a comparison of the previous data to theoretical
predictions based on Flagan's model of particle formation involving (1) high
temperature volatilization of ash components followed by (2) condensation and
coagulation of the volatile material as the gases cool toward the boiler
exit# Note that for comparison purposes between theory and field measure-
ments, the results are presented on a particle number basis in contrast to a
mass basis used in the previous figure.
Since the data in Figure 1 were obtained, a number of boilers have been
tested and they all to some degree exhibit an aerosol spike and a bimodal
particle size distribution. This leads to a conclusion that the aerosol spike
is a generic feature of fly ash generated in combustion processes occurring in
conventional pulverized coal boilers.
Figure 3 summarizes particle size distribution data from six utility
boilers of varying designs burning different coals. An obvious feature of
these data is that the large particle mode seems to be highly variable from
unit to unit. Careful examination of these data suggest that the large parti-
cle mode may in fact be a superposition of one or more discrete particle
modes. The shape and variability of the large particle mode is more than just
a simple effect of variations in total coal ash content and is an area of
research beyond the scope of this paper. For simplicity, the large particle
mode as discussed here is considered as a single particle mode.
The fine particle distribution is also highly variable which is more
obvious in a blow up of the fine particle region as shown in Figure 4. The
important aspects to note are (1) the aerosol spike is very monodisperse
(narrow) which is consistent with the homogeneous gas phase condensation
formation theory and (2) the spike appears to occur at essentially the same
particle diameter in all units.
Table 1 is a summary of the submicroo mass data shown in the previous
figures. Note that the ash content as evidenced by total mass emissions
varies nearly an order of magnitude from unit to unit. The percentage of
total mass represented by the fine particle spike is also variable. However,
188
-------
there is no simple correlation apparent between ash content and mass of
aerosol spike.
CORRELATION BETWEEN NOv AND FINE PARTICLES
X
It was implied earlier that there was an observed connection between NO
emissions and the quantity of fine aerosol generated in the boiler. Figure 5
is a summary of the aerosol spike mass data generated to date, plotted vs.
N0X« There is considerable scatter in the correlation which can be attributed
at least in part to a very poor quantitative understanding of fine particle
generation mechanisms. However, it is quite apparent that as N0X increases
the relative mass of the aerosol spike also increases. Although a correlation
between N0X and fine particulate matter was not originally sought or expected,
as more and more field data were collected it became apparent that a correla-
tion may exist. Theories of particle formation are still evolving and an
explanation can only be speculative. From a very simplistic standpoint one
might hypothesize that the temperature time history during combustion
processes may be a critical factor common to both formation of submicrometer
aerosol and N0X« One might argue that the hotter temperature which may
enhance N0x formation (especially in the predominately pre-NSPS boilers repre-
sented here) may also volatilize more coal ash constituents leading to the
relationship observed. Nevertheless, based on information to date it appears
that attempts to reduce N0X via combustion control may have a beneficial side
effect, i.e., reduce fine particle generation and emissions. The possibility
of continued beneficial effects at NO emission levels below the range of NO
X X
data presented here remains to be verified.
To illustrate the very rudimentary state of understanding of fine particle
generation and its connection with N0X, Figure 6 indicates the day to day
variability of the aerosol spike emissions observed at one plant. In this
particular case the quantity of subraicron aerosol varied significantly under
repeated boiler conditions. N0X and boiler load remained constant during the
tests.
189
-------
SUMMARY
Figure 7 reemphasizes the significance of the aerosol spike as it relates
to particulate emission control. Here the particle collection performance of
a very high efficiency baghouse and a large electrostatic precipitator are
compared.Note that in the region of the aerosol spike there is a minimum
in collection efficiency for both control devices. Obviously then the
presence of fine particles can have a pronounced impact on control device
performance and selection. It appears that this deficiency inherent to par-
ticulate control may in part be overcome by application of N0X combustion
control.
190
-------
REFERENCES
Ensor, D.S., et. al. Evaluation of the George Neal No. 3 Electro-
static Precipitator. EPRI Report FP-1145, August 1979.
Flagan, R.C. and S.K. Friedlander. Particle Formation in Pulverized
Coal Combustion — A review. In: Proceedings of the Symposium on
Aerosol Science and Technology at the 82nd National Meeting of
the American Institute of Chemical Engineers, Atlantic City,
New Jersey, August 29 - September 1, 1976.
Carr, R.C. Performance of Electrostatic Precipitators and Baghouses.
Proceedings of EPRI Topical Conference: Focus on Particulates,
EPRI Report P-80-2-LD, June 1980.
191
-------
dM
d log D
2800
(mg/m3)
2400 f—
200
100
2000 \— 0
1600 \—
1200
800
0.1 1.0
Lower section expanded by
5 times in magnitude
400 \—
Submicrometer
Mode
(Aerosol SpiKe)
I I I 11'
Large
Particle
Mode
0.01 0.1 1 10
Particle Diameter (micrometers)
Figure 1. Differential mass particle size distribution at
boiler outlet illustrating bimodal nature of
particulate matter (520 Mw coal fired boiler)
192
-------
dN
d log D
10^6
1014
1013
1012
1011
1010
109
(number/m3)
^— Theoretical
(Flagan and Friedlander, 1976)
_/
/
/
t
108
0.01
520 MW opposed
wall fired
I I I I I II ll
I I I Mill I I I I III
0.1 1 10
Particle Diameter (micrometers)
100
Figure 2. Comparison of measured differential number size
distribution with theoretical prediction
193
-------
dM
d log D
5000
(mg/m3)
4000
3000 —
2000 —
1000
Boiler designs
360 MW, front wall
540 MW, tangential
520 MW, opposed wall
113 MW, roof fired
360 MW, tangential
25 MW, front wall
'/Cascades
/ //* impactor \
'/V
Mobility
analyzer
0.01
0.1 1
Particle Diameter (micrometers)
Figure 3. Differential mass particle size distributi
for six coal fired uitlity boilers
194
-------
dM
d log D
GOO
250
200 —
(mg/m3)
360 MW, front wall
640 MW, tangential
520 MW, opposed wall
113 MW, roof fired
360 MW, tangential
25 MW, front wall
0.01
0.1 1
Particle Diameter (micrometers)
Figure 4. Aerosol Spike Differential Mass Particle Size
Distributions for Six Utility Boilers
195
-------
Relative Mass of Aerosol Spike
40
35
30
25
20
15
10
®
360 MW, front wall
~
A
540 MW, tangential
>.V
~~ 0
520 MW, opposed wall
4 ~
~
113 MW, roof fired
v'"'"'. -V'
v.\.
— D
360 MW, tangential
O
25 MW, front wall
• - - ^ 5L.V' -
00
T
Of
T
—
$
~
1 . •"•tr
«>
~
<8
D
A ££>
K
i
1
0
500 1000
Nitric Oxide, 3% O2 (ppm)
1500
Figure 5. Relationship Between NOx Emissions & Aerosol
Spi ke
196
-------
fM n (mg/m3)
d log D
300
250
200
113 MW, roof fired
150
Day 1
Long-term
average
Day 2
100
1
0.1
0.01
Particle Diameter (micrometers)
Figure 6. Variations in Aerosol Spike at Constant Boiler
Load
197
-------
Penetration (%)
100
0.1 —
0.01
s&w •
Aerosol
spike
' Efficiency (%)
-n 0
— 90
Electrostatic precipitator -
SCA = 745 ft2/kacfm -
520 MW
— 99
Fabric filter baghouse
A/C = 1.7 acfm/ft2 —
25 MW
99.9
i i mil ii I I I llllll I I i i mi
99.99
0.1 1 10
Particle Diameter (micrometers)
100
Figure 7. Size Dependent Collection Efficiency of Particulate
Controls Illustrating Minimum Performance in Region
of Aerosol Spike
198
-------
TABLE I
TOTAL & FINE PARTICULATE EMISSION SUMMARY FOR SIX UTILITY BOILERS
Boiler Description
Total Mass
grams/m3 (grains/ft3)
Mass <2fim
(percent)
Mass of Aerosol
Spike (percent)
360 MW, front wall fired
Western, low sulfur, subbituminous
10.5 (4.6)
7
0.3
540 MW, tangential fired
Eastern, high sulfur, bituminous
9.5 (4.2)
4
0.2
520 MW, opposed wall fired
Western, low sulfur, subbituminous
6.4 (2.8)
4
1.3
113 MW, roof fired
Western, low sulfur, subbituminous
3.4 (1.5)
8
2.2
360 MW, tangential fired
Western, low sulfur, subbituminous
2.3 (1.0)
20
0.9
25 MW, front wall fired
Western, low sulfur, subbituminous
1.7 (0.75)
8
0.5
199
-------
COMMERCIAL EVALUATION OF A LOW NOx COMBUSTION
SYSTEM AS APPLIED TO COAL-FIRED UTILITY BOILERS
By:
S. A. Johnson and T. M. Sonmer
Babcock & Wilcox Company
Alliance, Ohio 44601
200
-------
ABSTRACT
Development testing of an advanced, two-stage combustion system, capable
of limiting NO emissions from pulverized coal-fired boilers to less than 0.2
fi
LB NO^/IO Btu has been completed. Test programs have been conducted
on both a 1.2 MUy and 10.2 MW^. system. These tests have confirmed that NO
emissions can be correlated to a dimensionless parameter proportional to the
second stage flane temperature. In addition, scaleup criteria were formulated
which allowed the design of commercial scale low NO combustion systens.
A
Subsequently, a detailed engineering evaluation was performed on two
candidate applications of this technology. The objectives of that study
were: 1) to refine scaleup correlations and design procedures; 2) to concep-
tually design a steam generator incorporating the two stage combustion concept;
3) to economically evaluate that design as compared to a conventional, post-
NSPS steam generator design; and 4) to identify areas of commercial concern
with the new designs and to recommend further research to address these concerns.
This paper summarizes the significant results and conclusions from the
test programs and the engineering study. The favored Venturi furnace system
is expected to limit NO emissions from coal-fired boilers to less than
f
0.2 LB N02/10 Btu, while increasing the capital cost of the boilers by signifi-
cantly less than the projected costs of tail end NO removal systems to meet future
Strict NO emission standards.
X
201
-------
INTRODUCTION
In 1976, The Babcock & Wilcox Company (B&W) entered into a contract
with the Electric Power Research Institute (EPRI) to investigate practical
methods for minimizing the emission of nitrogen oxides (N0X) from coal-fired
utility boilers. The goal of this research was to achieve 100 ppm of NO^
using only combustion modifications. It was achieved, as reported in the
previous EPRI N0X Control Seminar [1], by utilizing an advanced staged com-
bustion technique called the primary combustion furnace.
The primary combustion furnace was conceptualized to limit N0X emissions
through effective control of temperature and oxidant availability. Unlike
conventional staged combustion, where separation of stages may be ambiguous,
the B&W process consists of two individual, physically separated combustion
chambers. The first stage is operated with substantially less than the
amount of air necessary to complete combustion. Under these conditions, the
fuel-bound nitrogen compounds tend to react to form molecular nitrogen within
the fuel-rich regions of the flame. After giving up some of their heat to
combustor surroundings, the products of partial combustion then pass into
the second-stage furnace where additional air is added to complete combustion.
N0X formation in the second stage can be further suppressed (or first-stage
N0X can be reduced) by limiting the second-stage, flame-zone temperature.
Initial testing of this concept took place on a 4-million Btu/hr
(1.2 MWT) device. These tests, aimed toward understanding of the parameters
which affect N0X formation and reduction, revealed that N0X could be limited
to 100 - 150 ppm by operating with a first stage stoichiometry of 60% - 80%,
as long as the second-stage, flame-zone temperature (measured) was less than
about 1800°F. A plot of NQX emissions, as a function of a characteristic,
measured flame temperature for these small-scale tests is reviewed in
Figure 1.
202
-------
Subsequently, prototype-scale testing was completed on a 35-million
Btu/hr (10.2 MWj) scale. The purpose of this portion of the test program
was twofold:
• To confirm that the effects of the various operating
parameters on N0x, noted during model furnace testing,
would occur in a similar manner on the prototype system.
• To develop a scale-up strategy, based on the results of
model and prototype testing, to be used to predict NO
A
emissions from a commercial system.
This information was then factored into a detailed engineering design
study performed by B&W's Fossil Power Generation Division. This study in-
cluded the following tasks:
t Refinement of scale-up correlations and design procedures.
• Design of a 650-MWg steam generator to accommodate the
advanced staged combustion concept.
• Economic evaluation of the new design as compared to a
conventional 650-MWe, coal-fired boiler built to meet
the original EPA New Source Performance Standard of
0.7 lb NO /I06 Btu.
/\
203
-------
• Identification of potential commercial concerns or un-
knowns included in the new design, as well as a recom-
mendation of research to address those concerns.
This paper is an update of the work accomplished since the last EPRI
NO Seminar in 1978. Results from the prototype test program are presented
A
and compared to previous model-scale data. Pertinent highlights of the boiler
design study and commercial evaluation will also be discussed.
PROTOTYPE-SCALE TEST FACILITY
The prototype-scale test program was conducted in a four-drum Stirling
boiler rated at 40,000 Ib/hr of steam. This boiler is located at B&W's
Alliance (Ohio) Research Center.
To transform this boiler into the Low-NO Combustion System (LNCS)
shown in Figure 2, a separate first-stage furnace was mounted on the burner
wall. This furnace consisted of a rectangular tunnel with a five-foot-square
cross section by about 13 feet long. It differed from the 1.2 MWy device in
that it was inclined at a 30° angle to minimize the accumulation of solids
on the furnace floor. The transition between the first and second stage
consisted of a converging section having a 55° included angle and terminating
at a rectangular opening 19-1/2 inches wide by five feet high. This furnace
was designed to accommodate the same residence time as the 1.2 MWj device
when firing at 10 MWy (35 million Btu/hr).
Cooling of the first-stage furnace was achieved by circulation of
pressurized water through the membrane-wall tubes. Total first-stage heat
absorption could be determined from the measured temperature gain of the
subcooled water, while chordal thermocouples were used to measure local
204
-------
heat flux at 24 locations in the first stage.
Second-stage air was injected through pivoting, rectangular-slot nozzles
located on either side of the first-stage furnace exit. These slots were sized
to duplicate the second-stage air velocities from the 1.2 MWy system.
As in the small-scale test facility, an ample number of sample probe
access ports were located in both stages of combustion. These ports are
diagrammed in Figure 3. As shown, N0X, CO and COg concentrations in the
stack flue gas were routinely measured and continuously recorded. Particulate
samples were also extracted from the stack for determination of combustion
completeness. Temperature profiles in the second-stage ignition zone were
taken for the majority of the test runs, while less routine measurements
included HCN, NHg, N0x» 0^, CO, C02» inflame solids and temperature profiles
in the first-stage furnace. It is impossible to report all these results in
the space allotted. (A full report is forthcoming from EPRI [2].) However,
the following sections describe the results having the most impact on the
design of a commercial unit.
COMPARISON OF MODEL AND PROTOTYPE RESULTS
Effects of input variables were investigated during both the model-
scale and prototype-scale tests. In this section, parametric influence
of pertinent variables on N0X emissions, as well as a comparison between
test results from the two combustion systems, are presented.
205
-------
Effect of First-Stage Air/Fuel Stoichiometry on N0x
The prototype tests verified that an optimum overall air/fuel
stoichiometry of 60% - 75% exists at which NO emissions could be mini-
A
mi zed. The effect of stoichiometry on NO is shown in Figures 4 and 5,
A
first, for the prototype system with and without flue gas recirculation
to cool the second-stage flame, and second, for comparison of the model
and prototype furnaces. At very low stoichiometrics, first stage NO
formation was low, but stack NO increased because the large amounts of
A
cyano species (>500 ppm) produced in the first stage were converted to
N0X through oxidation reactions in the second stage. This conversion was
enhanced by high, second-stage flame temperatures, since less heat was
released and removed from the first stage.
At high stoichiometrics, N0x formation in the first stage was not
as effectively suppressed. Even though reduction of NO still took place
A
across the second-stage flame front as a result of higher first stage heat
removal and lower, second-stage temperatures, N0x emissions remained high
because first-stage NO formation was high under local oxidizing conditions.
The reduction of NO noted across the second-stage flame front can be seen
A
by comparing the first stage exit NO line to the stack NO curves in Figure 4
A A * •
It can be seen from Figure 5 that NO emissions from the two furnaces
were very similar at any given first-stage stoichiometry. The prototype
furnace generally produced slightly higher N0X at a particular stoichiometry
due to decreased heat removal surface in the larger furnace.
206
-------
Effect of Gas Phase Stoichiometry on NO
Although it was convenient to correlate N0X emissions with overall
air/fuel stoichiometry, this parameter can be misleading when applied to
coal combustion because a significant fraction of the coal may remain in
the solid phase. This would be especially true in combustion systems
characterized by low temperatures or short residence times in the flame
zone.
The first stage of the B&W low NO combustion system was fired
A
with a low-turbulence burner under low-temperature conditions (2000-2500°F).
Stratification of the mixture inside the first-stage combustor can be seen
from Figure 6 in which contours of constant gas-phase stoichiometry are
plotted for a test run at an overall stoichiometry of 70%. The gas-phase
stoichiometrics were calculated from the measured gas composition at each
sampling plane located 3.3 feet (1 meter) and 6.7 feet (2 meters) from the
burner outlet.
This figure shows that local gas-phase stoichiometrics very seldom
approached the overall value. In fact, much of the flame periphery remained
under oxidizing conditions.
Since N0V formation occurs primarily in the gas phase [3, 4], it
/v
would seem that to limit first stage NO , one should minimize the gas-phase
stoichiometry. This can be seen from the contours of constant fuel nitrogen
conversion plotted on Figure 7. Since low conversion closely corresponded
to low, gas-phase stoichiometrics (as illustrated by the contour having a
gas-phase stoichiometry of unity superimposed on the plots), an attempt was
made to Increase the uniformity of the mixture by using an impeller to
207
-------
disperse the coal into the air stream. The resultant gas-phase stoichiometry
contours are shown on Figure 8. It can be seen that the local gas-phase
stoichiometires were much more uniform, but that little improvement was
made in achieving more reducing conditions. As a result, first-stage NO
X
changed very little. The largest effect of the change, however, was a 4%
decrease in combustion efficiency due to severe flame impingement on the
first-stage furnace side walls.
Effect of Second Stage Temperature on NO
Early in the model-scale test program, it became obvious that second-
stage, flame-zone temperature had an overriding effect on NO emissions from
A
this combustion system. With constant conditions in the first stage, the
second-stage NO could be increased or decreased, depending on the second-
A
stage, flame-front temperature. Additionally, the first-stage NO concentra-
tion could be increased by nearly a factor of three without significantly
changing stack NO concentrations. This startling result has been reported
A
elsewhere [1, 5] and need not be explained in detail here.
Unfortunately, the early correlation between N0X and second-stage
flame temperature involved a single characteristic measured temperature.
Since such a temperature could not be easily estimated during the design
of large units, another more meaningful correlation was necessary.
As a first attempt, NO emissions were correlated with a calculated
theoretical, second-stage temperature, which was defined as the temperature of
the second-stage mixture if combustion were completed under adiabatic second-
stage conditions. This correlation is shown on Figure 9. It can be seen
from this figure that a consistent relationship existed between NO and
X
208
-------
theoretical second-stage temperature. But, by segregating the data, distinctivi
curves could now be drawn for each furnace load. It is believed that the load
effect, which was not noted in the correlation with measured temperatures, was
probably caused by the fact that second-stage heat losses (clearly a function
of load) were not accounted for in the theoretical calculation. The success-
ful design correlation which normalized the load effect is described below.
DEVELOPMENT OF AN N0X CORRELATION PARAMETER (NCP)
An expression incorporating the parameters influencing second-stage
temperature, while normalizing the load effects, is given below:
Q 0 - P)
Qn (1 + f)
In this parameter, Q and QN are the load and the nominal load of the
furnace; 3 is the amount of first-stage heat removed, expressed as a
fraction of the fuel-energy input; and f is the amount of recirculated
flue gas expressed as a fraction of the total coal and combustion air-flow
rates. Nominal load was chosen as that load which would maintain mean-gas-
residence time similarity between model and prototype furnaces.
A closer look would reveal that this parameter is actually the com-
bination of the following two terms:
Q (1 - 3) Q_
Q (1 + f) Qn
The first term is representative of the theoretical second-stage temperature
209
-------
while the second term is a modifier showing the effect of load changes.
A linear correlation of NO emissions with NCP, established from a
A
least-squares fit of all data, is shown in Figure 10. Data plotted include
both model and prototype-scale Pittsburgh No. 8 bituminous coal test results.
These data cover a first-stage stoichiometry range from 50% - 75%, and FGR
from 0% - 30%. Load variation covered a range of roughly 2 to 1. It's
apparent the correlation is excellent.
Figure 11 is a plot of the Montana subbituminous coal data as a
function of NCP. Superimposed on this graph is the linear correlation
developed from the Pittsburgh No. 8 bituminous coal data. It can be seen
that the Montana subbituminous coal data are generally higher than the
correlation line by roughly 25 ppm, and that the scatter of data is slightly
greater. However, since the deviation is not severe, a separate correlation
line for Montana subbituminous coal has not been considered necessary at
this point.
It should be noted that the correlations shown in Figure 10 and
Figure 11 are valid only within a restricted range. In other words, the
correlation was developed for the optimized conditions only. All of the
data plotted are results from tests with optimum-slot, air-nozzle direction
(15°) and excess air level (about 3% excess 02). Also, since the correla-
tion is empirical in nature, extrapolation outside a certain range of
applicability would not be appropriate. The recommended ranges of applica-
bility for a set of input variables are listed in Table 1.
210
-------
It is also evident as shown by the data plotted in Figure 10 and
Figure 11 that low NO emissions can be consistently obtained with the LNCS.
A
As long as the operation was optimized with respect to slot angle, excess
air and reasonable load, N0x emissions were well controlled at 100 - 200 ppm.
ESTIMATION OF FIRST-STAGE HEAT REMOVAL
Determination of the NCP is straightforward. All terms in its make-up,
except the first-stage heat removal term, are basic input information. How-
ever, to complete the determination of the NCP, a method had to be devised
to predict heat loss in the waterwall, coal-fired, fuel-rich, first-stage
furnace. To achieve this end, a simple algebraic predictive equation was
developed by combining the energy-balance equation and the radiation-balance
equation for the first-stage furnace. This equation can be expressed as:
b + c - e HA/SC
(1 ~ a,) ^ "d —
In this equation, 3 is the amount of heat loss to the water-cooled furnace
wall expressed as a fraction of the fuel-energy input; is the first-stage
stoichiometry; a, b, c and d are coefficients determined by coal composition,
heating value and combustion air preheat; HA/SC is the furnace heat release
rate (the ratio between furnace-energy input and furnace surface area); F
Is the system mean, gray-body emissivity. Derivation and discussion of this
model equation has been detailed elsewhere[6].
211
-------
The accuracy of this predictive equation was evaluated by comparing
the predicted values with the experimental data. During the model and pro-
totype-scale testing of the LNCS, measurements of first-stage heat removal
were taken for two coals (Pittsburgh No. 8 bituminous and Montana subbituminous)
fired in two different size furnaces at air/fuel stoichiometry ranging from
50% - 90%, and load changes from 50% to 120% of nominal load. The performance
of the predictive equation was found satisfactory for all cases. Examples of
the measured and predicted heat transfer for the bituminous coal are shown
in Figure 12.
As the size of the first-stage combustor increases toward conmercial
scale, the furnace heat liberation rate increases. Therefore, at a constant
first-stage residence time, first-stage heat removal would be expected to
decrease as larger units are designed. This would increase the NCP and the
predicted NO emission. Fortunately, the predicted increase in NO emission
A A
as a function of furnace capacity is not expected to be sizeable, as shown on
Figure 13.
It can be seen that, under optimized conditions, scaling from 35 million
Btu/hr to 1 billion Btu/hr should only increase N0X emissions from 140 ppm
to 160 ppm. Alternatively, N0X emissions can be trimmed by providing more
heat removal surface area in the first stage. For the prediction of first-
stage heat absorption in very large furances, conventional predictive
procedures were shown to be valid.
212
-------
SCALE-UP PROCEDURE
The procedure for scale-up design of a low NO combustion system can
A
be outlined as follows:
1. Calculate nominal load using furnace dimensions (or vice versa).
2. Calculate first-stage heat removal using appropriate methods.
3. Calculate the NCP using the input information on load, coal type,
amount of F6R, etc.
4. Estimate stack NO concentration by using the correlation with
A
the NCP shown in Figure 8.
5. If the calculated NCP is less than 0.4 (lower limit of confidence)
take the NO value at 0.4. Do not extrapolate.
A
6. If the calculated NCP is greater than 0.8 (upper limit of confidence),
the designer has failed to take full advantage of the LNCS. Redesign
is recommended.
213
-------
DESIGN CRITERIA FOR COMMERCIAL APPLICATION
The laboratory data collected on both low-NO combustion systems
A
identified several design criteria essential for minimizing the emission of
nitrogen oxides. The most significant is the second-stage flame temperature.
A target adiabatic temperature of 2600°F (1700°K) was selected for the
commercial design to limit NO emissions to 0.2 lb/10 Btu. The heat trans-
A
fer correlation described earlier could then be used to determine the size
of the primary furnace necessary to achieve that second-stage adiabatic
temperature.
In order to ensure that the degree of second-stage mixing is similar
to that of the laboratory devices, the commercial system will be geometrically
similar in second-stage injection design. Primary furnace exit velocity, staged
air injection velocity, and the direction of both flows will be duplicated In
the commercial design. The design of the injection ports will be such that
velocity and direction can be controlled independently to accommodate changes
in load and second-stage combustion conditions.
The primary furnace will be fired with burners that are geometric du-
plicates of the laboratory versions. The heat input per burner will be on the
order of 150 - 300 x 106 Btu/hr (44 - 88 MWj). The small-scale operation
of the LNCS indicated that moderate changes in burner design had little
effect on N0X emissions. Therefore, the primary objective for the commer-
cial design will be for efficient, stable operation over a range of loads and
furnace conditions. The B&W dual-register burner has proven to be an effective
tool for limiting initial N0X formation in the primary furnace. Slightly
more than 20% of the total heat input to the commercial system will be absorbed
214
-------
by the primary furnace walls. Both the residence time and heat absorp-
tion were calculated at the optimum first-stage stoichiometric ratio of
0.7 to minimize NO emissions.
X
COMMERCIAL DESIGN
The boiler design task of the LNCS commercial application study resulted
in two different boiler designs. One design, the primary combustion furnace
(PCF), utilizes the multiple, primary furnace concept, while the Venturi
design has only a single primary furnace. A brief description of each design,
as compared with a conventional boiler design, will highlight the differences.
Conventional Design
The boiler chosen as a basis for comparing the LNCS designs to conven-
tional designs is shown in Figure 14. It is a 650-MWe, balanced-draft,
Babcock & Wilcox, Carolina-type radiant boiler of post-1971 design. It
is arranged with a water-cooled, dry-bottom furnace, along with conventional
superheater, reheater, economizer and air-heater components. The boiler was
designed to produce less than 0.7 lb/106 Btu of N0X (500 ppm) when firing a
midwestern bituminous fuel with 49 dual-register burners in a compartmented
windbox.
PCF Design
The first LNCS boiler design evaluated during this study is shown in
Figure 15. It is a balanced-draft, Babcock & Wilcox, tower-type radiant
215
-------
boiler also rated at 650 MWg. The tower design proved to be more economical
for this particular application because extra space was required around the
boiler to accormiodate the primary furnaces. This design eliminates any
interference by removing the large downflow section of the convection pass
found on the more conventional Carolina design. It is designed to be fired
with 28 primary combustion furnaces (PCF) and utilize conventional dry-bottom
furnace, superheater, reheater, economizer and air-heater components, while
producing less than 0.2 lb/106 Btu NO (150 ppm).
/\
An iterative, heat-rate analysis showed that the heat transferred from
the primary combustion furnace containments should be utilized in the boiler
itself rather than in the pre-boiler cycle. Circulation and manufacturing
considerations prompted the decision to make the PCF containments an integral
part of the furnace wall circuitry. However, to maintain adequate flow through
that circuit, five circulation pumps were added. In addition, a flue-gas
recirculation system was added for NO and steam temperature control.
A
Venturi Design
The design complexities associated with the PCF design prompted the
development of an alternative LNCS boiler design. The arrangement shown 1n
Figure 16 eliminates the primary combustion furnace containments by operat-
ing the entire lower portion of the boiler furnace at an air/fuel stoichiometric
ratio less than 1. The remainder of the combustion air is injected at the
constriction of the furnace formed by two wall arches. This arrangement has
been designated the "Venturi" furnace design.
216
-------
The primary furnace (below the arches) is fired with 28, high-input,
dual-register burners. There are no circulation pumps necessary and the
remainder of the boiler is almost identical to the base unit. By maintain-
ing proper first-stage heat removal, this design is also expected to meet
the 0.2 lb NO /1066 Btu emission goal.
A
ECONOMIC ANALYSIS
Assuming the data collected from the laboratory testing can be extra-
polated to the necessary scale, both LNCS steam generator designs should be
functionally satisfactory. However, an economic analysis indicated that the
Venturi design is much more cost effective. Table 2 is a summary of the
cost study.
The LNCS designs both have reduced gas temperatures exiting the
furnace. This requires additional convective heat transfer surface to be
installed to maintain design (1005°F) steam temperatures. This surface
accounts for the largest portion of the pressure-part cost increase.
Additional structural steel is also necessary to support the additional furnace
and convective surface and the weight added by the two furnace-wall arches in
the Venturi design. The PCF design, in addition to the items mentioned above,
has cost increases associated with the attachment of primary combustion fur-
nace containments and the need to use circulation pumps.
COMMERCIAL CONCERNS
In spite of the careful consideration given each of the system designs,
B&W is concerned that several phenomena, unique to staged combustion, could
217
-------
cause operating difficulties in the future. The primary concerns at this
time are first-stage furnace waterwall corrosion and carbon utilization. After
completing the engineering study of the conmercial LNCS systems, many early
concerns such as prohibitive cost, complicated control systems and operating
procedures, manufacturing and construction problems, development of flame
safety equipment and others proved to be unjustified.
The first concern arises from operating experience which has shown in-
creased tube metal corrosion in areas suspected of exposure to reducing
atmospheres. The nature of operation of the LNCS System will expose large
areas of furnace-wall tubes to reducing atmospheres.
The second concern, carbon utilization, is an area of concern due to
inexperience with staged combustion in utility boilers firing pulverized coal.
The second stage mixing of combustion air with combustion products and un-
burned fuel from the primary furnace, and its effect on carbon burnout,
have not been quantified. The relatively low temperatures in the second-
stage combustion zone could also reduce combustion efficiency.
CONCLUSIONS
As a result of a detailed engineering analysis of advanced staged
combustion techniques as applied to a new, pulverized coal-fired boiler,
it has been concluded that the EPRI/B&W Low N0U Combustion System could
be a viable commercial alternative to tail-end NO removal to meet future
A
strict EPA N0X emission standards (circa 0.2 lb N02/106 Btu). Due to its
relative simplicity and lower capital cost, the Venturi furnace described
previously most favorably embraces the concept. The PCF design, however, is
218
-------
still considered a retrofit possibility, especially for cyclone units oil-
fired boilers that a utility wishes to convert to coal.
Before offering this system as a commercial product, B&W will complete
several additional research programs to address the remaining concerns men-
tioned above. The results of these programs will help refine the LNCS
design and enable a more precise, second-generation estimate of material and
operating costs to be made.
FUTURE WORK
The commercial concerns noted previously could best be alleviated by
converting an existing utility boiler into a low-NO Venturi furnace based
A
on the design and scale-up criteria developed during the system analysis
study. If this were done, it would only take a few years to measure cor-
rosion rates under actual operating conditions to determine whether a
potential problem really exists. It would take a few weeks (using variable-
angle, second-stage injection ports) to determine whether trade-offs exist
between low NO emissions and acceptable carbon conversion. Most importantly,
A
such a test would provide day-to-day operating experience to finalize control
schemes and to assess reliability of system components.
Unfortunately, it is unlikely that such a test could occur in the
near future. Based on preliminary discussions with a number of utilities,
the following factors substantiate this opinion:
219
-------
t Retrofit of a Venturi furnace would be extremely expensive due to
pressure-part changes required in the lower furnace walls. Sub-
stantial funding from an outside agency (EPRI, EPA, DOE) would be
required for a project of this magnitude.
• A long outage in excess of 8 weeks would probably be required to
modify the unit. Most utility companies do not have enough reserve
capacity to afford such an extended outage.
• Because fewer new boilers are being built, most utilities are
stressing maximum availability of their existing units, even old
and inefficient units. Any test program which could disrupt the
production of electricity is not likely to be approved.
t At this time, many utilities are concerned that very strict NO
X
emission standards will prevent widespread committment to
coal because they perceive problems involving costs and plant
reliability. Until strict emission standards are promulgated
(requiring extensive combustion modifications or expensive tail-
end NO removal, they will probably remain unwilling to test
X
major modifications to reduce N0y.
In the meantime, B&W has undertaken the following programs to provide data
which will enable the risk associated with each concern to be quantified.
220
-------
Furnace Tube Wall Corrosion (Background)
Fireside corrosion of waterwall tubes in coal-fired utility boilers began
to be a problem in the late 1930s. The trend at that time of high furnace heat
release rates, coupled with increasing steam pressure requirements, resulted
in higher furnace-wall temperatures and led to increased deposits of slag on
furnace waterwalls. Presence of aggressive iron and alkali sulfates in the
slag layer coincided with severe loss of metal. It was further noted at this
time, that low, excess-air operation greatly aggravated this corrosion problem,
presumably by fluxing the slag layer and providing even more aggressive reduced
sulfur compounds (H^S, S, Fe$2) in the vicinity of the waterwall. Flame
impingement often accompanied corrosion in these units, and relatively high
concentrations of CO (0.5 - 6.0%) were measured in the vicinity of the wall,
indicating the presence of reducing conditions.
Staged combustion of coal also results in local, gas-phase reducing
conditions, as well as longer flames. During the testing of the primary
combustion furnace under this EPRI contract, for instance, CO concentrations
in a similar range (0.1 - 9.0%) were measured in both the model and prototype
first-stage furnaces.
Mechanisms for corrosion in a reducing atmosphere are still not totally
understood. Hydrogen sulfide (H2S) gas, which is present in quantities up
to 0.5% in the low NO^ system can be potentially harmful. Sulfidation attack
by HgS on metals in hydrogen-rich atmospheres has been observed in steam
reformers associated with many refinery operations. The questionable effect
of large quantities of hydrogen, however, makes it difficult to transfer the
refinery experience to utility boiler operation.
221
-------
Another possible mechanism for corrosion involves deposition of pyrite (FeSg)
on the metal surface, followed by reaction of the pyrite with the protective metal
oxide layer. Removal of the oxide layer by alternating exposure to both oxidizing
and reducing gases could lead to rapid metal loss. Deposition of other ash
constituents (like alkali sulfides and sulfates) can also cause corrosion by
diffusion of tube-metal iron into a slag layer under the deposit or by dissolving
the iron oxide layer into complex iron alkali tri-sulfates.
Tube-thickness measurements made during the prototype PCF test program in-
dicated that corrosion could be occurring on the floor of the first-stage furnace.
After about 300 hours of operation, the only points where measurable metal loss
took place were located on the furnace floor tubes where dry and partially fused ash
accumulated. It was also observed that flame impingement occurred on the furnace
floor, creating a local reducing atmosphere (see Figure 6). Gas samples taken
in a plane located 6.7 feet downstream of the burner showed that the roof and
sidewalls were exposed to local oxidizing conditions ranging in stoichiometry from
100% - 140%. The floor, however, maintained a gas phase stoichiometry of 85% - 95«g
Test Programs
It can be seen from the above discussion that, if waterwall corrosion takes
place in a low-N0x combustion system, it could occur as a result of one or several
imperfectly understood mechanisms. In the commercialization of the low NO
X
concept, it is of first importance to select potential materials for adequate
(>10 years) metal life. In order to achieve this goal, two fundamental corrosion
studies are planned:
222
-------
1. A screening test of candidate tube metals exposed to a gas-phase
atmosphere typical of that measured in previous low-NO combustion
A
tests.
2. A parametric variation of furnace conditions aimed at determining
the effects of various sulfur forms on corrosion rates of carbon
steel. Variables could include temperature, sulfur concentration,
alkali availability, oxidizing/reducing cycle, and gas phase
stoichiometry. Potential methods to control corrosion, such as
air curtains or flue gas inerting, could also be investigated.
In the metal screening tests, the following materials have been selected
for testing:
1. Carbon steel (SA210)
2. Croloy 2-1/4
3. Type 304 and 304L Stainless Steels
4. Types 309 and/or 310 Stainless Steel
5. Incoloy 800
6. Inconel 671
7. Aluminized carbon steel via diffusion coating, dipped coating, and
flame spraying
8. Chromized and Aluminized Croloy 2-1/4 and Carbon steel.
Two autoclave systems have been designed and installed at B&W's Alliance Research
Center to carry out corrosion tests of these materials.
223
-------
Three samples of each material will be exposed to the gas mixture (typical
of the concentrations measured at 70% stoichiometry near the exit of the first-
stage furnace) for three time periods. Corrosion rates for two of the three
samples removed after each time period will be determined by loss-of-weight mea-
surement. The third sample of each material will be examined by electron microscope
to determine composition and structure of the scale for comparison with unexposed
archive samples.
The objective of these tests is to compare the resistance of each material
to sulfidation attack under conditions of constant temperature and gas composition
The data should also provide some qualitative information concerning the effects
of time, temperature and localized sulfur activity on corrosion rates. Based on
this information, initial Venturi furnace designs will consider the cost/benefit
dichotomy of corrosion resistent materials in zones (such as the lower arch)
where highly reducing conditions are likely near the wall.
In the event that gas-phase sulfidation by H^S or S2 is not the dominant
mechanism of corrosion, B&W is planning a subsequent corrosion study beginning in
calendar year 1981. Additional tests beyond the screening tests will be required
at conditions more closely simulating the low-N0x furnace conditions, including
the effects of ash deposits. In this study, further information will be developed
concerning corrosion rates for only those materials that are deemed acceptable
from the material screening study. Although detailed plans have not as yet been
made, the following outline contains our current thoughts:
224
-------
• Three sets of samples should be provided for each material: bare
metal samples, samples coated with a uniform layer of FeS^ (iron
pyrite) and samples coated with a layer of sintered flyash from a
high-sodium western coal.
• Samples should be exposed at constant temperature to a typical
fuel-rich gas for 500, 1000, and 2000 hours.
• Samples should be exposed to various temperatures ranging from
500°F to 900°F.
• Samples should be exposed to periodic addition of 0^ to the gas
composition.
In addition to the measurement of corrosion rates, analysis may include
detection of S03 in the reactor exit gases, and sulfur forms existing in the
corrosion products. In particular, evidence of pyrosulfate or alkali tri-
sulfate will be sought. Sulfide penetration into the metal substrate will also
be measured since its presence could indicate future catastrophic failure of
materials that initially show low corrosion rates.
CARBON UTILIZATION (Background)
The concern that carbon burnout may be unacceptable in a full-scale Venturi
furnace operating in the low N0x mode stems from the following considerations:
225
-------
• During prototype testing of the Pittsburgh seam #8 bituminous coal,
large agglomerates of sticky particles which formed in the first-stage
furnace were not completely combusted in the second state. The sub-
bituminous coal, however, which did not agglomerate, achieved 99+%
burnout under all conditions. Carbon burnout for each of these
fuels is illustrated in Figure 17.
• Carbon burnout depends on the amount of time that the particle is
exposed to high temperatures in the presence of oxygen. The local
oxygen concentrations, in turn, depend on the penetration and
entrainment of the second-stage air jets into the first-stage exit
gases. In the prototype tests, the air had to penetrate across a
jet approximately 20 inches wide, but in the commercial Venturi
furnace, the fuel-rich jet is 21 feet wide.
Work done on the model and prototype low-NOx systems at B&W showed that there
was a slight trade-off between increased second-stage mixing (i.e., higher air jet
velocity or higher jet penetration angle) to achieve better burnout, and decreased
mixing to minimize second-stage NO . It is possible that the same conditions
considered optimum during the test programs would need to be refined in the
field. Therefore, B&W is studying the mixing process in a full-scale Venturi
furnace by utilizing advanced flow modeling techniques in order to show that com-
plete combustion is possible under low N0x conditions. Armed with the information
from this study, B&W will then optimize the second-stage mixing during the even-
tual field testing of the Venturi furnace concept.
226
-------
PRELIMINARY FIELD TESTING
The third major concern with commercializing the Venturi furnace concept
is that the use of flue-gas recirculation fans, along with operation in an
unfamiliar staging mode, may decrease the availability of early commercial
units. As stated previously, extensive field testing of a demonstration unit
is the only way to assess the impact of this concern. Before the demonstration
unit will be accepted by a customer, however, it may be advantageous to point
to preliminary field tests where boilers were subjected to "deep staging", i.e.,
burner air/fuel stoichiometries on the order of 0.6 - 0.8. Such tests could be
conducted on an existing boiler equipped with overfire air ports. B&W has begun
discussions with an electric utility that has expressed interest in participating
in such a test program.
SUMMARY
In the above sections, B&W has defined interim research aimed at further
defining the practical limit of NO control using advanced staged combustion
A
techniques. These programs result from a detailed engineering study of the
Venturi furnace, based on model and prototype test results. This low-NO
A
system, as designed, seems to be a viable alternative to either tail-end N0X
reduction or the EPA burner concept, both of which will be commercially demon-
strated in the near future. With the potential decrease in N0X emission
standards just around the corner, it would seem prudent for the utility industry
to carry this research to its final demonstration before the alternative technolo
are forced upon them.
227
-------
ACKNOWLEDGEMENTS
Much of the interpretation and analysis of the data presented here was
performed by Dr. Bob Yang. The authors are deeply appreciative of the technical
skills and philosophical guidance contributed by Bob.
REFERENCES
1. S. A. Johnson, P. L. Cioffi, T. M. Sommer, and M. W. McElroy. "The Primary
Combustion Furnace - An Advanced Low N0X Concept for Pulverized Coal Combus-
tion." Second EPRI NO Control Technology Seminar, 1978.
A
2. Subtask 3.2 Final Report: Results of the Prototype Primary Combustion
Furnace Tests, Palo Alto, Calif.: Electric Power Research Institute
Contract RP-899-1.
3. J. 0. L. Wendt, D. W. Pershing, J. W. Lee, and J. W. Glass. "Pulverized
Coal Combustion: NO Formation Mechanisms Under Fuel-Rich and Staged
Combustion Conditions.11 17th Symposium (International) on Combustion.
The Combustion Institute, August 1978.
4. D. W. Pershing and J. 0. L. Wendt. "Relative Contributions of Volatile
Nitrogen and Char Nitrogen to NO Emissions from Pulverized Coal
A
Flames." 83rd National Meeting of AIChE, Houston, Texas, March 1977.
5. T. M. Sommer, S. A. Johnson, and G. D. Lindstrom. "Further Development of
an Advanced Low N0x Coal-Fired Utility Boiler." ASME Winter Annual Meeting
Paper No. 79-WA/Fu-4, New York, New York, December 1979.
228
-------
6. R. J. Yang, S. A. Johnson, and M. W. McElroy. "Heat Transfer Modeling
for Two-Stage Pulverized Coal-Fired Combustors." Joint ASME/AIChE National
Heat Transfer Conference, ASME Paper No. 80-HT-114, Orlando, Florida,
July 1980.
229
-------
260 -
240
220
© ' O
coco
EQUILIBRIUM CURVE
3% O
LEAST
SQUARES
CURVE
FIT
O NORTH DAKOTA LIGNITE
O PITTSBURGH 8
~ MONTANA SUBBIT.
J 1 »
1500
1600 1700 1800 1900 2000
SECOND STAGE FLAME ZONE TEMPERATURE (°F)
2100
2200
Fig. 1 NOx emissions as a function of a measured second stage flame temperature
230
-------
STACK
COAL
BUNKER
COAL &
PRIMARY AIR
ARC STIRLING BOILER
PCF
12'
WEIGH
FEEDER
FIRST STAGE
AIR & FOR
RECYCLED
FLUE GAS
SECOND STAGE
AIR ft FGR
E-21
PA
FAN
PULVERIZER
air/
PREHEATER
Fig. 2 Prototype seal* tast facility
231
-------
2ND STAGE WINDBOX
rv>
oj
r\>
BURNER
WINDBOX
12'9" (TOP)
+ 30
FIRST STAGE
GAS, SOLIDS, TEMP.
3 PORTS, 15" VERT. SP.
1'K
M
STACK
GAS
SOLIDS
OPACITY
+
+
SECOND STAGE
6-INCH INTERVALS
GAS. TEMP.
FIRST STAGE EXIT
GAS, SOLIDS, TEMP.
1' INTERVALS
FIGURE 3 SAMPLE PORT LOCATIONS (PROTOTYPE TEST FACILITY)
-------
PROTOTYPE DATA
PITT. NO. 8 BITUMINOUS COAL
LOAD - 37 * 106 BTU/HR
3% EXCESS Oz
OSTACK NOv
15% FGR TO 2NO STAGE
| STACK NOx, NO FGR
i FIRST STAGE EXIT NOx /
v0/T
o o
• o
/
_l_
_1_
_1_
BO 70 90 110
FIRST STA6E STOICHIOMETRY (%>
4 Effect of first stage itoichiometry on NOx - prototype syitem
233
-------
PITT. NO. 8 BITUMINOUS COAL
3% EXCESS 02
OPROTOTYPE DATA
LOADS37 * 106 BTU/HR {106% NOMINAL LOAD)
FGR TO SECOND STAGE 15%
• MODEL DATA
LOAD=?4.1 x 106 BTU/HR (100% NOMINAL LOAD)
FGR TO SECOND STAGE 2*17%
\»
VSV
_L
_L_
_L
30 SO 70 90 110
FIRST STAGE STOICHIOMETRY (%)
130
Fig. 5 Effect of first stage stoichiometry on NOx - comparison of model and prototype
-------
TOP
A
BOTTOM
PLANE 1. 4 FEET FROM BURNER
TOP
WEST
EAST
BOTTOM
PLANE 2, 8 FEET FROM BURNER
Fig. 6 First stage flow patterns - prototype combustor
235
-------
TOP
BOTTOM
PLANE 1, 4 FEET FROM BURNER
TOP
WEST
EAST
BOTTOM
PLANE 2. 8 FEET FROM BURNER
7 First stag* fuel N conversion - prototype combustor
236
-------
TOP
WEST
EAST
BOTTOM
PLANE 1, 4 FEET FROM BURNER
TOP
WEST
EAST
BOTTOM
PLANE 2. 8 FEET FROM BURNER
FIGURE 8 GAS-PHASE STOICHIOMETRY PROFILES (RADIAL-VANE IMPELLER)
237
-------
300
PITT. NO. 8 BITUMINOUS COAL
FIRST STAGE STOICHIOMETRY - 60 - 75*
OVERALL EXCESS AIR — 3%
«LOAD - 6.2 * 106 BTU/HR
LOAD - 4.1 . 10® BTU/HR
2.6 H 10® BTU/HR
ALL MODEL FURNACE DATA
J L.
J L.
-I L
-I 1 I L
2200
2400 2600 2600
SECOND STAGE THEORETICAL TEMPERATURE [°F)
3000
Fig. 9 NO emissions correlated against a calculated second stage temperature
238
-------
PITT. NO. 8 BITUMINOUS COAL
Q MODEL FURNACE DATA
PCF STOICHIOMETRY - 60 - 75%
FGR ¦ 0 - 32%
PROTOTYPE FURNACE OATA
PCF STOICHIOMETRY - 50-75%
FGR *0-29%
_ 200
(N
o
j;
o
o
ui
6
«y
ff
ff
O
a
150 -
100
o
LEAST SQUARE
CURVE FIT
y • 13.7 ~ 208.4 x
50
I—
0.6
NCP
0.7
0.8
O.ft
Fig. 10 NOx correlation for Pittsburgh s««m #8 bituminous coal
239
-------
MONT. SUBBITUMINOUS COAL
O MODEL DATA
FIRST STAGE ST01CHI0METRY- 60 - 7B%
• PHOTOTYPE DATA
FIRST STAGE STOICHIOMETRY- 60 - 76*
o o
PITT. NO. • COAL
CORRELATION LINE
0.2
0.3
0.4
0.5
0.6
NCP
0.7
0.8
OS
1.0
Fig. 11 NOx correlation for Montana
240
(Dackar Saam) tubbituminous coal
-------
COAL; PITT. NO. t BITUMINOUS
MODEL FURNACE
LOAD sr 4.1 X 10® BTU/HR
0 MEASURED
- - ** PREDICTED
PROTOTYPE FURNACE
LOAD a 37 X 10® BTU/HR
O MEASURED
PREDICTED
_L_
J.
i
60 70 80
AIR/FUEL STOICHIOMETRY (%>
FIGURE 12 Comparison of actual and predicted haat removal
241
-------
CM
O
o
UJ
h
O
U1
c
cc
O
u
2
a.
a.
180
160
MEAN RESIDENCE TIME
IN THE FIRST STAGE FURNACE - 0.8 SEC.
AIR/FUEL STOICHIOMETRY = 70%
FLUE GAS RECIRCULATION = 15%
TO THE SECOND STAGE FURNACE
no
¦c*
rv
O
z
*
u
<
I—
CO
140
u
5
ill
ac
a.
120
100
10
100
1000
FIRST STAGE FURNACE CAPACITY (106 BTU/HR NOMINAL LOAD)
FIGURE 13 PREDICTED NOx EMISSIONS (FULL SCALE PCF)
-------
JACT
¦ii nr n
imtmyw
VQO *0» gN» uftmT
K I
litis.
FIGURE 14 CONVENTIONAL FURNACE DESIGN
243
-------
tl+v— q
'W "§ tjrttf
M
FIGURE 15 PRIMARY COMBUSTION FURNACE DESIGN
244
-------
FIGURE 16 VENTURI FURNACE DESIGN
245
-------
100
98
fc 96 —
3
0
£
1
O 94
o
K
<
u
K
V
<
PROTOTYPE DATA
EXCESS 023s3%
¦ MONTANA SUBBITUMINOUS COAL,
NO FGR
~ MONTANA SUBBITUMINOUS COAl,
15% FGR TO SECOND STAGE
# PITTSBURGH NO. 8 BITUMINOUS COAL
NO FGR
O PITTSBURGH NO. 8 BITUMINOUS COAL
15% FGR TO SECOND STAGE
92
90 —
B
88
50
60 70 80 90
FIRST STAGE STOICHIOMETRY <%l
100
Fig. 17 Carbon conversion efficiency - prototype furnace
246
-------
TABLE 1
Input Variable Range of Applicability
First stage stoichiometry 60-75%
FGR 0-20%
Slot Angle 0-30%
Overall Excess O2 2.5-3.5%
NCP 0.4-0.8
247
-------
TABLE 2. COST-COMPARISON SUMMARY
COST INCREASE
(* OF CONVENTIONAL BOILER COST)
Components Venturi Design PCF Design
Pressure parts 5.5 8.0
Flues and Ducts 0.75 1.3
Structural Steel 2.25 6.6
Primary Furnaces N/A 8.4
Circulation Pumps N/A 4.7
8.5% 29%
NOTE:
Percentages are based on a typical cost of $65/KW for the
conventional boiler design, which can vary depending on scope,
fuel, turbine heat balances, etc.
248
-------
THE DEVELOPMENT OF DISTRIBUTED
MIXING PULVERIZED COAL BURNERS
By:
D. P. Rees, J. Lee, A. R. Brienza and M. P. Heap
Energy and Environmental Research Corporation
Santa Ana, California 92705
249
-------
ABSTRACT
This paper summarizes work sponsored by the. EPA to establish generalized
design criteria for low emission pulverized coal burners. Data for single ami
multiple configurations in research furnaces at 50 and 100 x 106 Btu/hr are
presented with the current design of distributed mixing burners. N0X emissions
down to 100 ppm (0% O2, dry) have been obtained for bituminous coals under
acceptable burnout conditions by sub-stoichiometric burner staging. These data
show that the optimum burner zone stoichiometry is approximately 70 percent of
theoretical air for all burners tested to date.
250
-------
Section. 1
INTRODUCTION
Increased utilization of coal is a major factor in the U.S. efforts to
reduce imports of petroleum-based fuels. Although in the future gasification
and liquifaction will provide alternate coal-derived fuels, direct coal combus-
tion offers the only opportunity to make an immediate impact on fuel usage
patterns. This, paper addresses one problem associated with direct coal
combustion - the production of air pollutants in general and nitrogen oxides in
particular. The Federal New Source Performance Standards for boilers have
recently been revised, and for nitrogen oxides (N0X) these are dependent upon
fuel type (1, 2):
Anthracite, bituminous and lignite - 0.6 lb NOx/Million Btu;
Sub-bituminous - 0.5 lb NOx/million Btu;
Coal-derived fuels - 0.5 lb N0x/million Btu.
These limits are achievable by current commercial practice, however their
adoption will not prevent a significant increase in emissions of nitrogen
oxides if the amount of coal burned by stationary sources increases. Consequently,
there is considerable incentive to provide technology to reduce N0X emissions
well below those specified by the New Source Performance Standards.
Modification of the combustion process to minimize thermal and fuel NO is
the most cost-effective method of N0X control and is normally achieved by staging
the heat release process in such a way as to provide an initial fuel-rich zone
before all the fuel and air is mixed to allow burnout. Staged heat release can
be accomplished by physical staging i.e., dividing the combustion air into two
streams and injecting one of these streams into the combustion chamber some
d±BtPnr& away from the burner supplying the fuel and part of Che combustion air.
An alternate method of heat release staging involves use of the burner to provide
an initial fuel-rich zone. The EPA Distributed Mixing Burner (DMB) is one example
251
-------
of the latter N0X control technique. The DMB uses combined fuel injection
and air delivery characteristics to provide a flame envelope which can be
accomodated by existing boiler designs.
The major objective of EPA contract No.68-02-2667 is to provide infor-
mation which will aid the field demonstrations of the DMB and also allow the
design to be generalized to various firing configurations and the complete
range of U.S. coals. This paper concentrates upon pilot scale studies
concerned with burner development.
252
-------
Section 2
FACILITIES
The pilot scale test facilities used in these studies have been described
in detail elsewhere (3) and consist of two combustors:
• Small Watertube Simulator (SWS) with a firing range of 8-12 x 10s
Btu/hr.
• Large Watertube Simulator (LWS) with a firing range of 50-125x10®
Btu/hr.
Table 1 summarizes the dimensions and operating data for both combustors
which are kept cool during the experiments by spraying water on the outside
surfaces. Sample ports are located at the exit of both furances for collection
of gaseous and particulate samples. Continuous monitoring instruments are used
during the experiments for O2, CO, CO2 and NO effluent stack measurements. Both
furnaces are equipped with pulverizers that produce size distributions similar
to field systems. All input parameters (air temperature, flow rates, coal feed
rate, etc.) are monitored during the experiments.
Table 2 lists the properties of the test fuels. The baseline fuel is a
high volatile, Utah bituminous coal and to date two different shipments of this
coal have been test fired (Utah I and II). Other fuels tested are listed in
Table 2.
A sketch of the basic distributed mixing burner used in these studies is
shown in Figure 1. It consists of a central coal pipe surrounded by annular
channels for combustion air. The coal injector has a radial mixing device near
the exit plane and the secondary channel includes variable annular swirl vanes.
For burners consisting of two secondary channels, the amount of swirl in each
channel could be controlled independently. The fuel and air flow at the bumei
throat is overall reducing and tertiary air to provide burnout is supplied thr<
outboard air ports located as shown in Figure 1. The optimized dimensions and
settings for the DMB are listed in Table 3 for different scales.
253
-------
Section 3
SINGLE BURNER EXPERIMENTS
The effective stoichiometry immediately downstream of the burner throat
.s one of the most important parameters associated with low N0X burner design,
lata are shown in Figure 2 for a 50 x 106 Btu/hr burner in the LWS. NOjj emissions
lecrease steeply as the stoichiometry of the burner zone is decreased over the
range shown. CO concentrations remain unchanged until the burner zone stoichio—
netry reaches about 60-70 percent of the theoretical air required for stoichio-
netric combustion (SR3). At this point CO levels rise abruptly and overall
burnout decreases. This limit is associated with a failure to mix the reactants
adequately before their temperature has been reduced below some critical level.
Figure 3 shows the effect of excess air on NO emissions at fixed SRg for
two different burners in the LWS. (For these experiments, the level of excess
air was varied by changing the air flow through the tertiary ducts). It is
apparent from these data that NO concentrations are relatively insensitive to
the overall excess air level. Provided that the conditions of the burner zone
are chosen to minimize NO emissions, other parameters relating to the outboard
air (i.e., velocity, position of tertiary ducts, amount of air, etc.) have a
much more significant impact on CO levels and overall burnout than the N0X
emissions. This observation is consistent with data obtained using a scaled-
down version of this burner in a multiple burner configuration (see Figure 4) .
Other work on Distributed Mixing Burners (DMBs) to optimize the various
operating parameters has been reported elsewhere (3, 4). The parameters include:
• The amount of swirl of primary and secondary streams;
• The spatial location of outboard air ducts;
• The location of the coal injector;
• The effect of coal type fuel.
254
-------
As a result of these experiments, the design criteria for additional experimental
and prototype burners have been established. These are listed in Table 4. These
criteria were incorporated into burners to be tested in a multiple burner con-
figuration in the LWS. These data will be described in the next section.
255
-------
Section 4
MULTIPLE BURNER EXPERIMENTS
One important aspect of burner performance is the effect of burner
interactions which occur in a multiple burner array. These interactions
involve:
Shielding of flames in the center of the array from the cold walls
by other flames;
Variations in the mass and temperature of the gases entrained by
the flame jet;
Variations in mixing patterns caused by the confinement of one flame
by another.
In addition, only the smallest industrial boilers are fired by four burners or
less. Consequently, multiple burner experiments were conducted in the LWS to
study the performance of the distributed mixing burner in a multiple burner
array. A single-secondary DMB with a nominal heat input of 12.5 x 10s Btu/hr
was chosen for these experiments and the design criteria used were those listed
in Table 4. Four burners were installed in a 2 x 2 array on a single wall of
the LWS with a total heat input of 50 x 10s Btu/hr. Tertiary ports were placed
so as to service each burner in a manner similar to the single burner operation.
A schematic of the configuration is illustrated in Figure 4. The fuel used for
most of the tests was the baseline Utah bituminous coal (Utah 1).
The overall characteristics of the heat release pattern in the 2x2 burner
array were different from those observed when testing single burners in the LWS.
The individual flames ignited within the refractory divergent and extended into
the furnace as discrete flames for 2-3 feet. At the intersection of the indivi-
dual flames and the tertiary air jets, the four flames joined to form a single
large diffuse flame that extended into the furnace 10-12 feet. If a slight
perturbation affected one of the burners, this was not reflected in the larger
flame pattern. A large perturbation, however, such as a flameout in one of the
256
-------
burners, would significantly alter the flame shape and burnout characteristics
(CO levels and opacity of effluent gases). It was possible to test the multiple
burners over a larger operating map than was possible for single burners probably
because interactive effects improve burner stability.
NO and CO emissions are shown in Figure 5 over the range of excess air
levels studied. NO levels lower than 125 ppm (0% Oz, dry)0.14 lbs NO2/IO6 Btu
were obtained before the CO levels began to be affected. In addition to the
decrease in O2 availability in the burner throat, changing the SRg also changes
the velocity in the secondary and tertiary air channels and thus might alter
the overall mixing pattern. These NO levels are similar to those obtained with
single burners (see Figure 3) at the same heat input. Burnout characteristics
were generally better with the multiple burner configuration than with single
burners.
The variation of NO and CO with load is shown in Figure 6. It was not
possible to turn the burners down lower than 50 x 106 Btu/hr due to the capacity
of the coal mill. Therefore, the burners were overfired up to 147 percent of the
nominal design value (50 x 10® Btu/hr). The increased heat release resulted in
higher NO concentrations and slightly lower CO levels. The stability and other
operating characteristics were not noticeably affected. Single data points at
loads of 80-90 x 10s Btu/hr were tested and it was possible to maintain low NOx
levels (<0.2 lb NO2/IO6 Btu) by staging the burners to low SRg (0.5-0.6). Burn-
out and CO characteristics were the same as, or better than, those obtained
operating at the 50 x 10® Btu/hr heat input design point. Thus, the DMB design
can compensate for conditions which would result in higher NO emissions by
staging the burner to lower SRg values.
Effect of Tertiary Parameters - Two parameters associated with the injection
of tertiary air through the outboard air injectors were investigated: air
velocity and removal of some tertiary ports from operation. The effect of out-
board air velocity is shown in Figure 7. The nominal velocity is approximately
75 feet per second and the high velocity condition (twice normal), was achieved
by reducing the flow area at the duct exit. The N0X emissions were essentially
257
-------
the same at both conditions and CO concentrations were also very similar. How-
ever, overall flame stability was reduced by increasing the outboard velocity.
Small perturbations in the input conditions (air flow rates to the burner, coal
feed rate, etc.) usually resulted in pulsed flickering of the flame, increased
CO emission and on occasions, flameouts.
It was also possible to direct the flow of air to selected rows and/or
columns of tertiary air system in the burner array (see Figure 4). This
usually resulted in decreased flame stability and increased CO emissions. These
changes did not have a significant impact on N0X emissions. A summary of these
data is tabulated in Table 5. The performance of the multiple burner system was
most effected when the middle row or column was removed from operation. The
results obtained with four burners are consistent with data obtained in the
single burner experiments where NOx emissions were more sensitive to burner zone
parameters than to outboard air variables.
Fuel Effects - Three different bituminous coals were tested in the multiple
burner configuration. The fuels are listed in Table 2 and the results for the
fuels tested are shown in Figure 8. There were no observable differences in the
general burner performance with each fuel. The N0X emissions were nearly
identical over the range of conditions tested. This is in agreement with other
work in the SWS with these fuels. More work is required on fuels exhibiting a
wider range of fuel properties (nitrogen, ash, heat content, water content, etc.)
to formulate specific conclusions as design criteria relating to different fuels.
258
-------
Section 5
SUMMARY
The pilot-scale testing of low NQX burners has lead to formulation of
specific design criteria for prototype distributed mixing burners. It is
evident from these tests that burner and operational parameters affecting the
burner zone can have a significant impact upon N0X emissions. The most
predominant variable identified thus far is the stoichiometry of the burner
zone. Approximately 60-70 percent of the theoretical stoichiometric air for
a given fuel is the optimum operating point. This same trend was observed for
both single and multiple burner configurations in the LWS research furnace. It
is also evident from the tests that factors affecting the outboard air have
little impact on N0X emissions over the range studied. These parameters do,
however, have an impact upon fuel burnout, flame stability, and the stable
range of burner operation. Results from single and multiple burner configuratic
are very similar. For the bituminous coals studied, the NOx emissions appear tc
be relatively independent of fuel properties. This was demonstrated in both
single and multiple burner configurations.
259
-------
Section 6
REFERENCES
1. Federal Register, Vol 44, page 33580 (June 11, 1979).
2. Tabler, S.K., "Federal Standards of Performance for New Stationary
Sources of Air Pollution", J. A. Pollut. Control Assn. 29(8).803 (1979).
3. Brienza, A. R., et al., "Development of Criteria for Extension of
Applicability of Low Emission, High Efficiency Coal Burners", Second
Annual Report, EPA Contract No.68-02-2667, G. B. Martin, Contract
Officer, Research Triangle Park, N.C. (July 1980).
4. Folsom, B. A., L. P. Nelson, J. Vatsky and E. Campobenedetto,
"Distributed Mixing Burner (DMB) Engineering Design for Application to
Industrial and Utility Boilers", Special Report, EPA Contracts 68-02-3127
and 68—02—3130, G. B. Martin, Contract Officer, Research Triangle Park,
N.C. (1980).
260
-------
Tertiary Air Port
Adjustable
Swirl Vane
Primary Air
and Coal Inlet
ro
O
Oil Gun
for Ignition
Fixed Vane
Primary Air Swirl
Cast
Refractory *-
Exi t
Secondary
Air Inlets
Figure 1. Single Secondary Burner(SSB) Used in Pilot-Scale Experiments.
-------
i 1 1 1 1 r
50 x lQk Btu/hr SSB
UTAH COAL
SRp= 0.25
SRt= 1.2
t—I—r
Q
J I I I 1 I I L
10
60
80
100
120 WO
40 60 80 100 120 WO
BURNER ZONE STOICHIOMETRY (X TA)
Figure 2. NO/CO Emissions from A Single-Secondary Burner in the LWS-
-------
500
400
ac
Q_
Q_
300
(X
a
ot
S 200
o
—i r~
SSB 50M Btu/hr
UTAH COAL
SRp - 0.26
SRb - 0.61
100
110
J L.
120 130
110
_J
150 160
500
400
Q.
a.
300
ae
a
o
o
o
200
100
1
I
1 1
SSB 100M Btu/hr
UTAH COAL
_ SRp - 0.26
SRb* 0.70
-
A
"O—50^
r
—
x srb-
0.50
—
1
l
1 1
110 120 130 M0 150
OVERALL ST0ICHI0METRY (Z TA)
160
Figure 3. N0X Emissions for Single Burners in the LWS.
263
-------
Columns
II
Qt 12.5—O
30
Burner1
j.
Tertiary
Ports
M2.5
pp o
o
III
o
Rows
A
O
o
Dimensions: Inches
Figure 4. Schematic of Four Burner Array in the LWS.
264
-------
N)
as
on
500
£400
i | r
SSB: 2x2 ARRAY
UTAH COAL
SRp - 0.23
R
S% = 0.7
a 100
01 L_
110 120
X
X
1000
130 140 150 160
OVERALL ST01CHI0METRY
A A A
0 120
(Z TA)
150 160
Figure 5. Emissions Profile of Multiple Burner Array in LWS.
-------
400
T
T
300
200
Q£
a
^ 100
UTAH COAL
SRp = 0.25
SRb - 0.71
W EXCESS 02
o
a_
0
50
60
70
80
90
UJ
o
X 106 Btu/HR
LOAD
Figure 6. Emissions at Various Loads in
Multiple Burner Array in LWS.
266
-------
500
g-100
~300
o
t i r
SSB: 2x2 ARRAY
UTAH COAL
SRp = 0.22
SRfi = 0.70
^ 100 -
o
<_>
El
O NORMAL
~ 2 x NORMAL
X
110
120
130
140
150 160
OVERALL ST0ICH10METRY (Z TA)
Figure 7. Effect of Outboard Air Velocity upon NO/CO Emissions from
Multiple Burner Tests in LWS.
-------
500
1 1 1 1 1
1 1 1 1
SSB: 2x2 BURNER ARRAY
400
_ UTAH COAL
2EZ
O-
SRp - 0.22j SRj =
1.26
Ql_
£300
Q
—
—
CN|
O
g200
—
o
O
UTAH I
100
- ~
UTAH II —
A
W. VA,
0
1 1 1 1 1
till
50 70 90 110 130
BURNER ZONE STOICHIOMETRY « TA)
Figure 8. NOx Emissions from Bituminous Coals- in
Multiple Burner Tests in the LWS.
268
-------
Table I. Dimensions of Research Furnaces
•
Small Watertubes
Simulator (SWS)
Large Watertube
Simulator (LWS)
Designed Firing Rate (Btu/hr)
12 x 10s
50 to 125 x 106
Geometry
Horizontal Trapezoidal
Tunnel
Vertical Rectangular
with Tapered Bottom
Dimensions
5'6" H x 410" W x
10'0" L
59' H x 16' W x 28' D
Firing Depth (ft)
10
18
Surface Area (ft2)
225
5050
Furnace Volume (ft3)
207
20950
Firing Arrangement
Wall
Wall/Corner
Burner Arrangement
Single
Single/Multiple
Wall Cooling
Water-wall
Water-wall
Air Preheat (secondary)
70-500°F
70-700°F
Pulverizer (both on-line)
Hammer Mill
Bowl (C-E Raymond)
Primary Air
Unvitiated, Ambient
Temperature
Vitiated, 150-170°F
269
-------
Table II. Analysis of Fuels Used in Pilot Experiments
PROXIMATE,
% AS RECEIVED
UTAH I
UTAH II
W. KENTUCKY
W. VIRGINIA
Moisture
6.39
7.41
5.46
1.29
Volatile Matter
38.89
38.84
36.63
31.01
Ash
7.40
8.83
8.33
13.76
Fixed Carbon
47.32
44.92
49.58
53.94
Heating Value (Btu/lb)
12,340
11,877
12,392
12,500
ULTIMATE,
(% DRY)
•
Carbon
73.17
72.24
73.42
72.37
Hydrogen
5.55
5.76
5.12
4.88
Nitrogen
1.42
1.54
1.59
1.34
Sulfur
0.68
0.76
3.46
1.77
Ash
7.91
9.54
8.82
13.76
Oxygen (by difference)
11.37
10.16
7.59
5.88
270
-------
Table 3. Design Information for Single Secondary Burners
Design Variable 10
Fuel Injector
Diameter (in.) 5
Area (in.2) 19.63
Swirl Angle (degrees) Variable
Setback (in.) 0
Secondary Air Channel
Outer Diameter (in.) 9.5
Annulus Thickness (In.) 4.5
Area (in.2) 51.25
Swirl Vanes (degrees) Variable
Setback (in.) 0
Tertiary Ducts
Distance from Burner Q. (in.) —
Spacing Around Burner (degrees) 22.5
Number of Ports 16
Injection Angle (degrees) 0
Axial Position (in.) 0
Diameter (in.) 4.5
Total Area (in.?) 254.5
Throat and Exit
Throat Diameter (in.)
Throat Area (in.')"
Half Angle of Exit (degrees)
Length of Exit (in.)
Burner Capacity (10^ Btu/hr)
12.5 50 100
5.5 10.5 15.75
24 86 195
45 45 45
0 0 0
10.5 20 28
2.5 4.75 6.1
63 228 421
60 60 60
0 0 0
22 44 62
90 90 90
4 4 4
0 .0 0
0 0 0
6 12 16
113 452 804
10.5 20 28
87 314 616
25 25 25
9 (18.5)* 18.5 27
-------
Table 4. Prototype Distributed Mixing Burner Design Parameters
Parameter
Temperature (°F)
Stoichiometry (% T.A.)
Velocity (ft/sec)
Swirl
Fuel System
Primary
Primary
Primary
Primary
Air System
Secondary
Temperature (°F)
Burner Zone Stoichiometry (% T.A.)
Inner Swirl
Outer Swirl
Axial Velocity (ft/sec)
Inner/Total Area
Inner/Total Flow Rate
Tertiary
Temperature (°F)
Swirl
Axial Velocity (ft/sec)
Angle (degrees)
Number
Location (radius/throat diameter)
Divergence (degrees)
Exit
Half Angle (degrees)
Length/Diameter
Setback - Inner Secondary (inches)
Setback - Outer Secondary (inches)
Operational Variables
Capacity (10^ Btu/hr)
Turndown {% capacity)
Overall Stoichiometry {%
T.A.)
Nominal
Oesign
Point
S. 0.
S. 0.
75
450 Vanes
S. 0.
50-70
Variable
Variable
• 60
0.33
0.33
S. D.
None
50
0
4
2.2
0
25
1.0
0
0
S. D.
S. 0.
S. 0.
Testing
Range
130-180
17-30
50-90
Variable
400-650
40-120
Variable
Variable
50-90
Variable
Variable
400-650
40-60
Variable
Variable
Variable
»5Q
100-150
272
-------
Table 5. Summary of Effects of Tertiary Air
Upon Multiple Burner Operation
ROW OR COLUMN* RESULTS + COMPARED TO
CONDITION OUT OF SERVICE BASELINE OPERATION
NO CO
I
Col. I or III
18% Low
42%
High
II
Row C
25% Low
68%
High
III
Row B
18% Low
32%
High
IV
Column II
No Change
45%
High
V
Row B and
Column II
No Change
45%
High
VI**
All in Service;
Velocity 2 x Normal
Operation
No Change
No Change
* Refer to Figure 4 for designation.
+ Burner zone and overall stoichiometry were kept constant when
rows or columns were taken out of service.
** Stable range of operation not as great as normal baseline.
273
-------
JAPANESE TECHNICAL DEVELOPMENT
FOR COMBUSTION NOx CONTROL
By!
K. Mouri and Y. Nakabayashi
Electric Power Development Company, Ltd.
8-2, Marunouchi 1 chome, Chiyoda-ku,
Tokyo 100 Japan
274
-------
ABSTRACT
Electric Power Development Co., Ltd. has been executing a research and
development program on combustion N0x control for coal-fired boilers in
cooperation with Japanese boiler manufacturers. Target NO emission levels of
100 ppm or below (O2 ¦ 6%, N ¦ 1.8%) have been defined and to date have not
been achieved.
However, results obtained from this program have been applied in a
stepwise manner to existing or new coal-fired power plants. NO emission
levels for existing plants have been reduced to 160 - 300 ppm ($2 ™ 6%,
N = 1.2%) in comparison to uncontrolled levels of 400 - 500 ppm. Regarding
new coal-fired boilers, EPDC is constructing two 2 50-MW boilers with target
emission levels of 250 ppm (0 = 6%, N * 1.7%) at the Matsushima Thermal Power
Station. In addition, plans are to construct a 700 MW boiler at Takehara
(No. 3 unit) with target emission levels of 200 ppm. Presently, EPDC believes
that N0X emission levels will be 150 ppm (O2 ¦ 6%, N ¦ 1.8% design base) for
the No. 3 unit boiler.
This report describes ongoing R&D programs, and the results of combustion
modification efforts (such as the technology development of low NO burners,
two-stage combustion, gas mixing, etc.), and low N0x countemeasures for new
boilers.
275
-------
ACKNOWLEDGEMENTS
We sincerely appreciate the joint researchers of EPDC for assistance in
writing this paper.
276
-------
Section 1
INTRODUCTION
As coal use is expanded, environmental issues become more important. In
general, most of the environmental risks associated with coal combustion are
amenable to control.
These environmental impacts differ because of regional characteristics
such as meteorology, population density and resource distribution. Thus, it
is not surprising that nations and regions take different approaches to
environmental control measures.
Japanese government environmental standards are the most stringent in the
world. Local governments may impose standards on pollutant sources which are
more stringent than required by national law. Almost all local governments
have independently enacted pollution control ordinances. Thus, power util-
ities in Japan must take countermeasures to meet the stringent standards and
pollution control ordinances of local governments.
N0X emission levels of coal-fired boilers are remarkably high in
comparison to the other fossil fuel-fired boilers. Development of N0x control
technologies for coal to meet emission levels achievable for LNG or oil
combustion is very difficult.
Electric Power Development Co., Ltd. (EPDC) has been developing N0x
control and removal technologies in cooperation with Japanese boiler
manufacturers. Two methods to reduce N0x emissions from boilers have been
pursued. The first is postcombustion N0x removal, such as Selective Catalyst
Reduction (SCR) and Selective Non-Catalytic Reduction (SNR). The second
method focuses on combustion N0x control such as low N0x burners and combus-
tion modification.
EPDC and Japanese manufacturers have been has been conducting R&D
programs on both methods. This paper describes the results of the combustion
NO control program, while the results of using postcombustion methods are
rep0rted in another paper at this Symposium and for convenience summarized in
Appendix A. 277
-------
Section 2
N0x CONTROL REGULATIONS IN JAPAN
STATUS OF AMBIENT NO LEVELS
x
In August 1967, the Basic Law for Environmental Pollution Control was
enacted for the purposes of protection of national health and the preservation
of the of living environment.
The Japanese government selected the environmental quality standards in
May of 1973 and revised them in July of 1978. These environmental quality
standards are shown in Table I. The revised standards are slightly less
stringent than the initial standards.
After the revised standards were enacted, regional noncompliance areas
(over 0.06 ppm) consisted of 4.6% of all general environmental-atmosphere
stations (892 sites) and 36% automobile influenced stations (182 sites). The
current compliance status with the N0x environmental quality standard is shown
in Figure 1. Figure 2 shows that the emissions N0x density increased from
1971 to 1973, but has remained approximately constant since then.
N0x EMISSION REGULATIONS
NO emission regulations for stationary sources were established in
August 1973, and revised four times prior to the present environmental quality
standard. Table II shows the progressive changes in the NO emission stair-
dards to the current levels of 400 ppm (at 6% O2) for coal-fired boilers and
150 ppm for oil-fired boilers (at 4% O2).
POLLUTION CONTROL ORDINANCES OF LOCAL GOVERNMENTS
Pollution control ordinances imposed by local governments are generally
more stringent than the national standards. Examples of local ordinances
imposed on existing coal-fired plants are shown in Table III. Local
ordinances applicable to new coal-fired plants are shown in Table IV.
278
-------
Matsushima City is located in Nagasaki Prefecture of western Japan. Here
NO emission controls are less stringent than elsewhere since the area is not
x
heavily industrialized. Thus, the Matsushima thermal power station does not
employ SCR technology.
However, at Takehara thermal power plant where the No. 3 unit is cur-
rently under construction, another power company is planning to construct a
large thermal power plant nearby. In this instance, the local government has
requested the power companies not to increase total NO^ emissions beyond
present levels. Thus emissions from Takehara Unit No. 3 and the new power
station must comply with the stringent controls shown in Tables IV and V.
This situation has required EPDC to install both SCR equipment and combustion
NO control technology on Takehara Unit No. 3.
TOTAL MASS N0x CONTROL
Six industrial areas in Japan (e.g. , Tokyo and Yokohama) do not now meet
the established N0x ambient air quality standard of 0.OA ppm. Local govern-
ments in these areas are considering stringent NOx controls to achieve this
standard.
In August 1977, the city of Yokohama (the leading local government for
pollution control), proposed an N0x Guidance Plan that would reduce NOx emis-
sion levels 36% by 1981 from major industries. Since EPDC's Isogo thermal
power station is located in Yokohama, a 36% NO^ reduction may be required.
The NO Guidance Plan stipulated that the time limit to retrofit combustion
controls extended until April 30, 1979. If SCR was used, this time limit
extended to March 31, 1981.
EPDC has elected to apply combustion modification because the Isogo Power
Station is too narrow to install the SCR equipment. The combustion modifica-
tions chosen are described in this paper.
279
-------
Section 3
OBJECTIVES OF COMBUSTION CONTROL DEVELOPMENT
FOR COAL-FIRED BOILERS
EPDC has two objectives in NOx combustion control develoment. The first
objective is to develop a technology that will meet existing N0x control regu-
lations. The second is to reduce cost of control compared to SCR.
N0X emission levels from the combustion of oil or LNG are low in compar-
ison to coal-fired boilers. If advanced combustion control technologies were
applied to oil or LNG boilers, N0x emissions would be less than 100 ppm for
oil and 50 ppm for LNG.
Reductions to these levels may not justify the installation of SCR equip-
ment for oil and LNG boilers in most areas of Japan. To date, however, the
Shinkokura power plant and other oil-fired plants in Japan have installed SCR
equipment.
For coal, advanced N0x control technology would enable stack gas N0x
levels to be reduced to only 200 ppm (O2 ¦ 6%) because of the inherent high
nitrogen content of the fuel (Table VI). In Japan, environmental preservation
takes precedence over other factors such as the economy and the desire to
reduce oil and gas consumption. Thus, local governments and the public have
demanded lower N0X levels in order to approve new facilities. These actions
have forced N0x emission levels for coal-fired boilers to comparable levels
for oil-fired units. In most cases, this requires the application of SCR
equipment. However, EPDC is also developing combustion N0x control technology
to obtain the most economical and lowest emission coal-fired power plant
through a combination of combustion and SCR control technology.
EPDC'S TARGET OF COMBUSTION N0x CONTROL TECHNOLOGY
EPDC's target emission level obtainable with combustion N0X control
technology is 100 ppm (and below) at 6% 0£ and 1.8% fuel nitrogen content.
Figure 3 shows three cases of a model low pollution coal-fired thermal power
plant.
280
-------
Case lis Takehara No. 3 unit, which is under construction. Takehara
No. 3 will have SCR equipment and be capable of N0x emission levels of 60 ppm
and below.
Case 3 employs a Dry FGD system, which does not have SCR equipment, but
the N0x emission level is aimed at 60 ppm and below. This is due to capture
of NO within the FGD system to the degree that an inlet NO concentration of
X X
100 ppm will be reduced to 60 ppm at the FGD exit. Case 3 has been determined
to be the most economical, reliable, and least polluting coal-fired plant in
EPDC's analyses.
281
-------
Section 4
EPDC COMBUSTION NOx CONTROL DEVELOPMENT PROGRAMS
Since 1977, EPDC has been conducting R&D programs for low NOx emission coal-
fired boilers in joint research with Japanese boiler manufacturers, such as
Mitsubishi Heavy Industries (NHI), Babcock Hitachi K.K. (BHK), Ishikawajima
Harim Heavy Industries (IHI) and Kawasaki Heavy Industries (KHI).
This R&D, using test furnaces owned by boiler manufacturers is testing
various types of imported coals (from Australia and China among others)
planned for use in new coal-fired power stations. The range of nitrogen
content for coal stocks from these countries is shown in Table VI.
An outline of EPDC's R&D program for combustion N0x control technology Is
shown in Table VII.
Testing is being performed as follows:
• Survey tests to achieve lower N0x emissions by optimizing conditions
of two stage combustion, gas mixing, etc., and
• Confirmation that CO, unburned hydrocarbons, unburned carbon and the
other pollutants do not increase.
Recently, our joint researchers built a multi-burner test furnace. We
anticipate obtaining good data concerning the interacting effects between
burners.
282
-------
Section 5
R&D RESULTS OF TEST FURNACES
NOx REDUCTION LEVEL
One sample of test results is shown in Figure 4. Test conditions are foi
1.7% and 0.9% nitrogen content in coal and for (1) a conventional burner,
(2) a new low NO burner without Two Stage Combustion (TSC) or Gas Recircula-
tion (GR), (3) a new low N0x burner with TSC but without GR and (4) a new low
NOx burner with both TSC and GR.
Under the best conditions, NOx emission levels are between 160 ppm and
210 ppm, (unburned carbon content below 5%) for 1.7% N, and between 90 ppm and
150 ppm for 0.9% N.
An outline of the results for the test furnaces is as follows:
• The new burner Improves N0x levels by roughly 40 ppm to 60 ppm.
• TSC improves N0x level 60 ppm to 100 ppm.
• TSC combined with GR improves NOlevels by 80 ppm to 100 ppm.
• Higher fuel nitrogen content increases NO emissions.
• Total N0x reduction is 270 ppm to 280 ppm from the levels emitted by
a conventional burner.
EPDC and its joint researchers will Investigate more effective combustic
conditions and further advanced low N0x burners to achieve the target 100 ppn
N0X emission level.
283
-------
P.O.M. MEASUREMENT RESULTS
Table VIII shows P. O.M. measurement results at one particular test
furnace.
This measurement was performed using the EPA No. 5 modified sampling
method developed by Battelle and a liquid chromatography analyzer.
These measurements indicate the following:
(1) Total emissions of P.O.M. with two stage combustion (reported in the
United States as an increase of 35%) increased 32%.
(2) The increase of P.O.M. emission owing to combustion modification is 20
to 60%.
(3) Absolute P. O.M* values increase at partial load.
(4) Measured P.O.M. emissions were much less than the values reported in
the U.S.A. It is not clear whether the observed difference is due to
the measuring method (GC-MS method in the U.S.) or the type of
coal. This difference shall be studied in the future.
Another observation is that as N0X emissions decrease with combustion
modification, P.O.M. emissions increase. Therefore, some adjustment is needed
to alleviate this problem.
This correlation is shown in Figure 5, and according to the data taken by
Hitachi, P.O.M. emissions are around 10 ug/Nm when N0X emissions are lowered
to 100 ppm. This is about five times that of normal combustion.
284
-------
Section 6
COMBUSTION MODIFICATION EXPERIENCE
AT ISOGO THERMAL POWER STATION
The Isogo power station is located in city of Yokohama about 30 km south
of Tokyo in an area known as the Keihin Industrial Area. The City of Yokohama
is very anxious to reduce pollution, and thus the pollution prevention
agreement between Yokohama city and EPDC is uniquely severe.
In 1977, Yokohama city proposed a guiding principle for restricting N0X
emissions, resulting in a N0x emissions reduction of about 36%. Accordingly,
the proposed N0x emission level from a coal-fired plant in Yokohama is
159 ppm.
HISTORY
Isogo No. 1 and No. 2 boilers (capacity of 265 MW each), manufactured by
Ishikawajima-Harima Heavy Industries Co., Ltd. (IHI), were completed in 1967
and 1969, respectively. A summary description and general boiler arrangement
are shown in Table IX. The original burner arrangement is shown in Figure 6.
In advance of the first restrictive regulations on NO emissions from
X
stationary sources in 1973, Isogo Units 1 and 2 were equipped with two stage
combustion with an over air port system (OAP) for N0x control.
As a second step and in accordance with successive, more stringent regu-
lations, IHI-FW dual flow pulverized coal burners (DF-CN burners, Figure 7)
and a boundary air system (Figure 8) were applied to the No. 1 boiler in 1976
and to the No. 2 boiler in 1977.
As a third step, burners out of service (BOOS) and steam injection from
oil guns were tested on the No. 2 boiler in 1978.
A fourth step to decrease N0x emissions was based on the test results of
the third step and the results on IHI's, test facility (capacity of 2 ton/h
pulverized coal firing). It was decided that the installation of eight
285
-------
additional OAPs close to both side walls and a division wall (Figure 9) and
installation of the steam injection nozzles on each oil burner gun would be
carried out for No. 2 boiler. The target completion date was Spring 1979,
aiming at NC>x reductions less than 160 ppm (at 6% 02) under routine operation
conditions. The same modifications were performed successfully on the No. 1
boiler in 1980. A summary of these countermeaeures and the results are shown
in Table X.
PRIMARY COMBUSTION CONTROL TEST
Figure 10 shows the relationship between TSC air ratio and NO level, and
the predicted OAP damper open ratio for N0x emissions less than 159 ppm. At
100% damper open ratio, TSC air is about 15% of total air volume.
Adapting these primary test results and results of IHI's test furnace to
the Isogo power station, we predicted the N0x level to be 150 ppm and below
under the condition of 30% TSC air volume to total air volume.
RESULTS OF N0x REDUCTION AFTER IMPROVEMENT OF COMBUSTION MODIFICATION
After installing OAP as illustrated in Figure 9, we had 190 ppm of NO as
X
the upper level in normal operation and 150 ppm as the minimum level. At that
time, boilers operating under routine conditions performed almost
satisfactorily without slagging and fouling trouble.
However, when an N0x emission level of 150 ppm is reached, superheater
tube metal temperature increases beyond the recommended operating limit, and
to protect the tubes, we have to restrict the two stage air volume (OAP air
volume).
Figure 11 shows actual data for N0x and indicates for increasing or
decreasing load, NO levels seldom exceed 200 ppm. The predicted NO reduction
A X
was almost obtained. However, the increase in superheater tube temperature
exceeded our expectations. Therefore, the achieved NOx levels are normally
less than 190 ppm.
UNBURNED MATERIAL
Other potentially serious problems accompanied N0x reduction. These
included an increase in CO, unburned solids in the fly ash, and increased
286
-------
emissions of other hazardous organic materials such as P.O.M. In fact,
these problems were recognized during the early stages of the field test funs.
To reduce the unburned material mentioned above, we have tried to improve
the mixing conditions between the secondary staged air and the combustibles
produced in the primary zone through the installation of additional OAPs.
By applying side OAPs, the remarkable N0x reduction could be achieved
while keeping unburned material at a constant level.
FUEL ANALYSIS
The fuel coal burned at Isogo Power Station is a blended variety, con-
taining Taiheiyo coal as the main constituent. The typical analysis is shown
in Table XI.
IMPROVEMENT OF SH TUBE TEMPERATURE AND SPRAY QUANTITY
To prevent the SH tubes from increasing in temperature, we modified the
superheater layout, the final configuration of which is shown in Figure 12.
In addition, the spray attemperation quantity was increased from 66 to
100 ton/hr. Currently, the spray quantity being employed is the same as that
used before modifications for N0X reduction. Unfortunately, the trend of high
superheater temperatures is not remarkably improved.
287
-------
Section 7 1
COMMERCIALIZED LOW NO BURNER
FOR NEW COAL-FIRED POwEr PLANTS
EPDC is constructing new coal-fired power plants such as the Matsushima
power station and Takehara Unit No. 1. Each plant has incorporated N0x
combustion control technology developed with the aid of results from test
furnaces.
MATSUSHIMA POWER STATION
Matsushima power station No. 1 unit is now performing trial operations
until January 1981, and shortly thereafter, it will start commercial
operation.
A description of the Matsushima power station is following:
• Output 500 MW x 2 unit
• Fuel Coal (imported)
• Boiler Manufacturer MHI
• Type of Boiler Supercritical pressure reheat type
U.P. boiler
• NO (guaranteed) 285 pm (0„ "6%)
X (N - 1.7%)
• NO (target) same as guaranteed
• Burner type SGR burner (illustrated in Figure 13)
NO^ emission levels from the Matsushima power station are about 200 ppm
(O2 =6%, N ¦ 1.7%), measured after trial operation without burner adjustment
optimization.
288
-------
COMPARISON OF NOx EMISSIONS BETWEEN LABORATORY TEST RESULTS AND FIELD DATA
Figure 14 shows test furnace results and field data, the latter from EPDC's
Takasago and Sunagawa of Hokkaido Electric Power Co.
We find good correlation between actual field data and test data using domes-
tic coal, and therefore anticipate reducing NO emission levels to approximately
X
250 ppm (O2 * 6%, N ¦ 1.7%) as suggested by Figure 14.
TAKEHARA POWER STATION NO. 3
Construction has recently initiated on Takehara Unit No. 3 and commercial
operation is scheduled for March 1983. An outline of Takehara No. 3 is as
follows.
Output 700 MW x 1 unit
Fuel Coal (imported coal)
Boiler Manufacturer BHK.
SCR Manufacturer BHK
Type of Boiler B & W Supercritical pressure reheat type
UP boiler
NO (guaranteed 250 ppm (Oj ¦ 6%, N ¦ 1.8%)
at boiler)
NO (target 200 ppm (0« ¦ 6%, N - 1.8%)
at boiler)
N0X (guaranteed 60 ppm and below (0£ ¦ 6%, N - 1.8%)
at stack)
• Burner type PG dual burner
The PG dual burner is shown in Figure 15.
COMPARISON OF N0x EMISSION BETWEEN LABORATORY TEST RESULTS
VS. FIELD DATA
Figure 16 shows the correlation between actual field data and test furnace
data with heat release rate at the burner zone. This relation will be con-
firmed when Takehara No. 3 boiler is operating. EPDC anticipates achieving
the target NOx level.
289
-------
REFERENCES
Y. Nakabayashi, Status of R&D on NO removal in Japan and results of EPDC's
R&D for DeNO Process. Second EPRIXNOx Control Technology Seminar, Denver
Colorado, Nov. 1980.
Environmental White Paper, 1980, Japanese Environmental Protection Agencies,
290
-------
Appendix A
Summary of Emission Control
Research and Development in Japan
for Coal-Fired Boilers
291
-------
Test Subjects
Companies and
Joint Resear-
chers
Capacity
Fiscal
L975
4 7 Lfl i
1976
1977
1978
4 7 10 1
1980
1981
1. Desulfuri-
zation
technology
r>o
ro
2. Demtration
technology
(1) Soot separat-
ing desulfu-
rization
system (dust
collecting)
a) Hitachi
b) Mitsubishi
c) IH1
(2) Dry flue
gas desulfu-
rization
system
(Low Nox burner
development
test)
(1) low NOx bur-
ner test
(2) Ultra-low
Ncx boiler
test
Takehara
Thermal Power
Station
Isogo Thermal
Power
Station
Takehara
Thermal
Power
Station
Takehara
Thermal
Power
Station
Isogo
Thermal Power
Station
Babcock Hitachi
Kure works
mi Hagasaki
Laboratory
IHI tec i
Works
Hitachi
Mitsubishi
IHI
SHI
IHI
Hitachi
Hitachi
Mitsubishi
X H I
Approx.
2#OOCNm3/h
10,000Nn3/h
Actual
boiler
Actual
boiler
Test
furnace
Test plant
Put into practical use
(Dec. '75>
timber of machines
equipped(#1 Mar. *76)
(#2 May *77)
SU!21ARY OF EMISSION CONTROL
RESEARCH PROJECTS FOR COAL FIRED BOILERS
-------
Test subjects
Location
Companies and
Joint Resear-
chers
Capacity
Fiscal
1975
1976
1977
1978
1979
1980
1981
Remarks
4 7 10 1
4 7 1C 1
4 7 10 1
4 7 L0 J
4 7 10 1
4 7 L0 1
4 7 1C 1
(3) 2-stage com-
bustion
device
(Research and
development of
denitration
technology)
<1> SCR test
a) Hitachi
Shipbuilding
& Engineering
Co.
-------
^-1-
.Location
Joint Resear-
chers
Capacity
£
1976
4 7 10 1
1977
4 7 1(1
4 7 1C 1
4 7 L0 1
1980
4 7L0 1
1981
4 7 L0 1
Remark*
(2) Denitration
a) Air preheater
test
FGD Equipment Cor
practical use
(1) Isogo Thermal
Power Station
(#1)
(#2)
(2) Takasago
Thermal Power
Station
(ID
(#2)
(3) Takehara
Thermal Power
Station
(il)
(4) Matsushita
Thermal Power
Station
(iX)
(«2)
Takasaqo
Thermal Power
Station
Takehara
Thermal Power
Station
Isogo Thermal
Power Station
Gadelius
Z H i
10,OOONe3/h
5r000N»3/h
2,OOONm3/h
Mitsui Mi ike
265MWx2Units
250MWx2Units
I H I
Hitachi
250MWxlUnit
SOOMMxlUnit
SOOMHxlUnit
II Unit (scheduled)
start up in MR/16,
#2 Unit (scheduled) to
start up in HAY/76*
#1 Unit (scheduled) to
start up in FEB/75.
#2 Unit (scheduled) to
start up in MAH/76.
Scheduled to
start up in Feb,/77
II Unit (scheduled) to
start up in Dec./81.
(3/4 Capacity)
#2 Unit (scheduled) to
start up in Jul.'81.
SUMMARY OF EMISSION CONTROL RESEARCH PROJECTS FOR COAL FIRED BOILERS
(Continued)
-------
Test Subjects
Location
Companies and
Joint Resear~
chers
Capacity
Fiscal
1975
1976
1977
1978
1979
1980
1981
Remarks
4 7J.0 1
|
4 7 LO 1
4 7 10 1
4 7 LO 1
4 7 LC 1
4 7 10 1
4 7 D 1
3. Dust Collect-
ing
Technology
4. Most* Water
Treatment
Technology
3. Attachment
techniques
CI) H. ESP
(2) Ditto, but
test plant
(3) Bag filter
(1) Reverse
osmosis
process
denitrifi-
cation test
(2) Biological
treatment
denitrifi-
ed t ion test
(1) Desulfuri-
zation
a) GGH test
b) De SOx fan
test
corrosion-
resisting
material
. test
Isogo Thermal
Power Station
Takasago
Thereal Power
Station
Takehara
Thermal Power
Station
Isogo Thermal
Power Station
Takasago
Thermal Power
Station
Matsushiaa
Thermal Power
Station
Not decided
yet.
isogo Thermal
power Station
Manufacturer•s
laboratory
Takasago
Thermal Power
Station
Takehara
Thermal Power
Station
SHI
Gadalius
Hitachi
I H 1
NHI
SHI
NHI
Mot decided
yet.
SHI
Mitsui Hiike
Ebarar
HZ
Gadelius
Hitachi
M H I
Fuji Resin
10,OOONn3/h
5,OOONm3/h
2,300Ne3/h
2,000NnVh
900Nn3/h
500MHxlUnit
500MHxlUnit
Not decided
yet.
Max. 7»3/day
Laboratory
scale
10,000Km3/h
^10,000Nn3/h
Test Piece
#1 Unit (scheduled) to
start up in DEC/60.
#2 Unit (scheduled) to
start up in JUL/81.
SUMMARY OF EMISSION
CONTROL
RESEARCH PROJECTS FOR COAL FIRED BOILERS
-------
PO
VO
o*t
Teat subjects
Location
Companies
Joint Resear-
chers
i
Capacity
Fiscal
1975
1976
1977
1978
1979
L980
..12
Remarks
4 7 LO 1
4 7 10 1
4 7 1C 1
4 710 1
4 7 LO 1
4
7|lC 1
4 7
10 1
Denitratlon
Technology
Demostration
Test
Takehara
Thermal Power
Station
800,000Nai3/h
•
Government
SUMMARY OF EMISSION CONTROL RESEARCH PROJECTS FOR COAL FIRED BOILERS
(Concluded)
-------
FIGURE I
STATUS OF COMPLIANCE WITH ENVIRONMENTAL QUALITY STANDARDS
RELATING TO AIR POLLUTION
I. Sulfur dioxide
1.500
(80.1 V
(stations)
(1975)
ri,
fT
87
353n
185;
.6%
Effective no. of
monitoring stations
No. of stations in
compliance with environ-
mental quality standards
i—l, 415—,
193.0?o
(1976)
H'|i
ill:
(1977)
2. Nitrogen dioxide (1977 status of compliance with new standard)
No. of stations in excess of 0.06 ppm
No. of stations in zone from 0.04
to 0.06 ppm (inclusive)
I I No. of stations below 0.04 ppm
900
-
225
(25.2%)
600
626
(70.2%)
300
(stations)
41 (4.6%)
166]
(36.3%)
•86
(47. 2% )•
30(16.5%)
200
100
Stations monitoring Stations monitoring
ordinary ambient automobile emissions
air quality (892) pas (182)
Notes; I. Classified according to the annual 98 percentile value of
daily average NO, concenlialion, with a Saltzman coef-
ficient of 0.84.
2. Monitoring stations for automobile emissions situated
over roadways have been excluded.
297
-------
FIGURE 2
CHANGES IN THE ANNUAL DENSITY OF MAJOR AIR POLLUTANTS
C
*5
(ppm)
0.06
0.05
0.04
0.03
0.02
0.01
Notes:
Carbon monoxide
(av. of three stations
monitoring automobile
exhaust piis in I lie Tokyo
Metropolitan Area)
Carbon monoxide
lav. of 2 stations)
Suller dioxid
(av. of I S stations)
Nitrogen dioxide
(av. of 6 stations)
(av. of
IS stations)
I ppm)
16. 5
6.0
5.0
4.0
3.0
2.0
J1.0
X.
o
c
0
E
c
o
x>
( ppm '
10.04
0.03
0.02
•65 '66 '67 '68 *69 '70 '71 '72 '73 '74 '75 '76 '77
1. Carbon monoxide measurements by automobile exhaust
pas monitoring stations arc given in terms of calendar
years.
2. Saltiman coefficient of 0.72 is used for nitrogen oxides.
298
-------
Eigute 3
Model Anti-pollution Coal-fired Thermal Power Plant
"Con-
centracion
Case
Syseeta
Power Plane Flow
Remarks
nh3
Low Dust
Denltracion
System
n
—o-
Boiler H.ESP SCR
Water L
-o-1-
Wcc
FCD
T7CI1
wit
H.ESP-.Hoc-side
Electros tatic
Precipitator
SCR: Selective
Catalyst
Reduction
Equipment
High Dust
Denltration
System
Boiler
SCR
Tm (~n
MI3 Ash
Treatment
Wa t c r.—
—o1-
Wet
FCD
r.CH
| uvr |
KH3
Dry FCD
System
Boiler
o-
a7T;
b':'./"au
E!'
hO-
Dry FCn
Kl!
A/H: Air Preheater
CCH: Cas-Cas
Heat Exchanger
FCD: Flue Gas
Dcsulfuriza-
L ion
W.WT: Waste Water
Treatment
Equipment
BH: Bag Ilous*
0: Kan
-------
FIGURE 4
One Sample of Test Results
A company
company
X E c ooI
IN. 1,7%)
(N « 1.7%)
t-
TSC + G R
Conventional
Burner
w
4>
e
9
m
*
tl
z
TSC
a.
o
•f
u
i/i
»-
o ~
c c
S 5
C CD
ft
>
8
u
4>
C
3
CD
*
4>
Z
TSC
ce
o
+
o
v>
©
©
©
©
©
®
©
©
©
©
300
-------
FIGURE 5
Corelotion of NOx conccntrol ion vs P.O.M. Emission
8.0 r
7.0
6.0
5.0
4.0
3.0
Z.O
Tf*t Fume
100
200
300
400
At Combustion
Modificotion
NOx (ppmj
At Normol
Combustion
NOx Conctntrot ion of Ttst Furnocc it the ovtrofi volu* of
rroiimum ond minimum volues.
301
-------
FIGURE 6
IHI 'S ORIGINAL BURNER
Secondary
Outer Sleeve
Inner Sleeve
Tertiary Air
Oil Burner
Secondary
Air Vane
Furnace Wall Tube
302
-------
FIGURE 7
IHI'S DF-CN BURNER
Fuel Lean
Combustion
Strendhened External
Recirculating Flow
Flame
Outer Sleev
Inner Sleeve
Movable
Sleeve
Oil Burner f
Inner vone
ffi
r
Secondary Air vane
Interna
Recirculating Flow
Wind Box
Fuel Rich
Combustion
Furnace Wall Tube
Cooli
ng Air Duct /
[Taper Noztle
303
-------
FIGURE. 8
I HI'S BOUNDARY AIR SYSTEM
Cooling Air NozzU>
\pver Air Port
Burner Throat
DF Burner
Burner Throot Cooling Device
(Figure 9 omitted by author)
304
-------
Fig. 1 0 OAP Damper Open Ratio Versus NOx Emission
p.p.m.
300
250
200
150
100
50
~ Predicted level with additional
OAP's and steam injection equipmer
O Test results
O Boos of two burner cut results
Actual
data
Steam
injection
Predicted data
50
OAP Damper Open Ratio
100%
305
-------
Load change (140*^ 265 )
Figure 1 1 Actual Data of
NOx
Load change (265^5^-140^')
Date Feb. 18
±:|!|
/yv^
iaa=6»
Time (O'clock)
-------
FIGURE 12
SH SYSTEM AND IMPROVEMENT
Steom Drum
-wyw-f
—V\M-
-Ww,
Primory
Rodiont
(Roof )
HRA Conyecti
Convection
cut tubes
Prymong
Sproy
) Secondary
Rodiont
Div Wall
of Furnoce'
Secondary
Sproy
Platen
)
Primory
Pendant
Secondary
Pendant
To Turbine
307
-------
AUXILIARY AIR
SGR
EL AIR
COAL
[LAIR
SGR
AUXILIARY AIR
FIGURE 13
SGR Burner
308
-------
FIGURE 1*4
Comparison of NOx Emission
- Laboratory Test Results vs. Field Data
600
250 ppm
0 = 6%
Kind of
Coal
u> is
Boiler
Actual
Boiler
Test
Boiler
CE Conventional Burner
u.
Hokkaido E.P. Co. Sunaaawa 13
EPDC Takasago II, 2
. SGR !,
Burner
EPDC
Matsushima
II. 2
- PM
Burner
309
-------
Figure 15 BHK Low NOx Burner
Secondory air resisler
Thertiory oir resisfer
Ventry
O
Oil burner
Primory gos vone
Secondary air
Thertiory oir
Inlet of primory gos
Secondory oir vone
Primary air and pulverlized coal
-------
FIGURE 16
HEAT RELEASE RATE AT BURNER ZONE VS NOx LEVEL
CB. Circular air register
Burner
DRB: Dual air register
Burner
PC-DRB: PG type-DRB
TSC' Two stage combustion
CM.' Gas mixing to
secondary air
TH 1 CB I
IPG-DRB
PC-DRB+TSCl
[PC-DRB+TSC+CMl
ITH 1 DRB
TH 1 DRB+
50 60 70 80 90 100 110" 120 130 140
Heat release race at burner zone (7.)
311
-------
TABLE I
no2 ENVIRONMENTAL STANDARDS IN JAPAN
NO2 Environmental Standards
announced as of May 8, 1973
N02 Environmental Standards
revised as of July 11, 1976
Remarks
NO2 high
polIution
areas
Intermediate tarqet level:
(effective through May, 1978)
NO2 level in the atmosphere
should be less than 0.0*» PPM
on the hourly level.
Final tarqet value:
(effective through May, 1981
NO2 level in the atmosphere
should be less than 0.02 PPM
on the hourly level.
Final target level :
(effective within seven years
of the date of revision)
NO2 level in the atmosphere
should be. reduced to a range
of 0.Oil PPM to 0.06 PPM or
less on the hourly average.
(1) The Japanese NO2
Environmental Standards,
which are much more rigid
than those now in force
in many foreign countries,
have been disputed in
Japan.
(2) Nine per cent of the
target level based on the
original standards was
attained as of fiscal
1976.
NOj Iow
pollution
areas
Final tarqet value:
(effective through May, 1978)
NO2 level in the atmosphere
should be less than 0.02 PPM
on the hourly level.
-------
TABLE II
CHANGES OF NOx EMISSION STANDARD
(Stondord Vfolve ppm )
( 1st Stondord Volue
( 2nd Stondord Volue :
I 3rd Stondord Volue 1
Aug. 10, 1973 s tori 1
Dec. 10, 1975 start \
Jun. 18. 1977start I
CO
CO
Scole
of \
Facilities \
Installation
Time
Before Aug. 9, 19 73
Aug. 10. I973~0ec. 9. 1975
Dec. 10. 1975-
Jun. 17, 1977
After
Jua 18.1977
Notes
Clorif icol ior
of
Restrict ion
1 St
Exist ing
2nd
Existing
3 rd
Existing
1 St
New
2 nd
New
3 rd
Existing
2 nd
New
3 rd
Existing
3 rd
New
Applicable
Time
From
Jul 1,1975
From
Dec.1,1977
From
May. 1.1980
From
Aug .10.1973
From
Dec 1,1977
From
May, 1,1980
From
Oct 10,1977
From
May 1,1980
From
Jun. (8.1977
Gas Firing
10, : 5V.I
mil lion
0.5 NmVh over
million
0.1<~0 5
I70
170
130
130
13 O
130
130
130
(30
130
130
130
(00
100
100
100
60
100
Solid Firing
(Oz 6%)
million
0.1 over
600
(7501
600
(750)
480'
480
480
460
480
480
400
( ) indicates combut -
lion with low
calorific value
cool(5.000KeolAg
under)
Ceiling burner
650 ppm
Devided wall type
550ppm
Fluid Firing
I02 : 4%)
million
1 over
million
0.5- 1
mil 1 ion
01 -0.5
230
1280)
230
(280)
230
(280)
230
(280)
230
1280)
230
(280)
180
180
1210)
190
C2I0)
180
180
130
180
180
180
180
180
180
150
150
150
150
150
150
130
130
150
1 ) indicoles combus-
tion with crude
oil and lar
( I indicates plant
equipped with
DeSOx system
-------
Table H
Present State of NOx Emission Control
for Existing Coal Fired Boiler
Isogo RS
Tokosogo PS
Takehora PS *i unit
Lo w
4 0 0
4 0 0
4 0 0
Ordfnonces
agreement
1 5 t
(under negotiatun)
30 0
4 1 0
Refferences
Output
Fuel
N content
265MWx 2
coal
1.1
250MWx 2
coot
l.O
250MWx 1
cool
l.O
Table W
Sample of NOx Emission Control
for New Coal Fired Boilers
Motsushima P. S
To ke hora P. S
Lo w
(Oj* 6*A)
4 00 ppm
400 ppm
i
Ordinances
agreement
t02« 67.)
30 0 ppm
No. 1
No.2
No.3
198 0
—
—
—
198 2
1 1 2
1 7 5
60
198 3
79
1 7 5
6 0
1 98 4
72
1 1 5
60
1 98 7
60
8 1
5 4
Remarks
250MW
350MW
700 MW
cool
oi 1
coal
314
-------
TABLE V
NOx REDUCTION PROGRAM
IN TAKEHARA AREA
(V.)
100
80
60
40
20
100
Tokehoro
No. 2
350MW
(Oil )
To kehara
No. I
2 50 MW
(Coal)
\
79
Takehara
No. 3
70 0MW
(Cool)
Takehara
No. 2
350 MW
(Oil )
Takehara
No. i
250MW
(Coal)
79
AP/56
[' | EPDCTokehara P/S
Total Emission Volume
79
73
Takehara
No. 3
700 MW
(Coal)
Takehara
No. 2
350MW
(Oil )
Takehora
Sfo1
Present
1
MW
Afllt
No. 3 Tokehoro
commercial
operotion(l982)
BP/58
AP/56
65
Takehara
No. 3
700MW
(CoaI)
Takehara
No. 2
350MW
(Oil)
Tokthoro
No. I
250 MW
fro"' )
A P/S
commercial
operation
(1983)
B P/S
commercial
operation
(1984)
315
-------
N i trogen
TABLE V!
Content of Planning
Coa 1
Chi na
A
B
N U)
1.4
0.8
Aus tra1i a
A
B C D E
F G H 1
N (*)
1.6
1.7 1.6 1.5 1.6
1.6 1.5 1.6 1.7
The others
A
B C-l C-2 C-3
D
N (%)
1.8
1.7 1.6 ].h l.i»
1.1»2
Domes tics
A
B
N
0.8
1.0
316
-------
TABLE VII
Outline of R D Programs on Combustion NOx Control Techroloqy
Test
Location
Test furnoce
Copocity
Test conditions
Soheduled
1977
1978
1979
1 980
198 1
M H 1
Nogosoki
1.3 k//H
(oil base)
TSC . GRF. GM.
SGR k PM burner
B H K
K ur e
2.0 t/H
(coal bose)
TSC, GRF. GM,
PG burner
I H I
A i 0 i
1.0 t/H
(coal base )
TSC . GRF.
OF (E) burner
K H 1
Shiga
3.7 t/H
(coal bose)
TSC. GRF , G M
V-D burner
r-r>-==
TSC ! Two Stage Combustion
G M 60s Mixing
-------
Table VIE
Measurement Results of P. 0. M. at Furnaces
Uni t ug / Nm*
Com bust ion
P. 0. M
Component
No rma 1
Combus-
t ion
2 stage
Combus-
t ion
2 stage +
F6 mix +
Primary 60s
PGM
Portia 1
Load
Naphthalene
0.22
0. 1 8
0. 22
0.24
0.24
Fluorant hene
-
0.40
0.88
0. 78
0.84
Pyrene
0. 65
0.64
0.87
0.87
0.84
Benzo (o)
pyrene
0.48
0.59
0.66
0.58
0.62
Totol P. 0. M.
1.99
2.62
2.46
3.15
3.49
1 Percentage when
Normal combus-
tion is 100 )
(100 )
(132 )
(124)
(158)
(175)
318
-------
TABLE IX. SUMMARY OF ISOGO 265 MW PULVERIZED
COAL FIRED BOILER (NO. 1 and NO. 2)
Type
Evaporation (at M.C.R.)
Steam pressure (at M.C.R.)
Superheater outlet
Reheater outlet
Steam temperature (at M.C.R.)
Superheater outlet
Reheater outlet
Fuel
Number of burner
Draft System
319
IHI-FW single drum, radiant type,
natual circulation, reheat boiler
(Indoor service)
840,000 kg/h
2
176 kg/cm g
2
34 kg/cm g
571°C
571°C
Bituminous coal
(Equipped with 50% MCR heavy oil
firing system)
24 set
(4 rows and 3 stages on boiler
front and rear walls)
Balance draft system
-------
v Countermeasure for NOx Control
TABLE X for isogo Power Station NO.2
Unit
Month. Year
NOx Emission Level
by Regulation, etc.
(ppm)
Achieved NOx
Level
(ppm)
Countermeasure
Before May,
1973
600
570
None
June, 1973
600
380 * 510
Two Stage
Combustion
Apr., 1977
480
240 -v. 250
Low NOX Burner
Curtain Air
Apr., 1979
400
170 -v. 190
Strength of
Two Stage Com-
bustion, etc.
Apr., 1980
159*
Yokohama city's
Requirement
170 -v 190
320
-------
TABLE XI
TYPICAL ANALYSIS ISOGO'S COAL
Higher heating value 6,200 kcal/kg
Ash 16.5%
Volatile matter 39-6%
Fixed carbon 39-
Sulfur O.A%
Hydrogen 5-5%
Nitrogen 1.0%
321
-------
EMPIRICAL EVALUATION OF SELECTIVE CATALYTIC REDUCTION
AS AN NOx CONTROL TECHNIQUE
By:
J. Edward Cichanowicz and D. V. Giovanni
Air Quality Control Program
Coal Combustion Systems Division
Electric Power Research Institute
Palo Alto, California 94303
322
-------
ABSTRACT
Selective catalytic reduction (SCR) has been proposed as a technique for
control of NO emissions to levels significantly below those mandated by NSPS
for coal-firea utility steam generators. EPRI is conducting an empirical
assessment of the feasibility and cost-effectiveness of SCR, using a pilot
scale system at the EPRI Arapahoe Emission Test Facility to simulate authentic
coal-fired utility operating conditions* The program is a logical extension
to earlier EPRI work defining economic feasibility of postcombustion control,
and complementary to other pilot scale studies in the United States and Japan.
The test program was initiated in September 1980 on a facility capable of
treating 5,000 scfm of coal-fired flue gas, equivalent to 2.5 MW of electrical
generating capacity. The facility employs a regenerative air heater in series
with a catalytic reactor to assess potential impacts on air heater perfor-
mance. The tests will focus on four major issues important to the evaluation
of SCR technology: (1) process performance as defined by N0X removal
capabilities at conditions representative of authentic utility application;
(2) process operating demands including the need for monitoring and control
systems, consumables such as ammonia, energy (pressure drop and auxiliary
power), operating and maintenance requirements to maintain process perfor-
mance, catalyst lifetime; (3) environmental impacts due to emissions of
residual ammonia, SO-j, and sulfates and bisulfates of ammonia; and the poten-
tial effects on SO2 and particulate control; and (4) systemwide operating
effects such as increased operating and maintenance of downstream surfaces
(particularly the air heater), heat rate penalty, limitations in load-
following, etc.
Results are presented for the initial tasks dealing with the evaluation
of measurement techniques, and preliminary data describing reactor and air
heater performance.
323
-------
ACKNOWLEDGEMENTS
The authors wish to acknowledge the efforts of the many individuals
volved at all levels in this project. Dave Naulty and Mike Moora of
earns-Roger, Inc., were primary individuals responsible for the procurement
d construction of the pilot facility. Assistance from Kawasaki Heavy
dustries was provided by Senji Niwa. Operation and maintenance of the pilot
ant was managed by Jim Parsons and John Serdinsky of Kaiser Engineers,
sting was conducted by Gary Shiomoto, Larry Muzio, and Robert Pease of
B. Finally, the key role played by EPRI staff members at the Arapahoe
ission Test Facility, namely Richard Hooper and Lou Rettenmaier, is most
atefully appreciated.
324
-------
EMPIRICAL EVALUATION OF SELECTIVE CATALYTIC REDUCTION
AS AN NOx CONTROL TECHNIQUE
INTRODUCTION
The Environmental Protection Agency (EPA) has cited as goals the reduc-
tion of New Source Performance Standards (NSPS) for oxides of nitrogen (N0x)
emissions from coal-fired utility power plants from current levels of 0.5-0.6
lbs/million Btu1s (depending on coal properties) to 0.2. Under certain cir-
cumstances, coal—fired power plants could be required to further reduce NO^
emissions below NSPS levels, such as in Air Quality Regions (AQRs) in noncom-
pliance with the proposed Federal ambient NO2 standard, or to prevent signifi-
cant deterioration of Class I AQRs. Such regulations have already been
promulgated for certain classes of oil- and gas-fired steam generators in
California's South Coast Air Basin. Postcombustion methods of N0X control
(alternatively known as flue gas treatment) have been proposed to achieve such
stringent standards for both clean fuels (such as natural gas and fuel oil) as
well as coal. Accordingly, the electric utility industry through EPRI is con-
ducting an empirical evaluation of the feasibility of postcombustion N0X
control processes for coal-fired utility boilers.
Development efforts in postcombustion control have been conducted predom-
inantly in Japan and in the last decade. This is due to stringent Japanese
emissions standards; and to the predominant use in that country of natural gas
and fuel oil, which due to low ash and sulfur content, offer greater pros-
pects for successful application. Extensive laboratory and pilot scale work
has been conducted for these "clean" fuels, and recently has been expanded to
consider coal. Most of this development work is typified by well-controlled
operating conditions necessary for fundamental process development, but not
representative of authentic utility application. Extrapolation of Japanese
pilot plant experience with coal to U.S. utility systems is complicated not
only by significant scale factors but also by differences in coal properties
325
-------
(i.e., ash characteristics and chemistry) between coals used in Japan and the
United States. Experience with full-scale utility systems is limited
exclusively to clean fuels, and thus does not reflect the problems inherent to
high particulate loading, SO2 concentration, and ash chemistry.
SUMMARY OF RELEVANT EPRI EXPERIENCE
Prior work conducted by EPRI has consisted of preliminary technical and
economic assessments of the feasibility of postcombustion processes for U.S.
utility coal-fired application. In two initial studies (1,2), approximately
50 processes were categorized as to technical basis and screened according to
potential merits and economics of operation. Follow-on studies considered a
limited number of processes in more detail through design studies of specific
applications. Seven postcombustion processes were evaluated by the Tennessee
Valley Authority (TVA) through a jointly funded program with the EPA; the
results identified the dry, ammonia-based selective catalytic reduction (SCR)
process as a leading candidate for near-term commercial acceptance (3,4). A
further study conducted by Stearns-Roger for EPRI examined two dry SCR pro-
cesses In addition to the Shell/UOP simultaneous S02/N0x and the Exxon Thermal
DeN0x process (5). Both the TVA and Stearns-Roger studies identified for each
process (a) the technical merits and potential obstacles to application to new
coal-fired utility boilers, and (b) the economics of operation, necessarily
limited to the costs of process equipment procurement, installation, and
operation. Only the most obvious impacts of SCR on powerplant design and
operation that could be identified with the limited level of experience were
included. A study similar in scope and depth has been recently completed,
also by Stearns-Roger, to assess costs of retrofit application of SCR systems
from two process vendors to an 80 MW boiler (6).
One of these preliminary investigations (5) has identified the minimum
cost of SCR application to new 500 MW coal-fired power plants, for the case of
90% NO removal from initial NO concentrations representative of present NSPS
X X
levels. These results, summarized in Table 1 for the Kawasaki Heavy Indus-
tries (KHI) process, are preliminary and therefore incomplete as all potential
impacts on plant operation attributable to SCR have not been identified. This
can only be accomplished through a pilot-scale evaluation of a commercially
representative system on a coal-fired utility boiler at operating conditions
326
-------
that reflect authentic utility application. The remainder of this paper
provides a status report of such an effort at the EPRI Arapahoe Emission Test
Facility.
POTENTIAL SCR EFFECTS ON FLUE GAS FROM COAL
Figure 1 presents a diagrammatic path for the potential fates of the
major nitrogen and sulfur species in flue gas subjected to SCR treatment. The
temperature range of interest spans from typical catalyst temperatures (speci-
fic for one process, but representative of other SCR processes) to typical air
heater exit temperatures. The process inputs are indicated at the reactor
inlet as are the desired products of the NH2/N0x reduction reactions, and the
residual and by-product emissions of NH^ and SO^" As the flue gas proceeds
through the reactor and downstream equipment and cools, the following fates of
major nitrogen and sulfur species are possible:
(1) SO2 can be
(a) oxidized to SO3 within the reactor (usually 1-5% inlet SO2);
(b) collected in the SO2 scrubber and converted to the liquid or
solid scrubber effluent (ultimately to be transferred to the
scrubber waste disposal site);
(c) emitted
(2) SO3 can
(a) be collected in the scrubber and transferred to the liquid or
solid scrubber effluent, and ultimately to scrubber waste
disposal site;
(b) be emitted, either in gas phase or liquid if the dew point is
reached;
(c) condense or be adsorbed on internal surfaces; or on fly ash, and
thus collect in the particulate control device (and ultimately
the ash disposal site), or penetrate the control device.
327
-------
(3) NH-j can remain in gaseous form and be
(a) emitted;
(b) absorbed in the scrubber (disposal site);
(c) absorbed by fly ash and thus collect in the particulate control
device (ash disposal site), or penetrate the control device.
(4) NH-j and SO^ can combine and form (depending on relative NH^, SO^, and
H20 concentration, and temperature) ammonium sulfate or ammonium
bisulfate (liquid or solid phase) which can:
(a) pass through the air heater and collect in either the scrubber
or particulate control device (disposal site), or penetrate the
control device;
(b) impact and lodge on air heater or other internal surfaces and
subsequently be removed by sootblowing and endure the fate
described in (a), or be removed through water washing and trans-
ferred to the wash effluent;
(c) be absorbed by fly ash, and collect in the particulate control
device (ash disposal site), or penetrate the control device and
be emitted.
Based upon these potential paths, at least four ultimate fates for
nitrogen and sulfur residual and by-product species can be identified:
• airborne emission
• retained on surfaces within system (possibly removed through routine
washing maintenance procedures)
• SC>2 scrubber/effluent disposal site
328
-------
• particulate collection device/ash disposal site
Each of these affected areas has been the focus of considerable atten-
tion, and the added impact of SCR by-products could raise new issues requiring
additional corrective action.
Table 2 presents a compilation of the issues of concern attributable to
SCR that must be addressed before the feasibility of SCR can be assessed.
These issues are attributable to the operation and maintenance of both the SCR
process and the entire power plant, as determined by the four potential fates
of major nitrogen and sulfur species. These issues are categorized as to
(a) process performance, (b) process operating demands, (c) environmental
impacts, and (d) system-wide operating effects. For these issues of concern,
the data necessary from the test program for an empirical assessment is cited,
as well as the economic implications for power plant operation.
This table has been used as the basis for the test plan which is presen-
ted in a later section.
DESCRIPTION OF FACILITY
The empirical assessment of SCR for coal-fired utility boilers is being
conducted at the EPRI Arapahoe Emission Test Facility. This facility,
operated in cooperation with the Public Service Company of Colorado (PSCCo),
is located in Denver, and was commissioned in 1977 to support the testing and
evaluation of advanced particulate control technology. The direction of work
at Arapahoe has been expanded to include the present evaluation of N0X
control, and in the future (1981) will involve testing and evaluation of SO2
scrubbers, and address the integrated design of all components in an emission
control train.
In December of 1978, EPRI contracted with KHI to provide a 2.5 MW
(electrical equivalent) SCR pilot plant. The selection of KHI does not
represent an endorsement or preference by EPRI for that process, but rather an
acknowledgment that the process is representative of SCR systems currently
available. Similarly, the C-E Air Preheater Company was selected to supply
for the same flow capacity a Ljungstrom regenerative air heater which is
329
-------
considered representative of air heater designs in current and anticipated
usage.
Figure 2 presents an outline arrangement of the pilot scale reactor and
air heater. Coal-fired flue gas is extracted from PSCCo Unit No. 4 at the
economizer exit (approximately 650°f) and ducted to a control valve, venturi
for flow measurement, and an electrical resistance heater for control of inlet
gas temperature (not shown in figure). The ammonia vaporizer and injection
assembly is located at the base of the flue duct riser. The catalyst is
housed in three reactor modules arranged in series, each with an equivalent
quantity of catalyst. The regenerative air heater is located downstream, and
oriented for a horizontal axis of rotation.
Specifications for the KHI SCR process are presented in Table 3. The
major active catalytic ingredient is vanadium pentoxide, stabilized or suppor-
ted in a titanium dioxide base. The material is formed in cylindrical shapes,
and arranged as shown in Figure 3, with spacing selected to minimize particu-
late accumulation and plugging. Total catalyst volume is 120 ft^, correspond-
ing to a design space velocity of 2500 hr~* at the rated flow capacity of
5000 scfm.
The ammonia vaporizer and delivery system have been designed to simulate
the hardware and operation of the full-scale systems proposed by KHI for
commercial application. The ammonia is transported from a storage tank to the
vaporizer, and then mixed with carrier air (approximately 20/1 air/NH^ ratio)
preheated to 120°F and injected into the flue gas via a mixing grid.
Design specifications are presented in Table 4 for the regenerative air
heater, which is designed to simulate the heat transfer and sootblpwing char-
acteristics of full-scale utility operation. In this context, the temperature
history of the wheel elements and the mass flux per unit time of aootblowing
media have been selected to represent full-scale values. An evaluation of the
tolerance and susceptibility of three materials of wheel basket construction
to plugging/fouling/corrosion by fly ash and SCR residual and by-product emis-
sions is possible. Stainless steel, hard-rolled steel with baked enamel
finish, and Corten are all employed in several arrangements of wheel basket
construction.
The facility is equipped with process control systems to (a) facilitate
pilot plant operation and provide for well-controlled test conditions, and
330
-------
(b) evaluate typical control systems that could be necessary for commercial
acceptance of SCR. Control systems for the catalytic reactor are necessary to
adjust the mass NH^/NC^ ratio according to changing input reactor conditions.
Five control modes will be employed at Arapahoe to both facilitate pilot
plant operation and simulate the actual conditions of SCR operation in a full-
scale commercial application. Each of these modes is listed in Table 5, which
identifies for each the continuous on-line processing required, and EPRI
experience to date. The least on-line processing is required for the first
two modes that maintain either a specified NH^/t^ ratio or NOx removal effi-
ciency. The most processing is required for the latter three (which all
require continuous measurement of NH^ at the reactor outlet) to limit the mass
of NHj injection to an amount that maintains NH^ carryover below a prescribed
level.
Continuous measurement of NH^ in flue gas is being attempted with a
special-purpose sampling probe/catalytic converter. This apparatus selec-
tively oxidizes NH^ to NO, permitting NH^ concentration to be inferred from
the difference in NO measurement before and after NHg oxidation. A schematic
of the sampling probe and the analytical system is presented in Figure 4.
Results to date with this system have not been successful, thus prohibiting
any of the latter three control modes in Table 5 to be employed.
Two control systems are employed to drive air heater operation, and are
primarily used to provide well-controlled test conditions over the long time
periods required for assessment of air heater operation. Both controls vary
air side flow rate and static pressure to maintain a prescribed (a) flue gas
exit temperature, and (b) air side/gas side pressure differential (i.e.,
leakage) independent of flue gas inlet temperature, pressure, and flow rate.
TEST PLAN
A test plan has been developed to fulfill the project objectives in a
time- and cost-effective manner. Table 6 defines the tasks of the test plan
and identifies for each the primary and secondary variables, and those
parameters that remain constant. As indicated in Table 6 the tests will
evaluate the issues over the following operating conditions:
331
-------
• baseline conditions, or those at or near the design conditions that
could be encountered during commercial operation (Task 3);
• transient, or nonsteady conditions, simulating load swings, excess
air changes, changes in boiler operating conditions, etc. , (Task 4);
• upset conditions, characterized by high carbon monoxide (CO) and
unburned hydrocarbon (HC) concentration, such as experienced in rapid
shutdown or when encountering problems in fuel preparation (pulver-
izer or mill malfunctions) (Task 5);
• high particulate and SO2 conditions, providing an approximation of
the effects of ash loading and SO2 concentration (Tasks 5, 6),
The test plan will also provide information regarding:
• catalyst lifetime, by documenting changes In catalyst performance
over a 12 month period, employing well-controlled, repetitive perfor-
mance tests at comparable operating conditions (Task 7), and
• air heater performance, consisting of both continuous analysis of
daily performance data as well as repetitive tests at well-
controlled, comparable test conditions (Task 8).
Tasks have been scheduled in a parallel sequence so as to maximize the utili-
zation of the facility. The pilot facility operates on a 24 hour basis, with
most parametric testing conducted during daytime hours when boiler conditions
are generally steady-state and flue gas characteristics relatively constant.
During evening hours, boiler conditions are generally not steady providing
realistic transient conditions which can be used for evaluation of ammonia
injection control schemes.
RESULTS
The initial phases of this project which have been completed (Tasks 1 and
2) are concerned with the procurement of the sampling/analytical equipment and
development of appropriate test methodologies. Preliminary results at limited
332
-------
test conditions are available describing SCR reactor and air heater
performance.
Tasks 1 and 2 efforts served to (a) verify conventional sampling/analyt-
ical techniques for major stable combustion products, and (b) evaluate a
limited number of unconventional sampling/analytical techniques for the major
species of concern (NH^, SO^, and (NH^)HSO^ and (NH^^SO^). Once the specific
technique best suited for field testing was selected, the exact configuration
*
of sampling trains and the specific sampling/analytical methodologies were
determined. Detailed methodologies were identified for sampling gas phase NH^
and total NH^+ (impinger collection with analysis by specific ion electrode)
and gas phase SO^ (sampling by controlled condensation, analysis by titration
for S04= using lead perchlorate). These methodologies were determined for
application at the reactor inlet, reactor outlet, and air heater outlet.
A summary of the preliminary results is presented in Figure 5, which
indicates N0x removal, NH^ carryover, and SO2 to SOg oxidation (as percent of
inlet N0X, NH-j, and S02, respectively). Results are displayed from KHI tests
in Japan and preliminary Task 3 results from Arapahoe (data in parentheses).
Results from Arapahoe are presented for one temperature; the temperature
effect determined by KHI when available is indicated. Figure 5 indicates the
results from KHI and preliminary Arapahoe results are similar.
Simultaneously with Tasks 2 and 3, a preliminary evaluation of the influ-
ence of SCR residual and by-product emissions on air heater performance was
conducted. Initially, baseline performance was documented for over three
weeks, while flue gas passed through the reactor but without ammonia injec-
tion. Measurements confirmed air heater performance (thermal efficiency and
pressure drop) to be as predicted by C-E Air Preheater. Figure 6 presents the
inlet/outlet gas side pressure differential as a function of continuous
operating time for the baseline tests and after three weeks of operation with
NHj and SO-j in the flue gas. The significant increase in pressure drop
(approximately 3-6 inches HjO) was observed with sootblowing frequency main-
tained as recommended by the manufacturer, but with a reduced air pressure and
mass flux (due to limitations in the air delivery system, since rectified).
At the conclusion of the three week testing period with NHg injection, the
pilot plant was removed from service and the wheel baskets disassembled
(Figure 7), revealing significant accumulation of deposits at the cold-
333
-------
intermediate basket interface. Figure 8 shows a close-up of the deposit
buildup on an intermediate basket face constructed of mild steel coated baked
enamel. Qualitative analysis of a deposit sample indicated the presence of
ammonia suggesting ammonium sulfate/bisulfate constituents.
FUTURE PLANS
The test plan is currently being carried out at the Arapahoe Emission
Test Facility. It is anticipated that testing will be complete by Fall 1981.
334
-------
REFERENCES
(1) Faucett, H. L., Maxwell, J. D., and Barnett, T. A., "Technical Assessment
of NO Removal Processes for Utility Application," EPRI Report AF-568,
March 1978.
(2) Rosenberg, H. S. , Cur ran, L. M., Slack, A. V., Ando, J., and Oxley, J. D. ,
"Control of NO Emission by Stack Gas Treatment," EPRI Report FP-925,
October 1978.
(3) Maxwell, J. D., Burnett, T. A., Faucett, H. L., "Preliminary Economic
Analysis of NO Flue Gas Treatment Processes," EPA Report EPA-600/7-80-
021, February 1980.
(4) Maxwell, J. D., Burnett, T. A., Faucett, H. L., "Preliminary Economic
Analysis of N0X Flue Gas Treatment Processes—TVA and EPRI Premises,"
final report in preparation for EPRI Contract 783-3.
(5) Scheck, R. W., Damon, J. E., Campbell, K. S., and Jones, G. D., "N0X
Control For Western Coal-Fired Boilers—Feasibility of Selected Postcom-
bustion Systems," final report in preparation for EPRI Contract 783-2.
(6) Swann, D. W. and Drissel, G. D., "Feasibility of Retrofitting Catalytic
Postcombustion Controls on an 80 MW Coal-Fired Utility Boiler," EPRI
Report CS-1372, Feburary 1980.
335
-------
u>
GO
680
Rue Gas Temperature (°F)
600 520 440 360
280
Catalyst
no/no2
Potential Fate
Fly ash
>so2
e^so3
V t V V
.v.v
i # i
2^m^2S>(NH4) hso4
{M» 1)2 S04
>NH3
• SO2 scrubber
• Retained on surface
• Emitted
• ESP/FF
Gas phase
Solid
Liquid
Approximate air heater
inlet temprature
Approximate air heater
outlet temperature
Figure 1. Fate of Major Nitrogen and Sulfur Species
-------
Reactor module
#1
#2
Regenerative #3
wheel segments
Hot
Intermediate
Cold
Scale (feet)
Stair
/"tower
Ammonia/carrier gas
injection grid
Return duct
/ to station
Air heater
(horizontal flow)
NOx Ammonia vaporizer
reactor and control unit
Figure 2. NOx Reactor and Air Heater
-------
Figure 3. Cynical Shaped Catalyst
338
-------
CO
CO
vo
NO (NO + N02)
h2o
NO (NO + NH3) + N02)
-dP-
*
High/low temperature
converter
Figure 4. Continuous NH3 Analyzer
-------
2350 space velocity
NH3/NOx ^ 0.95
520
560
~r
600
Reactor Temperature (°F)
640 680 720
760
800
840
T
NOx removal (% inlet NOx) 93 —
Residual NH3 (% inlet NH3) «*-
1.5
(1.5)
SO3 oxidation (% inlet SO2) <1% (<2%) ¦
Minimum continuous; catalyst |
damage from deposited sulfates
Permanent deterioration I y
(minimum 12 hours)
Approximate economizer .
outlet temperature
(minimum load)
¦*>95
->•4%
Maximum continuous;
SO3 formation excessive
(full load)
Sintering
(maximum 1 hour)
Figure 5. Influence of Reactor Temperature on Operation and Performance
-------
Gas Side Pressure Drop On)
8
7
6
5
4
3
2
1
0
0 5 10 15 20 25 30
Days of Continuous Operation
Figure 6. Influence of SO3/NH3 in Rue Gas on Air Heater Pressure Drop
Sootblowing
Air
8-hour cycle
90-100 psig
With NH3/SO3
in flue gas
Without NH3
in flue gas
-------
Figure 7. Regenerative Wheel Basket
-------
Figure 8. Deposit Accumulation on Regenerative Wheel Basket
(cold face of intermediate segment)
343
-------
TABLE 1
MINIMUM IDENTIFIABLE COSTS OF KAWASAKI HEAVY INDUSTRIES
SELECTIVE CATALYTIC REDUCTION PROCESS
~CAPITAL REQUIREMENT: '
*LEVELIZED OPERATING COST:
41 - 60 $/Kw
4.0 - 5.9 mills/kwh
Study Basis
Design Premises
• 90% N0X removal (300 ppm $ 5% O2 initial N0X)
• 500 MW (net), 548 (gross) generating capacity
• Midwest location
• Economizer bypass employed for low loads (40 - 60% Maximum
Rated Capacity)
• Low-sulfur, sub-bituminous coal (typical of Powder River Basin
In Wyoming, Montana)
Economic Premises
• 1979 dollars
• Escalation factor 6% per annum
• Capital equipment inflation 6% per annum
• Levelizing period 30 years
* from Reference No. 5, "N0X Control for Western Coal-Fired Boilers-Feasibility
of Selected Postcombustion Systems"
344
-------
Table 2
SCR OPERATING ISSUES
Operating Issues
To be Addressed
1. Performance (de-n1tr1f1cat1on effi-
ciency)
2. Process Operating Demands
i. amnonia
Information Required
Economic Implications
for Plant Operations
b. heat rate penalty
c. reactor sootblowing and
Materwashlng
d. catalyst life
• Inlet/outlet NOx
• aitmonla demand vs SCR operating
parameters
• restrictions 1n flue gas exit
temperature due to limitations In
catalyst operating temperature
• auxiliary power, additional flue
gas pressure draft
• Influence of sootblowing frequency
on deposit accumulation
• change In catalyst activity
• quantity of catalyst required for
given NOx emission level
• cost of continuous aimonla
supply
• Increased fuel cost
• capital and operating costs of
sootblowlng facilities; and
effluent disposal
• catalyst replacement
3. Environmental Impacts
a. SO3
b. «h3
c. ammonium sulfate and
bl sulfate
d. Interference with particulate
collectlbll 1ty
• oxidation of SO2 to S03 across
catalyst, preheater exit concen-
tration
• carryover or breakthrough of ammonia
to reactor outlet; preheater exit
concentration
0 preheater exit concentration
• concentration In air heater
Mterwash effluent
• influence of NH3/SO3 and by-
products on
- resistivity
- physical and chemical
ash characteristics
«. N2O, HCN emissions
• Inlet/outlet N2O
and HCN concentration
e cost of SO3 removal and disposal
• cost of NH3 removal and disposal
• cost of ammonium sulfate and
blsulfate removal, and disposal
• capital and operating costs
of waterwash effluent treatment
• capital cost of electrostatic
precipitators
- specific collecting area
- ash hoppers
• capital and operating cost of
fabric filters
- air/cloth ratto
- cleaning cycle,
pressure drop
• ash disposal costs
• cost of KjO and HCN renoval
4, System-Wide Effects
a. air heater sootblowlng and
Materwashlng requirements
b. air heater operation and
maintenance
• degradation in thermal performance
or pressure drop with deposit
accumulation
• degradation 1n thermal performance
or pressure drop with pheel basket
corrosion/fouling
• Increased fuel costs (heat rate
penalty)
• capital and operating costs of
sootblowlng/waterwasMng facilities
• Increased capital and operating
costs of corrosIon/fouling resistant
air beaters
345
-------
TABLE 3
KAWASAKI HEAVY INDUSTRIES
SELECTIVE CATALYTIC REDUCTION
PROCESS-DESIGN CONDITIONS
• Flow capacity 5000 scfm (8000 Nm3/hr)
t Catalyst quantity,space velocity 120.1 ft3 (3.4 m3), 2500 hr^
• Flue gas temperature 680°F (360°C)
• Fly ash loading
• Flue gas composition
3.3 gr/scf (7.5gm/Nm3)
Specie
02
N0X
S0X
C02
H20
CO
Volume Concentration
5*
600 ppm
350 ppm
11%
9%
15%
• N0X removal - 90% of Inlet
for design flue gas conditions
" NH3/N0X<1
- NH3 <10 ppm (prediction only, not guaranteed)
346
-------
TABLE 4
LJUNGSTROM AIR HEATER DESIGN SPECIFICATIONS
Flue Gas
Mass in 23,000 lb/hr
Mass out 26,700 lb/hr
Temperature in
Temperature out
Pressure drop
675°F
300° F
3.2 in H20
Air
Mass in 19,850 lb/hr
Mass out 16,150 lb/hr
T + , 85°F
Temperature in
618°F'
Temperature out
1.3 in H20
Pressure drop
347
-------
Controlling
Parameter
Continuous On-Line
Measurements and
Processing Required
EPRI
Experience
to Date
NH3/NOx
1. Rue gas flow rate and
temperature
2. NOx concentration
3. Ammonia mass flow
Limited
NOx removal
1,2
None
Residual NH3
4. NH3 flue gas
concentration
None
IMH3/NOx, maximum
limited by NH3
1, 2, 3 and 4
None
NOx removal, maximum
limited by NH3
1, 2, 3 and 4
None
Table 5. Ammonia Injection Control Schemes Under Evaluation
-------
Table 6.
TASK DEFINITION
TASK
Primary Variables
Fixed Parameters and/
or Secondary Variables
3. Baseline Characterization
SCR operating parameters
- NH3/NO* ratio
- NO*
• ffue gas/catalyst temperatures
space velocity
• fuel type
• particulate characterization
• minor gas species
- SO2
- HC
- CO
4. Transient Node Operation
• control system logic
• excursions In process gas
. - flowrate
- temperature
- composition
• fuel type
e particulate characteristics
e minor gas species
5. Influence of Minor Species
• S02
• CO
• HC
• particulate characteristics
e fuel type
• SCR operating parameters
6. Influence of Particulates
particulate mass loading
particulate size distribution
sootblowing schedule
(accumulated solids)
minor gas species
SCR operating parameters
7. Catalyst Activity
• erosion, aging of catalyst
(hours of operation)
• particulate characteristics
• minor gas species
e SCR operating parameters
8. Air heater performance
air heater operating
parameters
- pressure differential
- wheel speed
- Inlet temperature
- flue gas/combustion
air flowrate
- sootblowing frequency
• particulate characteristics
e 100 tfcl owing. medium and momentum
349
-------
DEVELOPMENT OF FLUE GAS TREATMENT IN JAPAN
By:
Y. Nakabayashi, H. Yugami, and K. Mouri
Electric Power Development Company, Ltd.
8-2, Marunouchi 1 chome, Chiyoda-ku,
Tokyo 100 Japan
350
-------
ABSTRACT
The Electric Power Development Co., Ltd. (EPDC) has been executing a
research and development program on selective catalytic reduction (SCR) sys-
tems through joint research with the equipment manufacturers since 1975.
As a result of this R&D program, EPDC believes there is a strong
prospect for commercialization of the Low Dust SCR System (LDSS) for coal-
fired power plants. At present, EPDC is constructing demonstration test
equipment for such a system at the Takehara Thermal Power Station No. 1
Unit (250 MW coal-fired) and plans to construct full scale commercial
SCR equipment at the Takehara Thermal Station No. 3 Unit (700 MW coal-
fired) .
Alternatively, the High Dust SCR System (HDSS) is also a candidate
for commercialization, requiring only the establishment of NH3 removal
technology for ash collected by a cold-side electrostatic precipitator.
This report describes the results of the R&D program executed by
EPDC concerning the SCR and air preheater problems, overall flue gas
treatment technology for coal-fired boilers, and the outline of Takehara's
SCR Systems.
351
-------
Section 1
INTRODUCTION
The oil shortage which gripped the world in the fall of 1973 triggered
a series of oil supply uncertainties in the years that followed, emphasizing
the importance of renewed efforts to increase the utilization of coal.
Coal-fired thermal power generation will be the major focus of
this shifted emphasis to coal. A coal-fired thermal power plant, however,
must be well controlled in order to satisfy all environmental protection
requirements. Otherwise, the utility may encounter resistance from
governmental agencies or local residents, resulting in failure to obtain
a site.
A thermal power plant using coal as fuel, compared to one using other
fossil fuels, emits larger quantities of dust (particulates), SOx. and NOx
from its boiler, necessitating special measures to hold down these
emissions. Such controls include auxiliary equipment for dust collec-
tion, desulfurization and denitration, including combustion modifications.
Electric Power Development Co. (EPDC), in cooperation with certain
Japanese manufacturers, has been conducting research and development
on technologies for the control of emissions from thermal coal-fired
power generation plants. Currently, assured of successful commerciali-
zation of boiler denitration technology, EPDC is installing SCR demonstra-
tion equipment on a coal-fired thermal power plant. In addition, EPDC
has just started construction of a 700 MW coal-fired thermal power plant
featuring similar SCR equipment.
352
-------
In this paper, we introduce several problems encountered in placing
denitration equipment in commercial operation together with remedial mea-
sures, as well as details about denitration performance, ammonia injection
control, and other parameters. Also, we will outline our progress in
the development of emission control technologies for other emission from
a coalrjrired thermal power plant.
353
-------
Section 2
GENERAL DESCRIPTION
INSTALLATION OF SCR EQUIPMENT FOR ELECTRIC POWER UTILITIES IN JAPAN
Table 1 shows a list of Selective Catalytic Reduction (SCR) equipment
installed or in the planning stage by Japanese electric power utilities.
Of the plant sites shown in the table, two are operating with SCR equip-
ment for coal-fired power generation, and another two are being installed
with similar equipment. Of the coal-fired power generation plants listed
in the table, the Tomatoazuma Power Plant of the Hokkaido Electric
Power Co., Inc. is provided with SCR equipment designed on a 1/4-scale
capacity; the Shimonoseki Power Plant of the Chugoku Electric Power Co.,
Inc. with equipment designed for full capacity.
EPDC has been commissioned by the government to construct SCR equip-
ment designed for the full capacity of the No. 1 boiler (250 MW) of the
Takehara Thermal Power Plant. This demonstration SCR equipment is
designed to operate at 80 percent or more NOx removal. In addition,
EPDC plans to provide the recently constructed No. 3 thermal power
generation boiler (700 MW) with similar SCR equipment designed to
operate at 80 percent or more NOx removal. These plans should be
considered representative of future programs for construction of coal-
fired thermal power generation plants.
EVALUATION OF DENITRATION PROCESS
Studies of the denitration process directed to coal-fired thermal
systems have evaluated the following:
• economic efficiency
• performance
354
-------
o reliability
o secondary environmental emissions
o retrofit capability
Japanese NOx regulations are stringent; thus it was important that
the DeNOx Process be quickly evaluated, and brought to commercialization
in a short period of time. As a result of a comprehensive analysis,
EPDC concluded that selective catalytic reduction (SCR) using ammonia
was the most suitable process for near-term application.
The comparison of results between SCR and selective non-catalytic
reduction (SNR), including the practical use of wet DeSOx, is shown in
Table II.
CATALYST COMPARISON BY TYPES OF FUEL
Research and development of the SCR process for oil-fired boilers
has been executed since 1973, and for coal-fired boilers since 1975.
Process development for power plants has focused on the development of
a suitable catalyst. Coal combustion results in large quantities of
dust and S0X becoming entrained in the flue gas,thus, catalyst develop-
ment for coal combustion has special problems such as deterioration and
plugging. Table III shows catalyst features for different types of fuel.
DEVELOPMENT OF SCR FOR COAL-FIRED BOILERS
It is not possible to apply the results of SCR development from
LNG and oil-fired boilers to coal-fired boilers because of differences in
flue gas components; thus there is a need to assess the practicality of
several catalysts for coal-fired boilers.
Among power utilities in Japan, EPDC, Tokyo Electric Power Company
(TEPCO), and Hokkaido Electric Power Company (HEPCO), have individually
tested the SCR process through pilot tests at their own power plants.
Table IV shows the outline of the pilot tests.
355
-------
Section 3
DEVELOPMENT OF AN SCR SYSTEM
EPDC, in developing the SCR system has identified the following needs:
• catalyst life must be greater than one year
• the system must be appplicable to both high and low dust loading
• continuous measuring instruments for leak ammonia and counter-
measures for leak ammonia influence on downstream equipment
are required.
• recycle or reutilization of spent catalyst
DEVELOPMENT OF CATALYST LIFE OVER ONE YEAR
For practical use of SCR, the following targets must be satisfied
during the one year of pilot tests:
• maintaining more than 80% efficiency
• leak ammonia should be less than 5 ppm
The following discussion details how these targets have been met
by the 5 joint researchers of EPDC.
Catalysts with 2 years of life have been promised by the manufacture:
although no experiments have yet been performed testing catalysts con-
tinously for over 2 years. The catalyst cost is relatively high, com-
prising 40-60% of the total SCR equipment cost. Whether catalyst life
is 1 or 2 years will be a significant influence on electricity cost;
therefore, it is important to develop long life catalysts. At this
time no assessment of catalyst life is possible as pilot tests of
sufficient duration have not been conducted. However, the following
catalyst performance test results suggest extended life is probable:
• Performance decreased slightly from initial valires but
stabilized after 500 - 3000 hours.
356
-------
• Significant performance drop was observed using catalysts
without treatment for S0X poisoning; treated catalysts
performed slightly better.
Catalyst performance drop is thought to be caused by the follow-
ing:
(a) plugging of catalyst pores by NH3/SO3 compounds
(b) plugging of catalyst layer and catalyst pores by dust
(c) catalyst erosion by dust
(d) permanent posion by fluorine, chlorine and calcium
Our experience from pilot and laboratory tests indicates neither
significant plugging of the catalyst layer and pores nor dust erosion.
Thus, poisoning by calcium or similar elements is suggested, making it
necessary to check the fuel calcium content before applying SCR to power
plants.
Catalyst damage caused by temperature variation associated with
daily load changes or start-up and shut-down is of concern and also will
have to be evaluated through long term testing.
SCR SYSTEM
Two systems for SCR have been developed, a Low Dust SCR System (LDSS),
which removes NOx after removing dust (by a hot-side electrostatic
precipitator), and a High Dust SCR System (HDSS), which removes NOx
before removing dust (by a cold-side electrostatic precipitator). These
systems are shown in Figure 1.
Low Dust System vs. High Dust System
The low dust and high dust SCR systems each have their own merit and
demerits. For comparison of these systems, it is necessary to examine
not only the SCR equipment, but also the entire coal-fired thermal power
generation system, in accordance with Table V.
357
-------
Before details about the results of these studies are discussed,
it must be pointed out that the choice between HDSS and LDSS, aside from
the technical details involved, depends on a combination of several factors,
such as environmental goals, relevant regulatory standards, and the goals
of individual electric power corporations. Under these circumstances,
it is yet to be determined which of these two systems will be adopted in
Japan.
Overall Thermal Power Plant System
The following lists several problems characteristic of LDSS and
HDSS SCR equipment:
• Compared with HDSS, LSDD offers a greater potential for the air-
preheater being blocked, necessitating the conventional type
air preheater to be replaced with a special one intended for
SCR.
• The LDSS, in which a precipitator (hot side) is installed
upstream of SCR, will not contaminate collected ash with NH3;
the HDSS, in which a precipitator (cold-side) is installed down-
stream of SCR, will collect NH3 ash contaminated.
To use NH3 contaminated ash as landfill material could require
certain countermeasures, depending on the quantity of NH3 entrained and
the specific dlposal circumstances.
The Table V.I shows the results of examination with reference to other items.
Precipitator
A major problem associated with the HDSS system concerns the possibi-
lity of leak ammonia mingling with fly ash, leading to a possible
decline in quality of the fly ash. It seems technically possible to
minimize the leak ammonia to the low levels of 5 ppm or less, which,
in our opinion, will not cause any deterioration in fly ash quality.
358
-------
SCR Equipment
Table VII shows the results of comparing the HDSS and LDSS. Of
particular interest is the fact that both these systems are free from the
problem of the catalyst layer being blocked, provided special considera-
tion is given to the catalyst shape and the gas flow speed. LDSS also
appears more tolerant to surface masking by fine particulates than
the HDSS. The latter phenomenon was witnessed using a scanning electron
microscope to examine the catalyst surfaces. Thus, LDSS requires the
installation of a soot blower.
However, EPDC, in cooperation with certain Japaneses manufacturing firms,
has developed a catalyst capable of maintaining a specified NOx removal
capacity for a period of one year without a soot blower. Thus, EPDC
presently has no intention of using a soot blower at this time.
Desulfurization Equipment
Another problem characteristic to SCR lies in the possibility of the
ammonia entering the desulfurization system and becoming concentrated in
the rejected solid waste. In such a case, it is necessary to install
additional NH3 removal equipment for the LDSS system. However, the HDSS
allows the leak ammonia to be almost fully adsorbed by the fly ash
and removed through the cold-side ESP, thus eliminating the need to
install equipment for significant removal of NH3.
The Table IX shows the results of comparing LDSS and HDSS.
LEAK NH3 FROM SCR EQUIPMENT - POTENTIAL IMPACTS AND COUNTERMEASURES
The following discussion is directed to the problems attributable to leak
NH3 from SCR equipment. EPDC has been executing R&D programs to establish
countermeasures for:
• Concentration of ammonia in Wet DeSOx (HDSS and LDSS).
• Plugging of the air preheater (HDSS and LDSS).
• Fly ash contaminated with NH3 and collected by electrostatic
precipitators (Refer to Figure 2).
359
-------
Removal of NH^/N in Waste From Wet DeSOx
EPDC has executed R&D programs in cooperation with Japanese manufacturers
for removing NH3 and N concentrated in Wet DeSOx System. This R&D was
successful, resulting in total waste water nitrogen (N) less than 10 ppm.
Presently, EPDC is constructing N and NH3 removal processes for the
SCR demonstration test equipment at Takehara power station No. 1 unit.
Specifications of the process are as follovs:
• Capacity 1,000 nrVday
• Quantity of NH4+ 198 Kg-N/day
• Total N at Outlet 10 ppm and below
Plugging and Corrosion of the Air Preheater
Plugging and corrosion of air preheaters has been addressed since the
initial phases of SCR development. EPDC has been executing R&D programs for
this problem in cooperation with air preheater manufacturers, using Ljungstrom
or Rothemulle pilot test equipment. To date, results have been obtained on the
Ljungstrom equipment. For the HDSS case, plugging was not observed, because of
the self cleaning and scouring action of the dust. For the LDSS case, this
scouring action does not exist; thus some countermeasures to plugging are
necessary. Countermeasures investigated shown in Table XII.
Plugging tendencies of air preheaters with the LDSS system were confirmed
in 1976. Work since then has not identified the plugging problems even
after one year of testing. Additional tests to define countermeasures for
LDSS air preheater plugging and the effects of such countermeasures have been
conducted and the results are shown in Figure 5.
Major countermeasures being evaluated for plugging include:
• Pressure, steam quantity and time of soot blowing,
and improvement of the soot blower design
• Change from DF to NF type of inter-element, and a
one piece element for intermediate and cold segments.
360
-------
The air preheater for LDSS applications is larger than that for HDSS
and therefore less economical. Future plans call for continued development
of a new soot-blower, with pilot tests in near future. For the Rothemulle
air preheater, detailed results are described in IHI's paper in this symposium
proceedings.
REGENERATION OF THE CATALYST
EPDC has been executing an R&D project in cooperation with Kawasaki
Heavy Industries concerning regeneration of spent catalyst. The objective
is to imporve catalyst utilization and thus the economics of the SCR system.
Four methods of regeneration have been evaluated, based on (1) steam heating,
(2) purging with high temperature flue gas, (3) water cleaning, and (4) water
cleaning using sonic impulses. Each of these methods was applied to catalysts
forced to deteriorate in pilot test plants. The results have confirmed that
reactivation of the deteriorated catalyst to the initial performance level
is possible.
Tests are currently being conducted with the reactivated catalyst in
the pilot plant, and no deterioration of catalyst performance has been
observed after 3,000 hours.
SCR PERFORMANCE
EPDC has specified 80% N0X removal as design conditions for performance.
However, we are targeting for levels as high as 90% and above while simultaneously
maintaining a maximum of 5 ppm leak NH3 by employing changes in catalyst
volume and OT^/NO* ratio. An increase in NOx removal efficiency from 80% to
90% under these conditions requires an increase in catalyst volume of
30-50%. Thus, high removal efficiencies require greater capital expenditure,
and necessitate higher draft loses, and higher SO2 to SO3, conversion rates.
This trend is exhibited in Figure 6 which shows the relationship between space
velocity (SV) and NOx removal efficiency. These data indicate that the increase
in catalyst volume, required from to move from 80% removal (.8 NHj/NOx) to
90% removal (.9 NH^/NO^j) is approximately 50% (5,400 + 3,800 h ~1) .
361
-------
A systeipwide analysis of SCR operation at 80% removal efficiency will
be conducted on the demonstration test equipment of Takahara Unit No. 1.
Depending on these results, a similar analysis of systemwide operation at
90% removal efficiency may be conducted.
DEVELOPMENT OF A CONTINUOUS NH3 ANALYZER
Continuous analysis of NH3 in flue gas is required to (a) precisely
measure the concentration of injected NH3, and (b) monitor residual NH3
to indicate the potential for air heater problems and upsets in SCR operation.
Continuous analysis for NH3 on both a conductivity and infrared gas basis
has been explored. However, these methods appear to be unfavorable for use
as process-monitoring devices due to unacceptable response time, limits
of detection, and stability.
Recently, a method employing the conversion of NH3 to NO was developed.
This method infers the NH3 concentration by measuring the difference in the
sample before and after converison of NH3 to NO. Difficulties with this
technique include (1) a large measurement error due to high background NO
levels, and (2) short lifetimes of the NH3 to NO converter.
Conditions are most favorable for this method when the gas sample line
length is relatively short. Otherwise, ammonia is adsorbed in the inner
surface of the sampling line at low temperature, or lost by the chemical
reaction between NH3 and SO3. For coal-fired flue gas, this approach appears
capable of NH3 measurement with dust in the sample gas while requiring
acceptable levels of maintenance.
EPDC has also evaluated a continuous NH3 analyzer developed by Anritsu
Electric Co., Ltd. based on the self-modulated derivative spectrometry of
the ultraviolet absorption of NH3 molecules. The NH3 analyzer has been used
since 1977 in SCR tests executed by EPDC and the Hitachi Shipbulding and
Engineering Co., Ltd. (HZ).
362
-------
The principle of this NH^ analyzer is based on the following:
1. Ammonia molecules have a periodic absorption spectrum (shown in
Figure 7)
2. The light transmitted into the ammonia gas also has a periodic
spectrum (shown in Figure 8a).
3. The wavelength is modulated around the intensity minimum
(Figure 8b).
4. Intensity-Modulated Light Single (IMLS) is obtained (Figure 8c)
and the amplitude of IMLS is proportional to the depth of the
intensity minimum (the depth is proportional to the concentration
of NH3).
5. The concentration of NH^ can be measured by the amplitude of
IMLS.
A good correlation (correlation coefficient (q) of 0.997) is obtained
between the analytical values obtained by the NH3 analyzer and those by the
standard method (indophenol). The scattering deviation for coal-fired
flue gas observed by the standard method (s = 10.2 ppm) is larger than the
value obtained by the NH3 analyzer (s = 2.9 ppm). The results obtained by
using the NH3 analyzer for coal-fired flue gas are shown in Figure 9.
Interference from S02 is shown in Figure 10. By subtraction of the inter-
ference signal, good correspondence has been obtained between the analytical
values using the standard method and those obtained by the NH3 analyzer.
To prove the reliability of the NH3 analyzer, the device will be used
in the demonstration test at Takehara No. 1 unit.
Future study of the NH3 analyzer will concern:
1. The reliability of the measured value at low concentrations (such
as 1 or 2 ppm).
2. Interference from SO2 concentrations over 1,000 ppm compensated
by employing an automatic calculation method.
3. Locating the NH3 analyzer near the sample probe.
363
-------
Section 4
DEVELOPMENT STATUS FOR FT,HE GAS TREATMENT TECHNOLOGY
AND LOW POLLUTION COAL-FIRED POWER STATIONS
COMMERCIALIZATION CLASSIFICATION AND DEVELOPMENT STAGE OF FLUE GAS TREATMENT
TECHNOLOGY FOR COAL-FIRED BOILERS IN JAPAN.
EPDC has been executing R&D programs on flue gas treatment technology
concerning emissions of dust, SOx, and NOx from coal-fired boilers. As
a result there are numerous processes (specified in Table XIII) both under
development and in the commercialization stage for application in Japan.
The current status of development and/or commercialization is show in Table XIV.
MODEL OF A LOW POLLUTION COAL-FIRED POWER STATION
Numerous combinations of flue gas treatment systems are possible for
the control of dust, S0X, and N0X from coal-fired power stations. Models
of three combinations are shown in Figure 11, and are anticipated to meet
target emission levels similar to that from oil-fired power stations (i.e.,
60 ppm NOx, 50 ppm SOx, and 10 mg/Nm3 dust). Detailed specifications of
these three models are presented in Table XV.
Among the three cases for S02 removal, the Dry DeS0x System (Case 3)
is quite attractive to EPDC in terms of simplicity, economy, performance,
and reliability. Thus, EPDC is highly interested in R&D for Dry DeS0x
and it is a key project of Case 3. EPDC is planning to perform a demonstra-
tion test at the EPDC Matsushima Power Station (500 MW coal).
TREND OF EPDC'S R&D PROGRAM IN THE FUTURE
R&D on N0X and SOx is scheduled to be virtually complete by March
1981. However, dust emission control in Japan may be tightened, thus
altering this schedule. Accordingly, dust collecting technologies such as
bag houses and high performance ESPs with particle charging system are
drawing attention.
364
-------
EPDC has been promoting R&D of emission control technology for modern
coal-fired power plants necessary to meet social needs. Establishing such
technologies has received formost attention. In the near future, R&D will
be directed to optimizing such processes for economy, reliability, and
performance.
365
-------
Section 5
DEMONSTRATION TEST AND COMMERCIALIZATION PLAN
After assessing the commercialization potential through pilot-scale
development efforts, a demonstration test to identify commercial capabilities
is needed. The Japanese government is sponsoring demonstration tests
conducted by EPDC to achieve this end.
THE DEMONSTRATION DeNOx TEST OF TAKEHARA POWER STATION NO. 1 UNIT
EPDC's demonstration DeNOx plant sponsored by the Japanese government
and initiated in 1978, is now under construction. Construction is 20%
complete, and testing will be initiated in July 1981.
Tlje concept of the Takehara No. 1 unit demonstration test is shown
in Table XVI.
NHq Injection and Its Control
As the NH3 injection system of the SCR demonstration equipment, at Takehara
No. 1 unit shows (Figure 12), NH3 is mixed and diluted with air in the dilu-
tion mixing chamber and injected before the SCR. To keep the NH3 concentra-
tion constant, the NH3 injection system is designed as follows:
• Sufficient distance is maintained between the NH3 injection point
and the SCR unit to mix NH3 adequately.
• Uniform velocity distribution is maintained by means of guide
vane installation at duct bends, or installation of a gas adjustment
grid upstream from the NH3 injection point.
• A constant mixing ratio between NH3 and dilution air is maintained
or the diameter of the NH3 injector nozzle can be optimized to
keep this dilution concentration and dilution gas speed constant.
Control of the NH3 feed rate is accomplished by injecting NH3 at a
constant ratio to the N0X quantity in the flue gas. An input control
defines the set value for NH3/N0X ratio. NOx concerntration at reactor outlet
is monitored with the NH3 feed rate controlled by feed-back signals from the
reactor outlet N0X concentration. Figure 13 shows the control system for
NH3 injection.
366
-------
TAKEHARA NO. 3 UNIT PLAN
Takehara No. 3 unit initiated construction in May 1980. Commercial
operation will begin in March 1983. The Takehara No. 3 unit flow system
(LDSS) is shown in Figure 14, with the layout in Figure 15. This facility
will be largest coal-fired power plant in Japan. Levels of N0X, S0X, and
dust anticipated to be emitted are shown on Table XXII and, as can be seen,
are extremely low.
DRY DeS0x DEMONSTRATION TEST
EPDC is planning to install dry DeS0x demonstration test equipment
at the Matsushima Power Station (700 MW coal) in April 1983. This demonstration
test is funded by the Japanese Government. EPDC had designated Sumimoto
Heavy Industries (SHI) to build the dry DeSO^ system since EPDC has been
executing an R&D project on dry DeS0x in cooperation with SHI.
Table XVIII shows the schedule and Table XIX shows outline of this demonstra-
tion.
REFERENCES
1. Y. Nakabayashi, Status of R&D on N0X removal in Japan and results of
EPDC's R&D for DeN0x Process.
Second EPRI N0X Control Technology Seminar, Denver Colorado, Nov. 1978.
2. T. Kimura, Y. Nakabayashi, K. Mouri, Overall Flue Gas Treatment
Technology from Coal Fired Power Plant in Japan, ICCR 5th Conference
Dusseldorf, Sept. 1980.
367
-------
Figure 1
High Dust Denitration System (HDDS) and Low Dust Denitration System (LBDS)
HDDS
Air
Preheater
Boiler
C-ESP
SCR
->-To Stack
GGH Wet FGD
LDDS
Boiler
Air Preheater
H-ESP
SCR
To Stack
Wet FGD
-------
Figure 2
Problems Depend on Leakage NH3 from SCR Process
Plugging by NH3/S03
Compounds
SCR
H. ESP
A/H
Wet
FGD
To Stack
Coal
Boiler
A/H
SCR
C.ESP
Waste Water
Treatment
Fly Ash
Handling
Fly Ash with
nh3
Concentration of
NH3
-------
Figure 3
Biological Nitrogen Removal Method
in Waste Water from De-SOx Process
Simple Sludge Type Organic Carbon Source
.•Living Energy
Increasing Energy
pH Control
Settling
Chamber
Settling
Chamber
Sludge Return
Sludge Return
Nitrogen
Removal
Tank
Nitration
Tank
Increase of Increase of Nitrogen
.BOD Oxidation Germ Removal Germ
Nitration Germ
Figure 4
Experiment Flow Diagram of Air Preheater for Denitration Process
Boiler
— st*ck
Air
Preheatei
OF
All1
Preheater
'DF
370
-------
Figure 5
Draft Loss Trend of Test Air Freheater and Conditions (Ljungstrom Test)
CO
200
150
100
50
"Tokr
SO3 injection started
K
"zdou 3060 zrdoD sutoo 5000 Tote si'rao—gnrkr
Test hour (H)
Gas Condition
S02
1200 - 1300 ppm
| S02
1200 - 1300 ppm
S03
3-5
ppm
1 SO3
|
20 ppm
Dust
0.1 - 0.2 g/Nm3
| Dust
|
0.1 - 0.2 g/Nm3
NH3
10 ppm
| NH3
10 ppm
Gas Temp of AH inlet
325-337*C
J Gas Temp of AH inlet
1
325-337°C
Gas Temp of AH outlet
145-165°C
1 Gas Temp of AH outlet
_l
145-165°C
-------
Figure 6
One Sample of
SV vs Nox Curves
372
-------
Figure 7
Absorption Spectrum of Ammonia
6000
5000
4000
3000
2000
1000
180
200
220
250
Wavelength (mm)
373
-------
Figure 8
Principle
(a)
Spectrum of transmitted light
Wavelength
I(t)
Intensity-modulated light signal
(2wt)
60 -H
Xj Xo X2
Time
t
Wavelength
(b) (c)
374
-------
Figure 9
300
200
= 0.997
= 10.2 ppm
100
Concentration of NH3 (ppm)
Figure 10
10
0
500
000
-10
Concentration of SO2 (ppm)
375
-------
Figure 11
Model Anti-pollution Coal-fired Thermal Power Plant
Case
Sys tem
CO
O-l
Low Dust
Denitration
System
High Dust
Denitration
System
Power Plant Flow
NH3
Boiler H.ESP
SCR
*70
—o
Vatef
-o-1
*"TOGH
Wet
FGD
TJviT-
Boiler
SCR
NH. Ash
Treatment
Water
—C>*-
Wet
FGD
tea
WWT
Con-
centration
Remarks
H.ESP:Hot-side
Electrostatic
Precipitator
SCR: Selective
Catalyst
Reduction
Equipment
A/H: Air Preheater
GGH:
Gas-Gas
Heat Exchanger
FGD: Flue Gas
Desulfuriza-
tion
Dry TGD
System
Boiler
-o
NH3
L
A/H
BH/New
EP
-O—
Dry FGD
BH
W.WT: Waste Water
Treatment
Equipment
BH:
0:
Bag House
Fan
-------
Figure 12
NH3 Injection Flow
Gas adjustment grid
Flue gas
Air
Flue gas
SCR reactor
NH3 injection nozzle
NH3 dilution and
mixture chamber
377
-------
Figure 13
Fundamental Control System for NH^ Injection
Air volume
for combustion
NOx cone
at inlet
NOx conc.
at outlet
Outlet
O2 Conc.
NH3 inj ection
volume
Correct O2
Gain adjustment
/ Set target
of outlet NOx
NH^^rr^ctioi^jadjustment value
H/A
P+I
P+I
H/A
FG
z:
378
-------
Figure 14
Overall Flue Gas Treatment System
co
to
M
O
4-)
(0
)-i
a)
C
8
CO
o
2
High tem-
perature :
electric
precipitat: on
BUF
(Low-nitration)
FDF
Stack
De-
sulfu-
rizer
Deni-
zation
device
Gas/
gas
heater
Boiler
Air
p re-
heater
-------
Figure 15
Layout
blow tank
No. 1 unit outlet
ndoor switchyard
I
Ho. 2 unit outlet
unloading •!*.#
I
stockpile
1
oooo
No. 2 unit
Intake channel
No. 3 unit
intake channel
. 1 unit
intake channel
No. 3 unit intake «*.»
N/V \
unit Intake
!°* """Mi-a brido.
No. 3 onlt extension
-------
Table I DeMOx Installation In the Power Utilities of Japan
OJ
oo
Power Company
Process
Manufacture
Power Plant
Unit
Out Put
(WW)
Fuel
Start-up
Gas Volume
(Nm3/a)
Capacity
Efficiency
<11
Remarks
Hokkaido EPCO
SCR
BHK
Tomato-Asumi P/S
1
350
Coal
Sep.
1980
280,000
25
90
Test Equipment
Tohoku EPCO
SCR
IHI
Hlgashi-Niigata P/S
2
600
H.0, Crude,
Aug.
1981
1,660,000
100
—
LNG
Tokyo EPCO
SCR
MHI
Yokosuka
4
350
H.0
Feb.
1978
Chubu EPCO
SCR
BHK
Chita
5
700
LNG
Mar.
1978
1,910,000
100
80
SCR
BHK
Chita
6
700
LNG
Mar.
1978
1,910,000
100
80
SCR
MHI
Chita
4
700
H.0
Nov.
1979
1,960,000
100
80
SCR
mi
Atsutni
3
700
H.0
Dec.
1980
1,900,000
100
80
SCR
KHI
Atsumi
4
700
H.0
Feb.
1981
1,900,000
100
80
SCR
MHI
Shin-Nagoya
3
220
H.0
Jul.
1980
650,000
100
80
SCR
IHI
Shin-Nagoya
6
500
H.0, Crude,
etc .
Jun.
1980
1,316,000
100
Hokuriku EPCO
SCR
IHI
Toyama-Shinko
2
500
H.0, Crude
Nov.
1981
1,370,000
100
—
Kansai EPCO
SCR
BHK
Kalnan
450
H.0
Jun.
1977
300,000
25
80
Compact SCR
BHK
Anagasaki-Hlgashi
1
156
H.0
Jul.
1978
466,000
100
30
Compact SCR
MHI
Osaka
1
156
H.0
Jun.
1978
490,000
100
30
SCR
HHI
Osaka
3
156
H.0
Jul.
1980
552,000
100
80
SCR
MHI
Osaka
4
156
H.0
Dec.
1979
552,000
100
80
SCR
MHI
Sakaiko
1
250
H.0
Jul.
1980
800,000
100
80
SCR
MHI
Sakaiko
6
250
H.0
Dec.
1979
800,000
100
80
SCR
MHI
Osaka
2
156
H.0
Jul.
1981
530,000
100
80
80
SCR
Mil
Tanagawa
3
156
H.0
Aug.
1981
520,000
100
SCR
MHI
Tanagawa
4
156
H.0
Mar.
1981
520,000
100
80
SCR
MHI
Sakaiko
2
250
H.0
Sep.
1981
800,000
100
80
SCR
MHI
Sakaiko
4
250
H.0
Feb.
1981
800,000
100
80
SCR
MHI
Sakaiko
7
250
H.0
Sep.
1981
800,000
100
80
SCR
IHI
Hineji
5
156
H.0
Jun.
1978
490,000
100
—
Chugoku EPCO
SCR
Mil
Iwakuni
2
350
H.0
Dec.
1980
980,000
100
80
80
SCR
MHI
Iwakuni
3
500
H.0
Apr.
1981
1,400,000
100
SCR
tffll
Shimonoseki
1
175
Coal
Apr.
1979
550,000
100
60
(80% in near future)
SCR
IHI
Kudamatau
2
375
NGL, H.0,
Apr.
1979
1,050,000
100
—
Crude
SCR
IHI
Kudamatau
3
700
NGL, H.0,
Sep.
1979
1,900,000
100
—
Crude
Kyushu EPCO
SCR
MHI
Shlnkokura
3
600
LNG
Jun.
1978
1,690,000
100
>75
SCR
MHI
Shinkokura
4
600
LNG
Jun
1979
1,690,000
100
>75
EPDC
SCR
BHK, KHI
Takehara
1
250
Coal
Jul.
1981
400,000x2
100
80
SCR
Not decided
Takehara
3
700
Coal
Mar.
1983
2,414,000
100
80
Joban Kyodo TPCO
SCR
IHI
Nakoso
9
600
Coal, H.0
Apr.
1983
1,700,000
100
—
(80% in near future)
SCR
Mil
Nakoso
8
700
Coal, H.0
Dec.
1982
1,700,000
100
>50
BHK
——
Total
H.0
July,
1978%
Total
July,
1980
2,586,600
MHI
Total
Coal
June,
1982%
Total
306
July,
1982
1,166,000
—
IHI
-
Total
H.0, Crude
April,
1978%
Total
2387
May,
1982
6,848,000
-------
Table II Comparison of DeNOx Processes
Process
Economy
Performance
Secondary
Reliability
Development
NOx SOx
Pollution
Ope-
ra-
tion
Main-
tenan-
ce
Danger
SCR
3 '
>80% >90%
None
Easy
Easy
None
1^2 years
SNR
1
>30% >90%
None
Easy
Diffi-
cult
Hydro-
gen
use
1^2 years
SCR+SNR
2
>60% >90%
None
Easy
Diffi-
cult
None
1^2 years
Dry DeSOx/
DeNOx
(CuO Type)
A
>80% >80%
None
Dif-
fi-
cult
Diffi-
cult
Hydro-
gen
use
3^4 years
Wet DeSOx/
DeNOx
5
>80% >90%
By Pro-
duct as
no3
Dif-
fi-
cult
Diffi-
cult
None
1^2 years
382
-------
Table III Feature of Catalyst for LNG, Oil
and Coal
AV: Area Velocity
Fuel
Countermeasure
Shape
Cost to 1 unit
Dust
Sox
LNG
None
None
Pellet
1
Oil
None
None
or Need
High AV
Honeycomb, pipe
board etc.
2
Coal
Need
Need
Low AV
Honeycomb, pipe
board etc.
3
383
-------
Table IV SCR Pilot Tests for Coal Fired Boiler
Company
Joint
research
Power
plant
Capacity
(Nm3/H)
Shape of
catalyst
HDSS or
LDSS
Schedule (Fiscal Year)
1976
1977
1978
1979
1980
1981
EPDC
IHI
Isogo
1000 x '2
1000 x 2
Honeycomb
HDSS
LDSS
H Z
Isogo
230
140 x 2
Honeycomb
HDSS ¦
LDSS *
MHI
Takasago
200 x 2
600 x 2
Honeycomb
HDSS
LDSS
KHI
Takehara
250 x 2
4200
Pipe
HDSS
LDSS
BHK
Takehara
2300
1330
Board
LDSS
HDSS
^^m
Tokvo
EPCO
MHI
Nakoso ^
4000
4000
Honeycomb
HDSS
LDSS
Hokkaido
EPCO
BHK
Ebetsu
2000
Ring
(Moving
Bed)
LDSS
(1) Fuel of Nakoso is 70% oil and 30% coal.
-------
Table V Comparative Items between
HDSS and LDSS
Equipment
Items
(1) Overall System
Plant Efficiency, Layout, Duct Work
Countermeasure for GRF erosion
Countermeasure for A/H Plugging and
Draft Loss, Treatment of Fly Ash with NH3,
Behavior of SO3
(2) Precipitator
Treated Gas Volume, Efficiency, Size
Heat Expansion, Quality and Properties of
Fly Ash
(3) SCR
Erosion of Catalyst, Shape of Catalyst
Catalyst Volume, Catalyst Life, Production
Method of Catalyst, Gas Velocity
Reactor Structure, Draft Loss, Leakage NH3
(4) Wet DeSOx
Gas Heat Exchanger
Waste Water Treatment
385
-------
Table VI Comparison between HDSS and LDSS
(Overall System)
Items
HDSS
LDSS
° Plant Efficiency
Base
Decline
° Layout
Base
Larger
° Countermeasure for Erosion
Use Multi Cyclone
Not necessary
° Countermeasure for A/H
(D Reduction of leakage
ditto
Plugging
nh3
(2) Low S02/S03
conversion catalyst
(3) Improved A/H for SCR
° Draft Loss
Base
Increase
° Treatment of Fly Ash
Contained NH3 into Fly
No trouble
with NH3
Ash
Need NH3 removal treat-
ment under centain
circumstances.
386
-------
Table VII Comparison between HDSS and LDSS (II)
(Precipitator)
Item
HDSS
LDSS
O
Gas Volume
Base (130 - 150°C)
Approx. 1.8 times
(370 - 380°C)
o
Efficiency
Base
Applicable to various
kinds of coal.
o
Size
Base
Generally bigger
o
Heat Expansion
Small
Large
o
Quality of
Fly Ash
Contain NH3 but
No trouble
No problem
o
Treatment of
Fly Ash
Simple device as
low temperature ash
Need insulator, burn
protections and ash
cooler for high
temperature ash.
387
-------
Table VIII Comparison between HDSS and LDSS (III)
(SCR)
Item
HDSS
LDSS
O
Influence by Dust
Need anti- erosion for
catalyst
No problem
O
Shape/peculiarity
of Catalyst
Parallel passage type
Such as honeycomb, pipe
board etc. and low
erosion, material
Same
o
Catalyst Quantity
Base
Nearly Same
o
Catalyst Life
2-3 years
(Presumption)
3-4 years
(Presumption)
o
Product Method of
Catalyst
Base
Same
o
Gas Velocity
Base
Same
o
Draft Loss
Base
Same
o
Leakage NH3
Base
Same
•
388
-------
Table IX Comparison between HDSS and LDSS (IV)
( DeSOx )
Item
HDSS
LDSS
a DeSOx
No problem
No problem
o GGH Plugging/
Corrosion
Less countermeasure
for plugging/corrosion
to LDSS as NH3 and
SO3 is precipitated
by ESP
Need the counter-
measure such as
strong soot blow,
element material
etc.
o Waste Water
Treatment
Need De-N device to
blow dow
De-N device is 2 or 3
times to HDSS
389
-------
Table X De-N Development for DeSOx Waster Water
Company
Joint Researcher
Process
Capacity
Test Location
Schedule (Fiscal year)
1977
1978
1979
1980
EPDC
Hitachi Zosen
(HZ)
Biological
Labo. test
MAX 208,/D
Field test
MAX 5 m3/D
Labo. test
3J2./D
HZ's Labo
Takehara P/S
HZ's Labo (1)
-
Mitsui-Hi ike
Machinery
(MMM)
Biological
Labo. test
Field test
Labo. test
MMM's Labo
Ditto
Ditto (1)
—
—
Ebara Infilco
Biological
Labo.test
Field test
Labo. test
El's Labo
Ditto
Ditto ^
—
Kirita (K)
Biological
Labo. test
K's Labo
Organo
Biological
Labo. test
O's Labo
Note: (1) De-N and De-COD simultaneously.
-------
Table XI Outline of Air Preheater Test
Company
Joint Researcher
(Type of A/H)
Test Location
HDSS or
LDSS
Capacity
(Nm3/H)
Schedule (Fiscal year-)
1976
1977
1978
1979
1980
EPDC
Gadelius
(Ljungstrom)
Takasago P/S
HDSS
LDSS
LDSS
LDSS
10,000 1
5,000
5,000
8,000
Hitachi Zosen
(HZ)
(Rothemuhle)
Takehara P/S
Isogo P/S
LDSS
LDSS
2,000
IHI
(Rothemuhle) .
Isogo P/S
HDSS
LDSS
2,000
2,000
Tokyo
EPCO
MHI
Nakoso P/S
HDSS
LDSS
4,000
4,000
-------
Table XII Identification of A/H Plugging
Qndition^
High dust
No NH3 injection
High dust
NH3 injection
Low dust
NH3 injection
Low dust
NH3 injection
Plugging
None
None
None
Yes
Test Condition
15g/Nm3
15g/Nm^
10 ppm
0. lg/Nm^
0.1g/Nm3
10 ppm
392
-------
Table XIII Flue Gas Treatment Techniques
^ Coal Flue Gas
vo
co
Combustion modification-
Denitration-
ESP
H. ESP
L. ESP
New ESP-
<~— BH
Desulfurization/
dust collecting-
SCR
Two-stage combustion
Low NOx burner
Gas recirculation
Gas mixing
Combined measures
Others
High dust system
i- Low dust system
SNR
ih Hybrid denitration (SNR+SCR)
Dry simultaneous desulfurization
and Denitration
Wet flue gas
desulfurization
Dry flue gas desulfurization
— Wet flue gas denitration
Boxer charger
3rd electrode
Pulse charge
Ionizer
Mixing tower system
Separating tower system
Mixing tower system
Saparating tower system
-------
Table XIV Present Status of Research and Development (1/2)
CJ
vo
4*
Boiler
H.EP
SCR
A/H
Remarks
Combustion
modification
S N R
High dust
Low dust
Regeneration
High dust
Low dust
(1) Steps of re-
Under R&D
Ditto.
in practical
under R&D
practicable
R&D
practicable
practicable
SNR: Selective Non- SCR: Selective
search and
use
Catalytic Catalytic
development
Reduction Reduction
(2) Co-researchers
Babcock Hitach,
MH2, IHI
SHI,(Gadelius
HZ, IHI, MHI,
Ditto
KHI
Gadelius IHI
Gadelius, IHI,
IHI, MHI, KHI
Hitachi, »JI)
KHI, Hitachi
HZ
(3) Development
lOOppm and belov
i 40% and
100mg/Nm3
80% and above
To operate
targets
(6%02> 5% and b€
-above
and below
5ppm and below j
stably for
low in unburned
lOOppm
0.5 - 1.0% and f
one year or
Ditto.
carbon,with
and
below
1 Ditto
Ditto
more.
CO and HCN
below
(performance
showing no
after the lapse
increase.
of one year)
(4) Results so far
Guidelines for
0 NOx is
0 Specifica-
Development
Ditto
0 Fundamental
0 Tendency
0 Soot blowing
achieved
NOx reduction
not af-
tions
targets have
techniques
toward
has been in-
measures have
fected by
necessary
been cleared.
for regene-
plugging
tensified.
been set up and
fly ash.
for design
ration have
has been
And element
impr o vemen t s
0 Furnace
of H.ESP
been estab-
confirmed.
has been im-
have been made
tempera-
have been
lished.
proved .
through testing
tures hav«
obtained.
0 Secular
by type of
been ac-
changes in
coals.
tually
performance
measured.
are being
(Isogo
checked af-
P/S)
ter comple-
tion of flue
gas regene-
ration test-
ing .
(5) Subjects in
To develop a
6 To study
0 To investi
® To develop
(Optimum
B Secular chant
e 0 Concrete
To make a
the future
boiler of lOOppn
a simple
gate the
ammonia
design)
after regene-
ideas to
survey of
and below
denitra-
effects of
ash dispo-
ration
improve
the opti-
° To develop an
tion
Na and Ca.
sal
commer-
mum design
ultra-low NOx
means,
process
cialized
burner
0 To study
equipment
° To make a
cooling
survey of
systems
optimum condi-
and
tion
° To deve-
lop an
NH3 ash
disposal
process.
(6) Time of putting
200ppm cam be
0 After
Under cons-
After deve-
Demonstra-
Not fixed yet
Practicable
Demonstra-
into practical
guaranteed at
develop-
truction in
loping the
tion test
tion test
use
present.
ing the
Matsushima
aanonia ash
plant is
plant is
NH3 ash
Power
disposal
under cons-
under cons-
disposal
Station
process
truction in
truction in
process
No.1 unit.
No.l unit.
Takehara
Takehara
Power
Power
Station.
Station.
-------
Present Status of Research and Development (2/2)
Dust Collector
G G H
Desulfurization
Wet EP
A/B
B H
iRemarks
Boxer Charger
B.H.
Net
Dry
(1) Steps, of re-
search and
development
Under R&D
Ditto.
In practical
use
Already put
into prac-
tical use
Under R&D
Ditto
Already put
into prac-
tical use
Under R s D
(2) Co-researchers
IHI(Toshiba)
Hitachi
SHI(IHI)
Gade, IHI
-
SHI,(IHI)
IHI(SHI)
-
SHI
(3) Development
targets
lOug/Nn^ and
below
lOmg/NB^
below
To operate stably
for one year or
¦ore.
(95% and
above)
95* and
above
(T)nox
40% and
above)
10mg/Nm3
and below
10mg/Nm3 and
below
(4) Results so
far achieved
Basic data has
been gathered
Laboratory
tests have
been com-
pleted.
Design
conditions
with and with-
out NH3 have
been confined.
riSOx and
SOx perfor-
mance has
been already
confirmed.
Performance
confirmation
(5) Subjects in
the future
Testing with
pilot plant for
confirmation
of practical
use
To nake sure
of filter
cloth life
and perfor-
mance and
to optimize
design
conditions
Optimum design
(measures for
clogging with
HH3)
Performance
of collect-
ing sub-
micron par-
ticules,
SO3, etc.
Performance
confirmation
Optimum
design
(6)Time of putting
into practical
use
Mot fixed yet
Not fixed
yet
Under construc-
tion in
Matsushima Power
Station
(Case without
leakage of
Mo.3 unit under
construction in
Takehara Power
Station
(Case with
leakage of
anaemia)
Already put
into prac-
tical use
Demonstra-
tion test
plant is
expected to
be provided
in Matsushima
Power
Station.
(Entrusted
by govern-
ment)
Not fixed yet
Demonstra-
tion test
plant is
expected to
be provided
in Matsushima
Power
Station.
(In a combina-
tion with dry
PGD)
-------
Table XV Features of Anti-Pollution Coal-Fired
Power Plant by System
Case
System
Flue Gas
Emission Level
Prospects of
Practical Use
Denitri-
fication
Counter-
measure
for A/H
Ammonia
Ash
Treatment
Lay-
out
Combustion
Modification
1.
Low dust
denitration
NOx 60 ppm and
below
SOx 100 ppm
and below
Dust
30 mg/Nm^
Practicable
(700MW under
construction
in Takehara
P/S)
Required
Required
Not
required
Base
Combustion modified
to 300 ppm and
below in NOx.
Already put into
practical use.
2.
High dust
denitration
Ditto
Within several
years.
(Ammonia ash
treatment
techniques
have not been
established
yet.)
Required
Not espe-
cially
required.
Required
Better
Ditto
3.
Dry Desul-
furization
NOx 60ppm and
below
SOx lOOppm and
below
Dust 1 mg/Nm3
Within several
years
Not
required
Not
required
Required
Best
NOx must be
reduced to 100 ppm.
A modification
means is being
developed.
Existing
-------
Table XVI Outline of SCR Demonstration
Plant at Takehara No„ 1 Unit
Item
Manufacture
Type
Capacity
Efficiency
Remarks
(1) Existing Plant
(a) Boiler
BHK
Reheat Type
Radiant
Boiler
810 T/H
Fuel: coal
(b) Plant output
-
-
250 MW
-
(2) Demonstration
Plant
(a) SCR
BHK
KHI
Board Cat.
Pipe Cat.
400,000 Nm3/H
400,000 Nm3/H
>80%
>80%
(b) Air Preheater
Gadilius
Long element
Ljumgstrom
800,000 Nm3/H
-
Now Air
Preheater
(c) De-N
HZ
Biological
Process
1,000 m^/Day
<10ppm
as total N
-------
Table XVII Flue Gas Emissions from Takehara No. 3 Unit
ITEM
Unit
Numerical Value
Gas volume (Wet)
Nm3/H
2,414,000
(Dry)
2,169,000
Discharge temperature
°C
100
Stack height
m
200
SOx concentration
ppm
100
NOx concentration
ppm
60(02=6%)
Dust
g/Nm3
0.03
398
-------
Table XVIII The Demonstration Schedule
Item
1981
1982
1983
1984
Foundation
Installation
Trial Operation
Test Operation
1 8
8
8
, =3
Table XIV
Outline of Dry DeSOx Demonstration Test Equipment
Item
Contents
0
Process
Dry Type Activated Carbon Absorption
0
Gas Volume
300,000 Nm3/H
o
Gas Temp.
135°C
0
S02 (Inlet)
1,000 ppm
o
SO2 (Outlet)
50 ppm and below
o
NOx (Inlet)
300 ppm
0
NO* (Outlet)
225 ppm and below
0
Dust (Inlet)
300 mg/Nm3
0
Dust (Outlet)
10 mg/Nm^
0
By product
Element Sulfur (purity 99.9% <)
399
-------
TREATING FLUE GAS FROM COAL-FIRED BOILERS FOR NOx REDUCTION
WITH THE SHELL FLUE GAS TREATING PROCESS
By:
Jack B. Pohlenz and Albert 0. Braun
UOP Process Division
UOP Inc.
Des Plaines, Illinois
400
-------
ABSTRACT
Copper as copper sulfate (C11SO4) is one of the group of metals which at
tempeatures of 350-450°C catalyzes the selective reduction of N0X in flue gas
to nitrogen and water with ammonia (NH3). Conversions and efficiency (ammo-
nia utilization) are high, resulting in low concentrations of both NO and NH3
in the treated gas.
If flue gas containing both sulfur and nitrogen oxides and the reductant
ammonia is processed over copper at 400°C, the copper is converted first to
the oxide, then to the sulfate, and N0X reduction begins. As the conversion
to copper sulfate continues, the N0X content of the treated gas decreases to
a minimum value and the S0X increases.
Copper sulfate can be reduced with a variety of fuels, e.g., H2, CO,
CH4, etc., at 400°C, yielding a concentrated stream of SO2, along with water
and the copper in elemental form.
Thus, the copper system provides the technical base for flue gas
treating capable of SOx reduction, N0X reduction, and the simultaneous reduc-
tion of both. It offers the potential of a dry process, without by-products,
and with modest energy requirements.
401
-------
SFGT PROCESS
The application of this technology to a NOx-only process is in the form
of a fixed bed reactor with provisions for ammonia introduction and mixing.
The reactor is of special design to accommodate fly ash. The catalyst is
contained in woven wire baskets suspended in the gas stream in such a way
that the gas flows in the open channels between baskets (see Figure 1), The
reactants enter and the products leave the catalyst bed by radial diffu-
sion. The design is modularized in units or cells, one-half meter square by
one-meter long, which can be stacked one atop the other for the required
space time. The flue gas from one half MW, i.e., 1,000 SCFM, is processed in
a single stack of cells.
For S0X removal, the operation is cyclic and requires that the reactor
be isolated from the flue gas circuit for regeneration, producing a concen-
trate of SO2 and restoring the copper to elemental form. Continuous flue gas
processing is achieved with multiple reactors, at least one of which is
always in regeneration. A unit designed for SOx removal can operate with NOx
reduction only by elimination of the regeneration step, and as a simultaneous
N0x/S0x unit by addition of ammonia.
This reduction practice is called the SFGT process. Commercial
applications have been in operation since 1973 treating flue gas from various
fuels not Including coal.
402
-------
PILOT/DEMONSTRATION UNIT
UOP has operated an SFGT pilot plant for several years at the Tampa
Electric Company's Big Bend Station near Ruskin, Florida. The pilot unit
treats a slip stream of flue gas from one of the station's coal-fired boilers
as the gas leaves the economizer. A simplified process flow is shown in
Figure 2. De-SOx only, simultaneous de-SOx/de-NOx and, most recently, de-NOjj-
only operations have been conducted. The current program is sponsored by the
U.S. Environmental Protection Agency.
The demonstration has involved two acceptor/catalysts: the first is in
commercial application, the second, a prototype, is a more active formula-
tion. The standard experimental procedure for each has been to establish a
reference performance under fixed operating conditions using air plus pollut-
ants followed by an evaluation on flue gas. Reference performance is again
obtained with air plus pollutants before completion of the run. These tests
are supported by similar evaluations in a laboratory unit.
SUMMARY OF DEMONSTRATION PROGRAM
NOx-only evaluation of the first acceptor/catalyst was made on a fully
sulfated acceptor. For four weeks, 90% NOx reduction was achieved with a
1,25 ratio of ammonia to NOx. effect of the ratio on the de-NOx perform-
ance in the reference test with air is shown in Figure 3; with flue gas, the
result is the same but for a translation of the curve requiring an increase
in the ratio of 10%. It has been characteristic of this system in this
service that introduction of flue gas is followed by an immediate drop in
performance and then stable operation.
Another characteristic of the system is that space time has little
effect on NOx reduction, that is, an increase of gas rate by 50% is accom-
panied by a similar percent increase in the de-NOx reaction rate resulting in
the same NOx reduction. Both these characteristics suggest that the reactor
is limited by the depletion of a reacting component which in this case is
expected to be ammonia.
403
-------
In the simultaneous N0x/S0x testing with the first acceptor/catalyst,
the unit was operated for a period of approximately four weeks with 90% SOx
reduction and 70% NOx reduction with a 1.25 ratio of ammonia to N0x» As in
the N0x-only tests, introduction of flue gas into a clean reactor system
results in an immediate drop in performance followed by stable operation.
The change in performance is due to a partial blinding of the screens and the
filling of the interstitial volume with fly ash.
Reference performance for the more active acceptor/catalyst with air
plus pollutants is shown in the on-line analyzer strip charts in Figure 4.
Time is right-to-left and was arbitrarily chosen at 99 minutes. Half-way
through the acceptance, S0x/N0x instantaneous reduction is 98/91%; cumulative
reduction at 99 minutes is 96.0/89.7.
In Figure 5 is shown the same strip charts when processing flue gas at
the same conditions. The corresponding values of S0x/N0x reduction are
90.6/92.5 and 88.9/85.2. Note that de-SOx performance shows the usual
decline with high loadings of fly ash and SO2 content (loading of copper with
sulfur), but the instantaneous de-N0x performance remains high and the cumula-
tive reduction is decreased due to the "slip" of NO in the first ten minutes
of the cycle while the copper is converted to copper oxide.
In Figure 6, also with flue gas, the rate of formation of CuSO^ has been
reduced to achieve 90% de-SOx in 99 minutes, and the initial slip of NOx
decreased by delaying the injection of NH3 for a few minutes into acceptance
and with a preoxidation of the copper following regeneration. After 99
minutes, the cumulative reduction of SOx/NOx was 90.9 and 90.0.
The acceptor/catalyst has proven to be quite stable in this service.
During stability tests carried out in 1974-1976, there was 13,000 cycles of
oxidation-reduction with no significant deterioration in performance. This
stability has been further supported in the current demonstration work, both
in the field and the laboratory.
404
-------
Operation of the TECO unit will be concluded with the testing of a
de-NOx catalyst of high activity that can function over the temperature range
of 200-450°C. This catalyst is at present in commercial service at 200°C but
not on coal. The performance will be determined in the parallel-passage
reactor over a range of temperatures, space velocities, and NI^/NOx ratios
with coal-derived flue gas.
405
-------
FIGURE 1
THE PARALLEL PASSAGE REACTOR
REGEN. GAS
PURGE OFF-GAS
REGEN. OFF-GAS
TREATED
FLUE GAS
FLUE GAS
PURGE STEAM
UOP 163-3
UOP 576-1
406
-------
FIGURE 2
SIMPLIFIED FLOW SCHEME
DEMONSTRATION UNIT FOR COAL
UTILITY BOILER AT TAMPA ELECTRIC, FLORIDA
TREATED
FLUE GAS
SEQUENCE
TIMER
REGENERATION
OFF-OAS
I I !
1
"1 REGENERATION
GAS
S02, NO
INJECTION
REACTOR
FLUE GAS
FROM DUCT
UP/DOWNSTREAM
PRECIPITATOR
BLOWER
407
-------
FIGURE 3
NOx REDUCTION vs. NH3/NOx
RATIO AT TAMPA ELECTRIC
DEMONSTRATION UNIT WITH A
COMPLETELY SULFATED BED
Z 100
o
E
g 80
ui
(ft
a
S 40
z
£
S 20
(A
Z
£ 0
NH3/NOX MOLE RATIO UOP 576-3
yv
RX NLET CONDITIONS
PROCESS GAS SOURCE AIR
FLOW 1600 NM3/HR
S02 2000-2400 ppmv
NOx 350-400 ppmv -
TEMP: 400°C
RX BED LENGTH 6 METERS
0.6 0.8 1.0 1.2 1.4 1.6 1.8
408
-------
FIGURE 4
100
90
80
70
60
50
40
30
20
10
0
CYC
LI N
O. At
>
INLI
RATI
TEM
T NO
IO
P.
>C
79 n
.19
oo°c
xiiv
NOV OUT
U-:
BUU
ipmv
/
/
/
y
r
100
90
80
70
60
50
40
30
20
10
0
CY
CLK
MO. .
to
S02 ACC IN/OUT
W"
wwv
W1
IV
I-
409
-------
FIGURE 5
CYC
;le ll
K>. 5
3
INLi
rr nc
»X 3
70 p
MTIV
A
Nl>X OUT
O-SOO ppmv
RATIO 1.18
.TEMP. 400°C
j
/
/
/
f
CY
CLE
ACC
4000
NO. 53
S02
0-
IN/OUT
DDmv
J\
.. _»
410
-------
FIGURE 6
100
90
80
70
60
50
40
30
20
10
0
100
90
80
70
60
SO
40
30
20
10
0
| |
|
.CYCLE NO. 18
9
INU
T NOv 4
B5 D
omv
NOX OUT
RATIO 1.28
KF""
r»
/
/
/
f
i—
£—
|
CY<
ZIM NO. 1
B9
802 ACC IN/OUT
/
411
-------
THE DEVELOPMENT OF A CATALYTIC NOx REDUCTION SYSTEM
FOR COAL-FIRED STEAM GENERATORS
By:
Tadamasa Sengoku, Yoshinori Todo and Naruo Yokoyama
Mitsubishi Heavy Industries, Ltd.
Tokyo, Japan
Brooks M. Howell
Combustion Engineering, Inc.
Windsor, Connecticut, United States of America
412
-------
ABSTRACT
Work done recently by Mitsubishi Heavy Industries in-Japan has
resulted in the design and successful operation of a full scale
catalytic NOx reduction system for coal-fired utility steam generators.
This paper describes the program carried out to evaluate the
commercial feasibility of catalytic nitrogen oxides removal from
coal fired power plant flue gases. Also discussed is the design
of a catalytic removal system for a large modern coal-fired central
station.
Testing of pilot catalytic systems on coal fired steam generators
was initiated at the Takasago station of EPDC in early 1977 using
plate type catalysts. After two stages of testing, the plate type
catalysts were replaced by grid type catalysts and testing was
resumed (fall 1979). Since the conversion to grid type supports,
more than 5000 hours of operation has been logged while maintaining
more than 80 percent NOx removal in both low and high dust load
environments.
At the Nakoso station of the Joban Joint Power Co. a grid type
pilot catalytic system has been operating for over 10,000 hours with
a removal efficiency of 84 percent. Operating under both high and low
dust loads, draft losses (as at Takasago) have been maintained at
low levels with only limited soot blowing during low dust loading
and no soot blowing during high dust loading.
At the Shimonoseki station of the Chugoku Electric Power Co.,
the first full scale system for a coal-fired boiler (175-MW) in the
world has run smoothly since startup in April of this year. Removal
efficiency has been 51 percent, as expected, and the ammonia slip
less than 1 ppm.
The paper also discusses ammonia slip, gas flow requirements,
catalyst life and catalyst blinding from fly ash. The design of
a 500-MW commercial unit based on the results of the test program
is described and the various factors affecting large commercial design
are discussed as well.
413
-------
INTRODUCTION
In most nations, SOx, NOx and particulates contained in flue gas from steam
generators must be minimized to meet environmental regulations. Particulates
are normally removed by electrostatic precipitators or bag filters and SOx by
tail-end flue gas treatment systems, but NOx removal is somewhat different.
The NOx levels in flue gas have been reduced considerably by combustion modificat-
ions. However, there is a minimum level of NOx cencentration beyond which
further reduction is not possible using these combustion modification techniques.
By using a catalyst and a reductant (NH^), however, the NOx generated in steam
generators can be reduced by a process of dissocation to form N2 and 1^0. This
is known as the Selective Catalytic Reduction (SCR) process and is recognized
as the most effective method for NOx reduction in Japan. The SCR process is
performed at a gas temperature of about 300 to 400 C, which usually exists just
upstream of the boiler air heater.
Since 1952, Mitsubishi Heavy Industries, Ltd. (MHI) has manufactured boilers
for electric power plants under a technical license with Combustion Engineering,
Inc. Almost all the boilers are tangentially fired. While combustion modification
in tangentially fired steam generators has been adequate to meet the regulated
NOx emission limits in the United States, the more stringent NOx emission regulat-
ions in Japan have led MHI to develop additional methods of NOx reduction: the
very low NOx PM burner and the Selective Catalytic Reduction process.
To comply with increasingly stringent NOx regulations in the U. S. this MHI
technology was licensed by C-E in April 1980 and will be available through C-E
for steam generators in the U. S.
The very low NOx burner technology has been presented in another paper (1)
at this conference, while this paper deals with MHI's SCR technology, focusing
especially on its application to coal fired steam generators.
414
-------
HISTORY OF DEVELOPMENT
MHI began developing a dry NOx removal process for boilers in 1971, concentrating
efforts on the SCR process in 1973. The first application was for clean flue gas
from liquefied natural gas (LNG) fired units, the second for semi-dirty flue gas
from low sulfur oil fired boilers, and finally for dirty flue gas from high sulfur
oil and coal fired boilers. Table I shows the SCR systems supplied by MHI.
As the clean flue gas from an LNG fired boiler does not contain particulate,
it was possible to use a pellet type catalyst of 3 mm diameter packed in fixed bed
reactors. A pilot plant with a capacity of 10,000 Nm /h was constructed at a 350-
MW boiler at the Tokyo Electric Power Co's, Minami-Yokohama power station. This unit
demonstrated a 10,000-hr catalyst life by the end of 1975, without any performance
deterioration (2). This successful test was MHI's first experience with the SCR
process and it provided the basic technical information for further development. The
results of this test were applied in the first, large full-scale system at the newly
installed 600-MW unit at Kyushu Electric Power Co's, Shinkokura station. It began
operation in June 1978 and has operated successfully since then.
In applying the process to the semi-dirty light oil firing flue gas, an inter-
mittent moving bed reactor was first developed using a pellet type catalyst, but it
was later concluded that fixed bed reactors using parallel passage catalysts, such as
plate or grid types, would be more suitable and practical for both operation and
maintenance. MHI carried out an evaluation program for both types of catalyst
simultaneously and, after considerable studies, concluded that the grid catalyst is
more economical for both oil and coal firing, primarily because of the compactness of
the reactor chamber.
Grid shaped catalysts used in parallel passage reactors are manufactured by ex-
trusion molding. Two types are commercially available; one is a homogeneous type,
*Normal cubic meters per hour
415
-------
where the catalyst is formed by a uniform blend of active and support material,
the other is a coated type, in which the active catalyst material is coated on
a ceramic support structure (substrate). Either type of catalyst may be mass-
produced through a series of processes to be within a predetermined tolerance of
dimension, strength and activation performance. Figure 1 shows one element of the
grid type catalyst.
The first commercial system using a grid catalyst was for an oil-fired boiler
at the Sodegaura Refinery of the Fuji Oil Co., a 160 t/h boiler (Fig. 2) put
into operation in December 1977 (3). This unit has operated smoothly for more
than two years without the replacement of the original catalyst. In January
1980 a 700 MW oil-fired unit, Chita 4, owned by the Chubu Electric Power Co., was
put into service. It is the largest SCR application to an oil-fired boiler
in the world.
To develop the SCR system for coal fired boilers, MHI constructed a pilot
plant known as the "DM-600 project" on a coal-fired boiler at the Takasago Power
Station of the Electric Power Development Co. (EPDC) and put it into operation in
January 1977 as a joint development project with EPDC. In March 1979, the
second pilot plant was put into operation at Nakoso Power Station of the Joban
Joint Electric Power Co. sponsored by Tokyo Electric Power Co., Tohoku Electric
Power Co., and Joban Joint Electric Power Co. and called the TTJM project. This
pilot plant consists of a precipitator to be usable either in the hot or cold
temperature mode, two SCR systems, one for high dust and another for low dust
loads, a wet desulfurization scrubber, and water treatment systems. The purpose
of this project was to investigate the overall performance of the total pollution
control package as well as individual systems. This configuration permitted
testing with the low dust conditions in the case of the hot electrostatic pre-
cipitator upstream of the SCR reactor and with high dust conditions in the case
of the low temperature electrostatic precipitator downstream of the air heater.
The test results of the TTJM project were used in designing the first SCR
system for a full-scale coal-fired commercial unit, the Shimonoseki 1 boiler owned
by the Chugoku Electric Power Co. This is a 175-MW unit with a high concentration
of fly ash. It was put into operation in April 1980 and has operated trouble free.
These projects are summarized in Fig. 3 and the test results are described
in a later section of this paper.
416
-------
DEVELOPMENT OF CATALYST
FEATURES
The catalyst for the SCR process is made of a mixture of active material
and support material. The active material has to be highly active, durable,
and formed to have sufficient surface and porosity to be efficient. In
addition, it must not adversely affect the boiler. Both materials must be low
in captial and replacement costs. The overall structure must be strong, compact,
and low in draft loss. A special consideration is that in most chemical processes,
temperature and pressure can be controlled for the most suitable reaction
conditions for the catalyst, but in flue gas treatment, the catalyst must be
developed to meet the boiler's operating conditions.
PROCEDURE TO DEVELOP THE CATALYST
MHI's policy regarding the production of catalysts for SCR plants is
to develop them with well established catalyst manufacturers who then manufacture
them in accordance with MHI specifications. To do this, MHI established a total
evaluation program of the catalyst development including trial formulation, evaluat-
ion of chemical activity and physical properties, and testing for durability.
This is illustrated in Fig. 4.
The first step in developing a catalyst is to select its composition and
determine the preparation method,'then, with the catalyst manufacturer's cooperation,
to establish the specification after evaluation of bench scale testing program
carried out at MHI's Hiroshima Technical Institute.
The catalyst manufacturers provide MHI with catalysts suitable for each
specific application. MHI carries out pilot tests to confirm the catalysts'
durability in actual flue gas. These tests consist of accelerated deterioration
tests, such as alkaline dust heating to predict the life expectancy of various
417
-------
fundamental tests, such as adhesion, clogging, and abrasion with different kinds
of dust, and testing the rejuvenation response of deteriorated catalyst. Moreover,
MHI is making a considerable effort to find the causes of catalyst deterioration
by evaluating various results of analyses such as X-ray diffraction, X-ray fluor-
escence, X-ray micro probe analysis, X-ray photoelectron spectroscopy, and mass
spectrometry. Physical properties are tested, such as surface, area, pore volume,
and strength against crushing, bending, abrasion, thermal shock, etc.
CONDITIONS FOR SELECTING CATALYSTS
There are large differences in SOx and dust concentration among exhaust
gases from various fuels. It is essential to select the most suitable catalyst
for each specific exhaust gas. The basic approach to the selection of the
catalyst for use with coal fired boiler exhaust gas is summarized as follows.
RESISTANCE TO SOx
Since coal-fired boiler exhaust gas normally contains high concentration
of sulfur oxides, the catalyst should be tolerant to SO2, the formation of SO^
by the catalyst becomes a particularly important problem because of its detri-
mental effect on the equipment downstream. A catalyst should be selected that
is capable of minimizing the SO^ oxidization rate. Figure 5 shows a comparison
between restrained and non-restrained catalysts in the formation of SO^.
WORKING TEMPERATURE
With an exhaust gas containing NH^ and SO^, the catalyst can deteriorate
from a low gas temperature. The boiler may have to be modified so that the
NOx removal plant can be continuously operated at a temperature above the allow-
able minimum working temperature of the catalyst. These minimum working
temperatures for various SO^ concentrations are known for operational results
obtained in the past.
LONG LIFE
It is necessary to select a catalyst that is stable thermally and stable
with regard to alkali compounds (K20 in particular), contained in dust that
can cause deterioration. As shown in Fig. 6, according to the forced deter-
ioration test using the alkali compounds as a pollutant, a catalyst of MHI
418
-------
specification, which consists of TiC^ as the support material and a proprietary
blend of two other active materials X and Y shows a much more stable perform-
ance than normal catalyst of a single active component.
FREEDOM FROM DUST DEPOSITION
It is necessary to select a catalyst with a structure (pitch) that will
not become plugged by dust contained in exhaust gas. Experience with
catalysts with 7 to 10 mm nominal pitch have shown satisfactory results
even with high concentration of dust.
SOOT BLOWING RESISTANCE
If dust should adhere to the catalyst, soot blowing is the most effective
way to remove it. The catalyst must be strong enough to withstand this
soot blowing. Figure 7 shows a comparison between our improved erosion
resistant type and a common catalyst.
RESISTANCE TO EROSION
Erosion of the catalyst from fly ash in exhaust gases cause a
decrease in the life of the catalyst. It is necessary to give special con-
sideration to this, especially when exhaust gases contain a high concentration
of dust. Without any counter measures for the catalyst, the leading ends
of catalyst can be severely eroded. MHI uses a catalyst with a high erosion-
resistance made by a propietary method. In addition to this, a section of
dummy catalyst support located upstream of the active catalyst can be
used to absorb the end impact and act as guide vanes, thus protecting the
active catalyst downstream. If the above measures are used and the appro-
priate gas velocity provided, the erosion of the catalyst can be eliminated.
Figure 8 shows the test results comparing erosion resistant catalyst with
common types.
419
-------
OUTLINE OF RESULTS IN FIELD TESTS
The operation of three coal-fired test plants as shown in Fig. 3 has provided
us with extensive information in establishing the design philosophy of NOx
removal systems. We can summarize these operating results as follows:
DM-600 Project
This pilot plant project was initiated in January 1977, as a joint research
program between MHI and EPDC for the purpose of establishing the selection
and durability criteria for catalysts for coal fired boilers. A flue gas
slipstream was taken from the 250-MW No. 1 coal-fired unit at the Takasago
power station of EPDC. In the first and second stages of this program, the
plate type catalyst developed by MHI was tested and it was proven that parallel
flow, fixed bed type reactors would work very will for either high or low
ash laden gas. These two stages of the program involved continuous operation
of more than about 10,000 hours.
As the third stage of the program, a grid type catalyst with 10 mm
pitch square holes was tested beginning in August of 1979. The composition of
the flue gas at the inlet of the reactor for the high dust system included
3
^12 to 15 g/Nm of dust, about 1500 ppm of SO and about 250 ppm of NO .
x x
Analysis of fly ash showed about 48% Si02 and 23% A^O^.
As shown in Fig. 9, it was confirmed that the NOx removal efficiency
could be maintained at more than 84% with less than 3 ppm of ammonia slip
at NH^/N0x ratio 0.85. The draft loss across the reactor did not change
throughout the test period for the high dust system. For the low dust system,
3
where the dust loading was about 50 mg/Nm , there was a slight increase in
draft loss across the reactor after 1000 hours operation, and it reached a
40mm increase after 3000 hours operation. But after soot blowing, the
draft loss increase was completely eliminated. Except for this, the low dust
system operation showed almost the same performance as the high dust system.
420
-------
These tests are being continued until the end of 1980.
Prior to pilot plant operation it was feared that alkaline compounds espe-
cially those of sodium and potassium could impair actibity of the catalyst by
chemical reaction with active catalyst materials. Such alkaline compounds
in fly ash are known to form a glassy complex of A^O^-CaCM^O-I^O in the
liquid phase and become non-active when cooled. After laboratory inspection
of catalysts taken from the pilot plant, we found no accumulation of alkaline
compounds in the catalysts, indicating that alkaline poisoning of the
catalysts can be eliminated.
According to laboratory experiments, the rate of catalyst erosion in-
creases rapidly when the gas velocity is greater than 10 meters/sec or when
the particulate size is over 40 microns in diameter. In the case of the high
dust system, while the average diameter of fly ash particle is about 20 microns,
considerable particles of more than 40 microns are present. This results
in an erosive condition. In a test with the gas flow rectified by an up-
stream dummy layer of catalyst support material and the gas velocity below
10 m/s, we eliminated the erosion of catalysts.
The smaller the particle size of fly ash and the more alkaline compounds
it contains, the more adhesive the fly ash will be. In the case where the
SCR reactor was located downstream of the hot precipitator, the dust particles
that escaped capture were mostly 3 microns or less in size. This material
proved to be somewhat sticky therefore soot blowers have had to be operated
periodically to remove this fine fly ash that collected on the surface of
catalysts.
TTJM Project
This project was initiated to establish an integrated flue gas treatment
system for coal-fired power plants. The pilot plant was installed at the
Nakoso power station of Joban Joint Power Co., Ltd. This boiler used mixed
combustion of coal and heavy oil (Coal amounting to about 40 percent during
the day time and to about 70 to 80 percent during the low load period at
night.) As shown in Fig. 3, the pilot plant consists of not only high and
low dust NOx removal equipment, but also includes wet desulfurizing equipment
but also includes wet desulfurizing equipment and waste water treatment
equipment so that the overall performance of the integrated clean-up system
in various combinations could be thoroughly tested. Testing was started in
421
-------
April 1979. At full boiler load, the flue gas in the high dust system contains
about 7 graras/Nm of dust, about 240 ppm of SOx and about 300 ppm of NOx with fly
ash composition having about 55% SiO^ and 27% Al^O^. As shown in Fig. 10 with
a vertical reactor chamber using a grid catalyst with an Sv of 2200 H * NOx
removal efficiency was maintained at more than 84% and less than 2 ppm of
ammonia slip with NH^/NOx ratio of 0.85 after about 10,000 hours operation. The
draft loss across the reactor chamber did not increase during the period of
3
high dust load operation. In the low dust system (100-300 mg/Nm dust loading)
the draft loss showed a gradual increase and reached about 60 mm 1^0 increase
after 3000 hours operation. At this point, the draft loss was successfully
reduced to the initial value through soot blower operation.
The regenerative air heater in the high dust system has the normal
arrangement of heating elements, DU type for high and middle temperature zones
and NF type for the low temperature zone. Air heater plugging due to the
deposit of ammonia bi-sulfate did not occur. This was attributed to the sand
blasting effect of highly concentrated fly ash and operation of the soot
blowers once a day. In the low dust system, the arrangement of air heating
elements was revised to use the SNF type (3.5 nun pitch) for the middle and low
temperature zones because draft loss increase across the airheater had been
observed with original airheater. With this modification and by operating
soot blowers both up and dwonstream three times per day, continuous pluggage-
free operation was attained. This is shown in Figs. 11 and 12.
FULL SCALE DEMONSTRATION
The Shimonoseki Unit of Chugoku Electric Power Co. is a 175-MW coal-
fired boiler and was constructed in 1967. The NOx removal equipment was
installed in 1980 with grid catalysts having 10-mm pitch square holes. The
reactor chamber is a vertical down flow unit and is located at the side of
air heater and precipitator. High dust laden flue gas is taken from the
economizer outlet and directed to the top inlet of the reactor and returned
to the inlet of the air heater. A provision has been made to keep the inlet
gas temperature above 330°C at lower loads by gas bypassing of the economizer.
422
-------
The system began operation in April, 1980 as the first full scale coal fired
SCR system in the world. The plant has been running since then without any
trouble at above design NOx removal efficiency, and ammonia slip concentration
less than 1 ppm. The design removal target for this system was at least 50
percent. Design conditions are shown in Table II and plant layout in Fig. 13.
The initial test results for NOx removal efficiency and ammonia slip con-
centration at various boiler loads are shown in Fig. 14. This data was taken
after one month operation, when the catalyst was in a fresh condition. However,
we are sure that deterioration of catalyst activity will not occur at least with-
in one year operation, judging from our test results at other pilot plants.
NO REMOVAL SYSTEM FOR 500-MW BOILER
X
OUTLINE OF DESIGN
The following outlines the design of a NOx removal system to be applied
to a new 500-MW coal-fired steam generator. The flue gas capacity is 1,553,000
3
Nm /h at 4.3% 0£ when firing a' typical sub-bituminous coal at the maximum con-
tinuous rating of the boiler. Design conditions for two cases, NOx removal
efficiency of 80% (Case I) and 90% (Case II) are shown in Table IV.
MHI has decided to use a 7 mm pitched catalyst for high dust systems on
coal fired boilers since it was proven that dust clogging would not occur on
the TTJM project. In this instance the reactor is located between economizer
and air heater and high ash-laden gas will be treated.
The specification of the equipment is shown in Table IV. The reactor
chamber is a vertical type where flue gas enters the chamber from the top and
passes downward through three or four stages of catalyst (Case I & II respect-
ively) . Each catalyst layer consists of a group of catalyst modules having a
cross-sectional dimension of about 2x1 meters a height of 1.65 meters, the
weight of a single module is about 3 tons and 216 or 288 modules are required
for Cases I & II respectively. Each catalyst module can be carried in and out
of the reactor chamber using an electric traveling hoist installed within the
steel structures of reactor chamber. One element of the catalyst is 150 x 150
mm square, about 650 millimeters in length, and about 14 kg in weight. A
total of 25.920 elements are required in Case I and 34,560 elements in
Case II.
423
-------
Eighteen traveling frame soot blowers in Case I or 24 sets in Case II using
steam as a blowing medium are installed for cleaning the catalysts. Low pressure,
orifice-type ammonia injection nozzles are installed upstream of the reactor
chamber, followed by an ammonia/flue gas mixing device. The plant layout for Case I
is shown in Fig. 15 and Figure 15 shows a schematic flow diagram of the system.
The predicted NOx removal efficiency and ammonia slip concentration curves are
shown compared to boiler load in Fig. 17 for both cases, and Fig. 18 shows the
relationship between ammonia to NOx ratio and NOx removal.
PROCESS DESIGN CONSIDERATIONS
There are three types of fixed bed catalyst reactor chambers where grid
catalyst elements are used: vertical up-flow, vertical down-flow and horizontal
flow. The vertical down-flow type is generally preferred in SCR systems for coal
fired boilers because it offers the following advantages:
(1) Less dust deposition on the catalyst surfaces and less plugging risk
than the horizontal flow type reactor chamber.
(2) Less dust deposition on the tope ends of the catalyst element than
the gas up-flow type reactor chamber.
There are two major factors to be carefully considered when the gas velocity
through the system is determined; one is the erosion of catalyst elements and the
other is the dust deposition on the catalyst during low load operation with
reduced gas velocity. The following are the results observed in the pilot plant
tests:
(1) There is no accumulation of dust deposits on the ends of catalyst layers
in vertical flow reactors when the gas superficial velocity is 2 m/sec
or greater. (Even on the inside surfaces of catalyst there is no
growth of dust deposit with such a low gas velocity.)
(2) A low gas velocity is desirable for minimizing erosion of the catalyst
elements. However, the catalyst elements developed by MHI can be
used without trouble with dust laden gas flows of up to 10 m/sec.
In designing a NOx removal system, therefore, the gas velocities are selected
taking the anticipated minimum and maximum load of the plant into careful considera-
tion. In the case of heavily dust-laden flue gas, an erosion resistant dummy
layer should be provided ahead of the catalyst layers to rectify the stream
of gas and dust.
424
-------
If the flue gas temperature becomes too low, the reaction products of SO^
and HH^ are formed on the active surfaces of the catalyst, impairing the capability
of the catalyst. However, should the flue gas temperature return to a high
level the deposits can be carried away by the gas stream, restoring the capa-
bility of the catalyst. The degree of reduction in NOx removal capability and
the period of time required for its restoration largely depend upon the gas
temperature and the SO^ concentration of the flue gas. Thus, the operating temp-
erature range of the catalyst should be determined to be suitable for the flue
gas condition.
When the flue gas has a high SO^ concentration, there is a possibility of
corrosion and' plugging in the air preheater due to the ammonia-sulfur compound.
The dust in the stack gas will also be adversely affected by S03 leaving the
air preheater. To avoid such difficulties, the SO2 to SO^ conversion due to
the catalyst should be controlled to a minimum. Therefore, the catalyst should
be controlled to a minimum. Therefore, the catalyst used in the NOx removal
systems should be tested to meet a maximum rate of conversion criterion of
SO2 to SOg of about 1% when treating a 1500 ppm SO2 concentration flue gas.
STRUCTURAL DESIGN
Figure 19 shows a cut-away view of the vertical type reactor chamber.
As the reactor chamber is a massive high temperature structure, its support-
ing points at four corners are allowed to slide to permit thermal expansion
of the structure. The reactor chamber is fixed only at a point in the middle
of its front wall. The midpoints of the other three walls are provided with
guide stoppers through which the horizontal forces exerted to the chamber
are conveyed to the supporting steel.
A number of catalyst elements are packed into steel grids to form catalyst
modules and these catalyst modules are placed on a rack having several
shelves. The gaps between modules are sealed to prevent short circuiting of
the gas; attention is also paid to prevent ash deposits. The racks are
fabricated from steel and supported at four bottom corners by the main frame,
which forms a part of the boiler house supporting steel structure. The
catalyst modules may be moved in and out of the reactor chamber through
openings on the side of the chamber by removing cover plates. To facilitate
handling of the catalyst modules, platforms are provided at the heights of
the respective module shelves. A hoist is provided to lift the modules to
425
-------
and from these platforms. Soot blowers are provided for each shelf of catalyst
modules and located at a right angle to the direction of module removal and re-
placement.
The ammonia distribution and mixing devices are located immediately up-
stream of the catalyst. These are constructed from mild steel pipe and structural
members to form a grid arrangement which assures thorough mixing of ammonia with
the entering flue gas.
CONCLUSION
To meet the increasingly stringent NOx emission regulations in highly
industrialized countries, a combination of combustion modification and flue
gas treatment will soon be necessary. To that end the development of the
Selective Catalytic Reduction system for boiler application has been success-
fully demonstrated in full scale commercial units for all fossil fuels.
426
-------
REFERENCES
Kawamura, Tomozuchi and Frey, Donald J., "Current Developments in Low
Nox Firing Systems," Presented at the EPA-EPRI Joint Symposim on
Stationary Combustion NOx Control, Denver, Colorado, October 6-9, 1980;
Published as Combustion Engineering publication TIS-6711, Windsor,
CT: Combustion Engineering, Inc., 1980.
Mitsubishi Heavy Industries, ltd., "Development of NOx Removal Processes
with Catalyst for Stationary Combustion Facilities, Mitsubishi
Technical Bulletin No. 124, May 1977.
Mitsubishi Heavy Industries, Ltd., "Practical Application of Flue
Gas De-NOx System Using Honeycomb Type Catalyst," MHI Technical Review,
17, 1, 11-18, 1980.
427
-------
Fig. 1: Grid-typa catalyst
6oil<
9«i
duet
Burner
General arrangement of 160 t/h oil-fired boiler
at the Sodegaura Refinery of the Fuji Oil Co.
428
-------
PROJECT
NAME A
LOCATION
SCHEDULE
'77
'78
'79
'80
'81
TEST PLANT FLOW DIAGRAM
No. 1 STAGE
DM400 (POU)
TADASAGO 1
ELECTRIC POWER
DEVELOPMENT CO.
FROM ECO.
OUTLET
No. 2 STAGE
600 Nm /H HEATER oe.-NOx M/C
HIGH DUST SYSTEM
-0~
IDF
TO
' DUCT
No. 3 STAGE
FROM LT.
E.P. OUTLET
600
Nm3/H
Q
HEATER DE.-NO„
LOW DUST SYSTEM
-e~
IDF
TO
DUCT
TTJM-40 (PDU)
NAKOSO 7
JOBAN JOINT
ELECTRIC POWER CO.
HIGH DUST SYSTEM
No. 1 STAGE
4000
Nm3/H
No. 2 STAGE
FRgM
OUT-
LET
K}
-0~
IDF
BAG-FILTER
4000 Nm3/H
TO DUCT
OE-SO,
40Q
GF GGH
—D—
777 IDF
HT-EP DE-NOx AH
LOW DUST SYSTEM
i
CI
WATER
TREATMENT
SHIMONOSEK11
(DEMONSTRATION)
SHIMONOSEK11
CHUGOKO ELECTRIC
POWER CO.
TEST OPERATION
560,000 Htrfl/H
ITP
TO STACK
DE-NO,
2)
vy
BOILER
£&—
-©¦
DE-SO,
e
MC LTEP
IDF BF GGH
tt=o
Fig. 3: Test schedule for pilot and demonstration SCR systems for coal-fired boilers
-------
I
h-
CATALYST MANUFACTURERS
"i n
i i
i r
M.H.I.
i
Trial Manufacturing
Commercial Production
L.
¦h
Trial Formulation
0
Selection of active and
support materials
l
Screening Test
0
Activity (micro reactor)
0
Physical properties
0
Durability (bench scale test
using boiler flue gas)
No
Yes
Preliminary Specification
Composition, Shape, Process
3 '
I
Fundamental
Tests
I
Accelerated
Deterioration Tests
1
o Oust adhesion
° Dust abrasion
[o SOx
Dust
o Heat
Pilot Plant Tests
° Coal
Evaluation
Improvement
Specification of
Commercial Catalyst
Commercial Unit Design
Fig. 4: Development of NOx reduction catalyst
430
-------
14
O SO, RESTRAINED
A COMMON CATALYST
10
SO, 500 PPM
S
0
360
400
300
GAS TEMPERATURE, °C
Fig. 5: Generation of S0^ due to catalyst
1.0
0.8
^ 0.6
>
& 0.4
<
0.2
°0 20 40 60 B0 100
ACCELERATED TEST HOURS, HR
Fig. 6: Test results of accelerated deterioration test with alkalimetal sulfat
431
TREATMENT TEMP 400°C
-------
*
>•*
5 «o
20
SB RESISTANT CATALYST
COMMON CATALYST
GAS TEMP.
360°C
SB MEDIUM
SUPERHEATED
STEAM
STEAM
4kg/cm^G
PRESS.
STEAM
TEMP.
190 °C
—i 1 1
20
IS
10
200
800 900 1000
Fig.7:
300 400 500 BOO 700
NUMBER OF SOOT BLOWING CYCLES
Endurance" test results with
soot blowing
TEST CONDITION
GAS VELOCITY
ANGLE
ASH LOADING
AVERAGE DIA.
20 m/wc
0°
SOg/Nm3
20 MICRONS
C /
—"if
B + DUMMY EROSION COMMON COMMON
RESISTANT CATALYST CATALYST
CATALYST (IMPREGNATION (KNEADING TYPE)
Fig. 8: Results of accelerated
erosion test
f .
QASTEMP. Mt°C
NNyNOa OJft
— o A LOW OUST SYSTEM
4 HIGH DUST SYSTEM
J
r- SOOT BLOWING
>
k—i—-
m
100
o
it 90
<
>
CATALYST 10 mm PITCH
DUST 6 - 10 f/Nm3
NOx 200 ~300 PPM
SO„ 200 ~300 PPM
) n n—o-oo-o—ojRj-o-oo-o-
Ul
e
•j g
TEST HOURS
Fig. 9: Test results of pilot plant
at DM-600 project
80
o.oi
0.005
0
I
NHj SLIP (PPM)
at INLET NOx (PPM)
500 1000 2000 6000 10000
OPERATING HOI IBS HR .
Fig. 10: Test results of
TTJM project
432
-------
SOOT BIOUHMQ FRQOWAM
200
*
I
| 100
5
s 0
0 1000 2000 3000 4000 5000 0000 7000 1000
TEST HOURS
Fig. 11: Results of air heater air side draft loss versus soot blow-
ing at TTJM
a I0ET
3L
0 NONC Q OMCI A DAY PROM OUTLIT
(|) OMCC A OAV FROM BOTH IIOCS
(HIGH DUIT SYSTEM)
a
15
SOOT BtOWINO PMMAM
3ST
MONK ® STIMU A DAY PROMOUTLtT
(J) 1 TIMS A DAY FROM BOTH tIDCI
< LOW OUST lYtTKM)
WATIR WASHING
3000 4000 6000
TEST HOURS
Fig. 12: Results of air heater air side draft loss versus soot
blowing at TTJM
433
-------
BOILER
INJECTION
' 0.8
A a • 0.5
140 MW 175 MW
20 40 60 >0
BOILER LOAD. %
100
Fig. 14: Initial performance test results at Shimonoseki No. 1 unit
434
-------
S!
i
/
! •>
; t
\ ;>
10.2 m
12.4 m
8.4 m
0 5 10 15 20m
1 ... .1 .... I .... 1 ¦ m 1
SCALE 1:600
Fig. 15:
General arrangement of 500 MW coal-fired boiler
with SCR system (Case I) Doner
435
-------
TOTAL AIRFLOW
NH„ GAS
CZj
nh3 dilution air
FROM STEAM AIR HEATER
DE-NO
REACTOR
CHAMBER
AIR
HEATER
BOILER
A
I I
(I M---C*
Fig* 16: P & I diagram of the SCR system
firSY
p
PRESSURE
T
TEMPERATURE '
f
Flow
A
aNALVSIS
C
tdkTAOLLEft
E
tLfe^ENT
X
GENERATOR
Y
TESTING
436
-------
100
90
x 80
70
CASE II (NH3/NOx MOL RATIO 0.92)
• — A -X
©
CASE I (NH3/NOx MOL RATIO 0.82)
1/4
Fig. 17:
2/4 3/4
BOILER LOAD
100
90
1 & II *
266
320 360
376 380
GAS TEMPERATURE. °C
7.6
6.6 4.9
4.3 4.3
O; CONCENTRATION, %
10
? J" M
1 %
-O'l
n ee 70
X X
* i
60
60
40
4/4 MCR
Predicted NOx removal
performance versus boiler
load
LOAD -MCR
OAS TEMP 380°C
O SV - 3700H"1 (Cn* I)
A SV-3000H1 ICM II)
0.5 0.6 0.7 0.8 0.9
NH3/NOx MOL RATIO
1.0
Fig. 18: Predicted NOx
efficiency vs
NHg/NOx mol ratio
437
-------
FLUE GAS INLET
REACTOR CHAMBER SHELL
HOIST RAIL
STEEL STRUCTURE
FOR MAINTENANCE
FLOW DIRECTION
OF FLUE GAS
DUMMY
LAYER
FLUE GAS
OUTLET
Fig. 19: Cutaway view of Reactor Chamber
-------
table I
SUPPLY LIST OF MHI SCR SYSTEM
Fuel
Plant She
Boiler
Capacity
Type of
Catalyst
N/R
Time of
Delivery
Kyushu Electric, Shinkokura 3
600 MW
Pellet
N
6
1978
Kyushu Electric, Shinkokura 4
600 MW
Pellet
N
6
1979
Osaka Gas Co., Senpoku Rant
20 t/h
Pellet
N
12
1976
Osaka Gas Co., Senpoku Plant
20 t/h
Pellet
N
12
1976
LNG
Aito Co., Chita Factory
30 t/h
Pellet
N
5
1978
Aito Co., Chita Factory
30 t/h
Pellet
N
5
1978
Aito Co., Chita Factory
30 t/h
Pellet
N
5
1978
Dainippon Ink Co., Sakai Factory
30 t/h
Pellet
N
2
1980
TK Plant (LNG & BFG)
395 MW
Grid
R
5
1981
Sumitomo Chemical Co., Sodegaura
370 t/h
Pellet
R
1
1976
Fuji Oil Co., Sodegaura Refinery
160 t/h
Gr
d
N
12
1977
Tokyo Electric, Yokosuka 4
350 MW
Gr
d
R
2
1978
Kansai Electric, Osaka 1
156 MW
Gr
d
R
6
1978
Chubu Electric, Chita 4
700 MW
Gr
d
R
11
1979
Kansai Electric, Osaka 3
156 MW
Gr
d
R
7
1980
Kansai Electric, Osaka 4
156 MW
Gr
d
R
12
1979
Kansai Electric, Sakaiko 1
250 MW
Gr
d
R
7
1980
Kansai Electric, Sakaiko 6
250 MW
Gr
d
R
12
1979
Chugoku Electric, Iwakuni 2
350 MW
Gr
d
R
12
1980
Oil
Chugoku Electric, Iwakuni 3
500 MW
Gr
d
N
4
1981
Chubu Electric, Atsumi 3
700 MW
Gr
d
N
12
1980
Chubu Electric, Atsumi 4
700 MW
Gr
d
N
2
1981
Chubu Electric, Shinnagoya 3
220 MW
Gr
d
R
7
1980
Kansai Electric, Osaka 2
156 MW
Gr
d
R
7
1981
Kansai Electric, Tanagawa 3
156 MW
Gr
d
R
8
1981
Kansai Electric, Tanagawa 4
156 MW
Gr
d
R
3
1981
Kansai Electric, Sakaiko 2
250 MW
Gr
d
R
9
1981
Kansai Electric, Sakaiko 4
250 MW
Gr
d
R
2
1981
Kansai Electric, Sakaiko 7
250 MW
Gr
d
R
9
1981
KK Plant
600 MW
Gr
d
R
7
1982
Chugoku Electric, Shimonoseki 1
175 MW
Gr
d
R
4
1980
Joban Electric, Nakoso 8
700 MW
Gr
d
N
12
1982
Coal
Chugoku Electric, Shinube 1
75 MW
Gr
d
R
8
1982
Chugoku Electric, Shinube 2
75 MW
Gr
d
R
7
1982
Chugoku Electric, Shinube 3
156 MW
Gr
d
R
6
1982
KM Plant
156 MW
Gr
d
R
3
1983
Remarks: N/R N: New unit
R: Retrofit
-------
TABLE II
DESIGN CONDITIONS—SHIMONOSEKI NO- 1
SCR SYSTEM
Boiler
Type Mitsubishi—C-E Controlled
Circulation® boiler-CCRR
Evaporation 540 t/h
Fuel Australian bituminous coal
and oil
Unit capacity 175 MW
NO* Removal System
Type Selective Catalytic Reduction
Gas flow 550,000 Nm3/h
Reactor Vertical down-flow fixed bed
Catjlyst Grid type —10 mm pitch
Sv nominal 3000 H"1
Inlet NOx concentration 500 ppm
NOx removal efficiency 51%
Ammonia slip (target) 5 ppm max.
Dust concentration about 20 g/Nm3
TABLE III
DESIGN CONDITION OF AN
SCR SYSTEM FOR A 500-MW
COAL-FIRED BOILER
Case I Case II
NOx value at system inlet 500 ppm 500 ppm
NOx value at system outlet 100 ppm 50 ppm
NO* removal efficiency 80% 90%
Ammonia slip less than less than
5 ppm 5 ppm
Catalyst life (target) 24 months 24 months
Mole ratio (NH3/NOx) 0.81 0.91
Gas conditions leaving the boiler
Gas flow rate 1,553,000 Nm3/H
Gas temperature 380 "C
Oxygen 4.3%
Sulfur dioxide 1,350 ppm
Sulfur trioxide 14 ppm
Particulate 15 g/Nm3
TABLE IV
SCR SYSTEM REACTOR CHAMBER SPECIFICATION FOR A 500-MW COAL-FIRED BOILER
Case I Case II
Type Fixed bed Fixed bed
Number Two "^w0
Direction of gas flow Vertical Vertical
down down
Size Width 12-4 m 12.4 m
Height 8.4 m 11.0m
Length 10-2 m 10.2 m
Catalyst volume 404 m3 480 m3
Catalyst shape Grid— Grid—
7 mm pitch 7 mm pitch
Number of layers Three Four
Superficial gas velocity 5.5 m/s 5.5 m/sec
Draft loss thru reactor less than less than
75mmH20 90mmH20
440
-------
APPLICABILITY OF THERMAL DeNOx
TO LARGE INDUSTRIAL BOILERS
By!
B. E. Hurst and C. E. Schleckser, Jr.
Exxon Research and Engineering Company
Florham Park, New Jersey 07932
441
-------
ABSTRACT
Exxon Research and Engineering Company has developed and successfully applied
a process called Thermal DeNOx for removing oxides of nitrogen (N0X) from
flue gas in stationary combustion sources. This non-catalytic process is based
on a gas phase homogeneous reaction. The technology involves injection of
ammonia (NH3) and hydrogen (H2) into the hot flue gas within a prescribed
temperature range.
Thermal DeNOx has been commercially demonstrated in gas and oil-fired
steam boilers, utility boilers and process furnaces. Successful tests have also
been conducted on a municipal incinerator and an oil field steam generator.
Tests on flue gas generated by coal combustion have demonstrated the applica-
bility of the process to coal-fired boilers.
Cost effectiveness of the process is superior to other competing flue gas
treatment processes in most applications. Also, Thermal DeNOx is not as
capital intensive as competing processes and can be applied with similar cost
and performance effectiveness for either grass roots or retrofit applications.
442
-------
APPLICABILITY OF THERMAL DeNO*
TO LARGE INDUSTRIAL BOILERS
INTRODUCTION
Exxon Research and Engineering Company has developed and successfully
applied a process (3) called Thermal DeNOx for removing oxides of nitrogen
(NOx) from flue gas in stationary combustion sources. This non-catalytic pro-
cess is based on a gas phase homogeneous reaction (2, 5). The technology
involves injection of anmonia (NH3) into the hot flue gas within a narrow and
critical temperature range. This temperature range can be significantly ex-
panded through the additional injection of hydrogen. The N0X reduction is
essentially independent of the concentration of oxides of sulfur or particulate
matter in the flue gas.
Thermal DeNOx has been commercially demonstrated in gas and oil-fired steam
boilers, utility boilers and process furnaces. Successful tests have also been
conducted on a municipal incinerator and an oil field steam generator. Tests on
flue gas generated by coal combustion have demonstrated the applicability of the
process to coal-fired utility boilers, but there is no commercial experience to
date. In full-scale retrofit commercial demonstrations conducted by Toa Nenryo
Kogyo K.K. and Tonen Seklyu Kagaku K.K. at their Kawasaki, Japan, plants, NOx
reductions exceeding 60Z were achieved (see Figure 9).
Cost effectiveness of the process is superior tp other competing flue gas
treatment processes in most applications. Also, Thermal DeNOx is not as capi-
tal intensive as competing processes and can be applied with similar cost and
performance effectiveness for either grass roots or retrofit applications.
Therefore, Thermal DeNOx offers a practical alternative In the field of NOx
pollution control.
443
-------
In general, Thermal DeNOx can effect reductions In N0X emissions of up
to 70% compared to the 90% reduction achievable in other processes. The
specific level of DeN0x is dependent upon a number of factors, including the
design or type of fired equipment, operating mode, and initial NOx level. Kftp
work is continuing on this process to improve its DeNOx performance Biid overall
process design.
Thermal DeNOx may be applied to boilers for additional NOx reduction after
combustion modifications have been made. Alternatively, in the event combus-
tion modifications result in unacceptable losses in load capacity, Thermal
DeN0x may be applied without these modifications.
PROCESS CHEMISTRY REVIEWED
The process chemistry relies on the selective reaction between NH^ and N0X
to produce nitrogen and water. This reaction proceeds in the presence of
excess oxygen within a prescribed temperature range. The overall NO^ reduction
and production reactions are summarized in equations (1) and (2), respectively.
NO + NH3 + 02 + (H2) - N2 + H20 (1)
NH3 + 02 - NO + H20 (2)
In typical flue gas environments, the NO reduction shown as equation (1)
dominates at temperatures around 1740°F (950°C). At higher temperatures, the
NO production reaction shown as equation (2) becomes significant, and above
2000°F (1090°C), the injection of NH3 is contraproductlve, causing increased
NO. As temperatures are reduced below 1600°F (850°C), the rate of both
reactions becomes extremely low, the NO reduction falls off drastically, and
the NH3 flows through unreacted.
These very rapid changes of NO reduction with temperature are shown most
clearly in Exxon's published laboratory data (4). The data shown in Figure 1
is typical of laboratory results.
Exxon's technology also includes means of altering the usable temperature
range. The addition of hydrogen (H2) extends the temperature window over a
wide temperature range as shown by the laboratory data in Figure 2. At Hj/NH^
ratios on the order of 2:1, the NOj- reduction can be forced to proceed rapidly
444
-------
at 1290°F (700°C). By judiciously selecting the H2/NH3 injection ratio, DeNOx
performance can be optimized at any intermediate temperature such that the over-
all temperature window is as shown in Figure 3. This temperature flexibility
provides load following capability on boilers and Industrial heaters.
TEMPERATURE "WINDOWS" IN INDUSTRIAL BOILERS
A typical window location for industrial boilers is usually found either
within the superheater tube bank or between the superheater tube bank and the
steam generator tube bank. Figures 4 and 5 illustrate an injection location for
a package type boiler having two synmetrical generating tube banks situated on
each side of the combustion chamber. Three injectors are located Interbank on
each side to provide complete coverage.
The Injectors for a typical field erected, cross drum design boiler are
shown in Figure 6. In this boiler the Injectors are located within a super-
heater cavity such as might be supplied for sootblowers. Support guides for
the Injectors are provided from attachments to the superheater tubes.
MAJOR FACTORS AFFECTING PERFORMANCE
The major factors affecting Thermal DeNOx performance are shown In Table 1.
In addition to temperature and the use of NH3 and H2 these include:
• Residence time at temperature
• Temperature profile
• Initial N0X
• NHj/NO^ ratio
• Mixing
Maximizing residence time at temperature tends to enhance DeN0x performance.
Typical ranges of residence time are from less than 0.1 second to greater than
1 second. In order to achieve high performance without sacrificing significant
space in the boiler, residence times of 0.2 to 0.3 seconds are preferred. The
most conmon method of providing residence time at temperature is by injecting
NH3 at the upstresm side of a cavity In the tube bank as shown in Figures 4, 5,
and 6.
445
-------
Flue gas temperature profile at the injection plane is important in Thermal
DeNOjj performance since wide variations in temperature may result in a lowering
of overall performance. Variations of +100°F (56°C) can usually be accommo-
dated without effecting performance. Larger variations may require hydrogen
injection in the lower temperature zone(s) of the injection plane in order to
maintain acceptable performance.
In order to accommodate temperature changes in the post-combustion zone
of a boiler resulting from changes in boiler load, more than one ammonia
injection grid may be required in order to maintain satisfactory DeNOx perfor-
mance. For example, as load is reduced from full to 50%, the temperature for
optimum Thermal DeN0x will shift toward the fire box. Temperature shifts can
also occur in coal fired boilers due to slagging, type of coal, changes in
excess O2 and other operational variations. In addition, the use of hydrogen
with its ability to widen the effective DeNO* temperature window increases the
capability to deal with a wide range of boiler operating conditions. Selection
of grid locations can be accomplished by temperature measurement in the case of
existing boilers or in conjunction with boiler vendor performance predictions in
the case of grass roots designs.
The process is also sensitive to initial N0X and NH3 concentrations. The
NH-j injection rate is generally expressed as a mole ratio relative to the
initial N0X concentration. Nt^/NOx ratios of 1.5 are common for initial N0X
levels of 200 vppm and less. As initial N0X concentration increases, the
ratio is reduced toward 1.0. Ammonia breakthrough levels are generally below
50 vppm.
NEW MIXING TECHNOLOGY BOOSTS PERFORMANCE
Recent innovations in mixing techniques have led to substantially in-
creased performance capability with the Thermal DeN0x process. This is es-
pecially true in short residence time situations normally encountered in utility
and industrial boiler applications.
For comparison purposes, three NH3 injection/mixing techniques are shown
on Figure 7. You will note that substantial improvements in performance have
been achieved through development of an Improved NH3 injection technique. These
data have been taken in our 30 MBtu/hr (8.8 MW) pilot plant test furnace.
446
-------
In addition, pilot plant testing has shown that staged injection of
hydrogen utilizing two grids in succession results in further increases in
performance. These technological advances will result in performance pre-
dictions in the range of 70% or more for grass roots utility boiler applica-
tions and 60Z to 702 for retrofit applications.
The impact of these advances in technology are illustrated in Figure 8.
This figure shows performance predictions for a retrofit utility boiler
application for three different conditions:
• single grid, no H2
• single grid, zoned H2
• two grids with staged H2
A maximum performance improvement at 100% load of 24 DeNOx percentage points
(from 41% to 655!) is possible here through utilization of the new technology.
PROVEN PERFORMANCE
Thermal DeNOx has been successfully demonstrated in gas and oil-fired
utility boilers, package and field erected industrial boilers, process furnaces
and oil field steamers as summarized in Table 2. Actual performance often
represents a compromise between the technical limits of the process chemistry
and cost effectiveness. In many situations, performance is maximized at full
load operation, and smaller NOjj reductions accepted at reduced loads with the
lower reaction zone temperatures. In such installations, total NQx (missions
are generally at target levels over the full spectrum of operating conditions
because of the reduced N0X production rate at lower loads. Results from seven
demonstrations are shown over their range of operating conditions as a
function of flue gas temperature In Figure 9. Please note that these units
Incorporate earlier mixing technology and thus performance shown is generally
lower than would be achieved with updated technology.
Alternative designs are available to suit the load demands of the parti-
cular boiler or furnace. Of the units Installed to. date, some have been
designed for base load operation, while others accommodate all operating
conditions from 50-100% of design capacity. As another example, one boiler in
which Thermal DeNOx was installed operates between 35 and 50% of design, and
the process was designed for these load conditions.
447
-------
Applicability and effectiveness will vary from one unit to the next de-
pending on local flue gas conditions and on the configuration of the high
temperature zones. Preliminary estimates of performance expected to be
achieved for retrofitting Thermal DeNOx into existing units can be made on the
basis of engineering drawings and design specifications. Exxon can also pro-
vide recommendations for designing new boilers and furnaces and for modifying
existing designs so as to maximize the efficiency of the process.
Most of the development of the Thermal DeN0x process has taken place on
oil-fired sources, essentially free of fly ash. Fly ash from coal firing may
deposit on Injector grids, change gas flow patterns and temperature profiles
and foul or erode injection nozzles. The extent to which reliable performance
can be provided under these conditions remains to be proven. We do not feel
significant or insuperable problems will arise. However, we would propose a
demonstration test be performed to evaluate these factors on an existing
boiler. Details of this demonstration test can be provided upon request.
WHAT ABOUT REACTIONS WITH SULFUR?
Detailed laboratory experiments have shown there are no reactions between
the Thermal DeNOx process and sulfur compounds in the high temperature flue gas
regions. That is, sulfur or its oxides do not interfere with the NH3-N0x-02-H2
chemistry. Additionally, ammonia injection has been shown to cause neither
additional homogenous nor additional heterogeneous oxidation of S02 to SO3.
To the extent that the thermal reduction of N0X leaves some NH^ unreacted,
and as the combustion gases cool, NH^ reacts with SO3 and H2O to form ammonium
sulfate ((NH^)2SO4) and/or ammonium bisulfate (NH4HSO4). Ammonium sulfate is a
dry solid which forms directly from the gaseous reactants without passing through
the liquid atate. It ia not corrosive or appreciably hygroscopic. When the
sulfate is formed in flue gas the resulting particles are on the order of 1-3
microns. Thus the sulfate should pass through the air preheater with very
little deposition. When heated, ammonium sulfate decomposes to ammonium
bisulfate and gaseous asmonia.
Ammonium bisulfate is a sticky liquid at air preheater temperatures. The
melting point of pure ammonium bisulfate is 297°F (147°C), which is below the
temperature of the flue gas leaving most air preheaters, but is above the metal
448
-------
temperature. However, the presence of HjSO^ or small amounts of ammonium sulfate
depress the melting point to about 266°P (130°C). So ammonium bisulfate de-
posits should be liquid throughout most of the air preheater, with solid depo-
sits only at the extreme cold end. The sticky liquid is corrosive and tends
to trap ash and soot particles, accelerating the rate of deposit build-up.
Ammonium bisulfate formation can be minimized by either of three principal
means:
• Limiting NH3 breakthrough to 5 vppm or less.
• Maintaining flue gas temperature at the preheater outlet above 400°F (204°C).
• Maintaining an Nl^tSO^ molar ratio above 2.0 and providing sufficient
residence time for the reactants to form ammonium sulfate rather than
bisulfate.
In most cases the last method is the most practical for application with the
Thermal DeN0x process, since typical NH3 breakthrough levels (approximately
50 vppm) are usually twice that of SO3. Also, maintaining a high flue gas
outlet temperature is impractical since this represents a 2-3% loss in
thermal efficiency for many boilers. A summary of the fouling/corrosion
potential of ammonium sulfates as a function of NH3 and SO3 concentrations
in flue gas is presented in Figure 10.
Based on laboratory and commercial tests with oil firing, these sulfates
when in combination in ash deposits do not create either severe corrosion or
unacceptable air preheater fouling problems when Thermal DeN0x is used in
accordance with its design specifications. However, we would expect both
fouling and corrosion to increase if bisulfate formation is not curtailed. In
addition, long term tests conducted lxi two oil-fired boilers by Tonen Sekiyu
Kagaku K.K. in Kawasaki, Japan, revealed sulfate/bisulfate deposits could
easily be removed by watervashing the air preheaters at reasonable Intervals.
ENGINEERING CONSIDERATIONS
Although the chemistry is straightforward, certain difficulties must be
overcome when applying the process to commercial equipment. Performance is
generally limited by the access to the required flue gas temperature range of
the reaction, and the dependence of the reaction on the local concentrations
449
-------
of reactants, NH^, N0X, 02* and H2. Exxon's technology provides a means of
adapting the chemistry requirements to industrial equipment environments, and
significant NOjj reductions can be achieved by the use of Thermal DeN0x tech-
nology in existing boilers. Application to new, grass-roots designs is usually
easier because the internal configuration of the cavity at the required tempera-
ture can be readily adjusted to complement the process demands.
Accommodating flue gas temperature variations is important if high DeNOg
rates are to be achieved. Not only does the system have to accommodate flue
gas temperature changes caused by normal load and operating variations, but it
also must allow for fluctuations across the reaction zone caused by non-
uniformities in flow and heat transfer. It follows, therefore, that a case-by-
case evaluation of flue gas temperatures and local conditions is required for
the application of Thermal DeNOx for each installation considered.
Initially, ammonia was injected only into boiler cavities, boiler regions
between tube banks, which can be considered to be isothermal to a first approxi-
mation. Subsequent experimentation by Exxon Research has shown the feasibility
of injecting ammonia into boiler tube bank regions as well. Thus, satisfactory
NO reduction performance can be obtained by locating the injector grid in either
the boiler convection pass cavities or tube bank. The ability to inject ammonia
at virtually any post-combustion boiler location where temperatures range
from 1292 to 1994°F (700 to 1090°C) has substantially increased the flexibility
of the Exxon Thermal DeNOx Process.
SUPERIOR COST EFFECTIVENESS
The Thermal DeN0x process offers one of the most practical approaches to
significant N0X reduction from stationary fired equipment of any post-combustion
N0X removal process currently on the market. This is due to the fact that the
process is far less capital intensive and achieves superior cost effectiveness
even though NOx reduction levels are somewhat lower than other processes.
In order to demonstrate cost effectiveness of the process, a 200,000 lb/hr
(91 t/hr) industrial boiler has been chosen for illustration purposes. A
Thermal DeNOx equipment sizing basis is presented in Table 3 and a simplified
flow diagram is shown on Figure 11. It is assumed that the boiler is oil or gas
450
-------
fired and has an uncontrolled N0X level of 200 vppm corrected to 32 0£ dry. The
boiler is to be equipped with a single grid capable of Injecting both NH3 and H2
via a 8team carrier. A one month NH^ storage capacity has been provided, and H2
is supplied through an ammonia dissociator.
The total erected capital investment for these facilities is estimated to
be $472,000 as outlined in Table 4. Direct costs include the material and labor
for equipment outlined in Table 3 plus interconnecting piping. Indirect costs
Include field labor overheads, construction superivision and equipment, labor
wage taxes, erection fee, engineering costs, and licensing fee. Contingency
is based on 30% of direct, plus indirect cost less licensing fee. Costs are
expressed at 1Q81 southern California level. Excluded from these costs are
such items as land and owners charges.
Cost effectiveness for the sample boiler installation is also shown in
Table 4 for three different bases. Annualized costs are obtained by assuming
a five year payout on investment plus annual operating costs. On a heat fired
basis, cost effectiveness of the process is $0.13 per M Btu (.04 mill/kWh).
The cost effectiveness based on N0X removed is $1.51 k/US ton NO2 ($1.66 k/t N02)
for a DeNOx efficiency of 60% and $1.29 k/US ton NO2 ($1.42 k/t NO2) *or ?0Z
DeNO^. This cost compares favorably with some combustion modification techniques
such as burners out of service and flue gas recirculation, and is substantially
lower than selective catalytic reduction.
Annual operating costs are shown in Table 5. These costs are based on a
65Z annual load factor and include values for NH3, power and steam consumption
and maintenance material and labor.
SUMMARY
In summary, Thermal DeNOx represents one of the most practical and cost
effective means of flue gas treatment for NO* control currently available in the
market place. As the requirement for control of NOjj emissions becomes more
stringent and more widespread, application of N0X control technology will become
necessary in many areas of the U.S.A. Recent technological advances have
resulted in significant increases in Thermal DeNOx performance with negligible
increase In investment thus further improving its cost effectiveness. No
451
-------
technological difficulties are foreseen in applying Thermal DeNO* to all types
of industrial boilers including coal fired, but a test in a coal fired boiler
is needed to demonstrate the applicability of the technology and optimize
engineering design considerations.
452
-------
REFERENCES
BASIC CHEMISTRY
1. U.S. Patent 3,900,554.
2. Lyon, R.K., "Communication to the Editor: the NH3-NO-O Reaction,"
International Journal of Chemical Kinetics, Vol. VIII, 1976, pp. 315-318.
GENERAL THERMAL DeNOx
3. Jones, Stacy V., "Nev Process Cuts Air Pollution," The New York Times,
Aug. 16t 1975, Financial Page.
4. Lyon, R.K. and J. P. Longwell, "Selective, Non-Catalytic Reduction of N0X
by NH3," EPRI NOx Seminar, San Francisco, Feb. 1976.
5. Muzio, L.J., et al., "Gas Phase Decomposition of Nitric Oxide in Combustion
Products," EPRI N0X Seminar, San Francisco, Feb. 1976.
6. "Exxon Thermal DeNOx System Successfully Commercialized," The Oil Daily,
Oct. 5, 1976.
7. "1976 I-R 100 Award Vinners," Industrial Research, Vol. XVIII, October, 1976.
8. Lisk, Ian 0., "NO^ Decomposition Technique," Pollution Engineering. March,
1977, p. 8.
9. Practical Available Technology Report, "A Way to Lower NO* in Utility Boilers,"
Environmental Science and Technology, Vol. XI, No. 3., March 1977, pp. 226-
228.
10. Ricci, Larry J., "Nixing NOx Emissions," Chemical Engineering. April 11,
1977, pp. 84-90.
453
-------
11. Bartok, W., "Non-Catalytic Reduction of N0X with Ammonia Injection for Coal
Fired Utility Boilers," Second EPA Symposium on Stationary Source Combustion,
New Orleans, Aug. 29 - September 1, 1977.
12. Tenner, A.R., "Ammonia Injection for Utility Boiler NOx Control," N0X Control
Technology Workshop, Southern California Edison Company, Pacific Grove,
Caifornia, October 25-28, 1977.
13. Varga, G.M., M. E. Tomsho, B. H. Ruterbories, G. J. Smith, and W. Bartok,
"Applicability of the Thermal DeNOx Process to Coal-Fired Utility Boilers,"
EPA-600/7-79-079, March, 1979.
14. Bartok, W., and G. M. Varga, "Applicability of the Thermal DeNO* Process to
Coal-Fired Utility Boilers," American Flame Research Committee International
Symposium on N0X Reduction in Industrial Boilers, Heaters and Furnaces,
Houston, October, 1979.
454
-------
FIGURE 1
TEMPERATURE "WINDOW" FOR THERMAL DENOx
REACTION-NH3 ONLY
250
NH3/N0i =
O2 = 2.0%
RESIDENCE TIME
(@ 1900°F) = 0.1 sec
1.7
200
150
100
50
650 700 750 800 850 900 950 1000 1050
AVERAGE FLUE GAS TEMPERATURE, °C
-------
FIGURE 2
TEMPERATURE "WINDOW" FOR THERMAL DENOx
REACTION-H2+NH3
250
RESIDENCE TIME
(@ 1900°F) = 0.1 sec
200
Q.
Q_
>
N>
150
X
o
z
t-
z
Hi
ZD
_l
Li.
Ll-
O
O
100
UJ
to
50
650
700
750
850
800
900
950
1000
1050
AVERAGE FLUE GAS TEMPERATURE, °C
-------
FIGURE 3
TEMPERATURE "WINDOW" FOR THERMAL DENOx
REACTION
250
200
RESIDENCE TIME
( @ 1900° F ) = 0.1 sec
a.
X 150
o
z
Lii
100
u.
Ul
50
1050
950 1000
850
900
750
800
700
650
AVERAGE FLUE GAS TEMPERATURE, °C
-------
FIGURE 4
TYPICAL PACKAGE BOILER THERMAL DENOx
INJECTOR DESIGN
(Rear View)
Steam Drum
-t*
tn
00
Vertical
Injector
(TYP)
Horizontal
Injector
(TYP)
Flue Gas
Passage
-------
FIGURE 5
TYPICAL PACKAGE BOILER THERMAL DENOx
INJECTOR DESIGN
(Plan View)
Horizontal Injector
Tube Bank
Furnace
Vertical Injector
-ocooooooq
xv
—OCD-
o o
o o
—coo
-CCD
a>
-ooo
o o
o o
—ooo-
x>
Horizontal
Vertical
Injector Injector
-------
FIGURE 6
TYPICAL CROSS DRUM BOILER THERMAL DENOx
INJECTOR DESIGN
STEAM DRUM
INJECTOR
(TYP)
-------
FIGURE 7
DENOx PERFORMANCE IN PILOT PLANT TESTS
USING VARIOUS MIXING TECHNIQUES
100
80 -
1750
IMPROVED MIXING
CROSSFLOW OR
COUNTERFLOW
CO-FLOW
NOj 200 PPM
NH3/N0j 1.5
H2/NH3
%0o 3
0.0
RESIDENCE TIME 0.066 SEC.
TEMPERATURE PROFILE
UNIFORM
1800 1850
TEMPERATURE, 8F
1900
-------
FIGURE 8
UTILITY BOILER PERFORMANCE PREDICTIONS
EFFECTS OF IMPROVED MIXING AND HYDROGEN
ADDITION
X
o
o
OS
65
60 -
50
40
100% LOAD
51
41
A
59
49
65
55
IMPROVED MIXING
NOZZLE ARRANGEMENT
CONVENTIONAL
NOZZLE ARRANGEMENT
±
1 GRID
ZONED HYDROGEN
I
1 GRID
NO HYDROGEN
2 GRIDS WITH
STAGED HYDROGEN
50
40
30
20
80% LOAD
46
24 1 GRID
0 ZONED HYDROGEN
J I
48
IMPROVED MIXING OR
CONVENTIONAL
NOZZLE ARRANGEMENT
_L
1 GRID
NO HYDROGEN
2 GRIDS WITH
STAGED HYDROGEN
-------
FIGURE 9
PERFORMANCE OF THERMAL DENOx
SYSTEMS IN COMMERCIAL APPLICATIONS
CXI
OJ
70
60
50
O 40
H
O
=>
o
¦ 11
u. 30
x
o
20
10
'00
¦ 25 t/HR
• 70 t/HR
o 120 t/HR
*100 MW
~ 100 MW
~ 150 kbbl/d
*150 kbbl/d
NOTES:
DATA REPRESENTATIVE OF
INITIAL APPLICATIONS BASED ON
EARLIER MIXING TECHNOLOGY
• PERFORMANCE ON ANY SPECIFIC
UNIT IS FUNCTION OF DESIGN AND
OPERATING PARAMETERS
DESCRIPTION
PACKAGE BOILER
INDUSTRIAL BOILER
UTILITY BOILER
CRUDE HEATERS
800 900 1000
FLUE GAS TEMPERATURE, °C
1100
1200
-------
FIGURE 10
FOULING POTENTIAL OF AMMONIUM SULFATES
MODERATE FOULING/
CORROSION POTENTIAL
HIGH FOULING/
CORROSION POTENTIAL
M^^WI
LOW FOULING/
CORROSION POTENTIAL
S03 CONCENTRATION, VPPM
-------
FIGURE 11
SIMPLIFIED THERMAL DENOx SUPPLY SYSTEM
FLOW DIAGRAM
FIRED EQUIPMENT
STEAM
OR AIR
INJECTION GRID
HYDROGEN
SUPPLY
AMMONIA
DISSOCIATOR
CARRIER
SUPPLY
AMMONIA
EVAPORATOR
LIQUID
AMMONIA
STORAGE
-------
TABLE I
MAJOR FACTORS IN THERMAL DENOx
PERFORMANCE
• FLUE GAS TEMPERATURE -
+ WITH OR WITHOUT H?
• RESIDENCE TIME AT TEMPERATURE
• TEMPERATURE PROFILE
• INITIAL N0X
• nh3/nox RATIO
• MIXING
-------
TABLE II
THERMAL DENOx EXPERIENCE SUMMARY
(OIL AND GAS FIRED UNITS)
JAPANESE INDUSTRIAL BOILERS
CONSTRUCTION DESIGN
OPERATIONAL PHASE PHASE
JAPANESE UTILITY BOILERS
JAPANESE PETROLEUM HEATERS
5
4
^ CALIFORNIA OIL FIELD STEAMER
^ (DEMONSTRATION)
CALIFORNIA FLAT GLASS MELTING
FURNACE (DEMONSTRATION)
CALIFORNIA PETROLEUM HEATERS
19
CALIFORNIA
CALIFORNIA
CALIFORNIA
UTILITY BOILER
INDUSTRIAL BOILERS
INCINERATORS
1
3
2
-------
TABLE III
EQUIPMENT SIZING BASIS FOR PERMANENT
THERMAL DENOx FACILITIES FOR SAMPLE
INDUSTRIAL BOILER
BOILER DESIGN CONDITIONS
FUEL
INITIAL N0X
nh3/mox
h2/nh3
CARRIER REQUIREMENT
nh3 STORAGE TANK
NH3 VAPORIZER
nh3 DISSOCIATOR
INJECTION GRID
INSTRUMENTATION
200,000 LB/HR, 615 PSIG, 700 F
OIL OR GAS
200 PPM
1.5
0.5
STEAM, 1500 LB/HR, 15 PSIG
(MINIMUM)
30 DAY STORAGE CAPACITY
(8000 GALLONS)
(1) ELECTRIC ELEMENT,
DIRECT CONTACT
(1)750 SCFH
SINGLE LOCATION
AS REQUIRED
-------
TABLE IV
THERMAL DENOx INVESTMENT COST AND
COST EFFECTIVENESS FOR SAMPLE
INDUSTRIAL BOILER
CAPITAL INVESTMENT^
MATERIAL AND LABOR^* $168 K
INDIRECT C0STS(3) $215 K
CONTINGENCY^ $ 89 K
£ 5472 K
COST EFFECTIVENESS
$/M BTU FIRED 0.13
K$/T0N N0X REMOVED 1.51
@ 60% DeN0x
K$/T0N N0X REMOVED 1.29
@ 70% DeNOx
NOTES: (1) INVESTMENT COSTS ARE EXPRESSED AT IQ81 SOUTHERN CALIFORNIA LEVEL
(2) INCLUDES EQUIPMENT ITEMS LISTED IN TABLE III PLUS INTERCONNECTING PIPING
(3) INCLUDES FIELD LABOR OVERHEADS, CONSTRUCTION SUPERVISION AND EQUIPMENT, LABOR WAGE
TAXES, ERECTION FEE, CONTRACTOR ENGINEERING, EXXON RESEARCH AND ENGINEERING CO.
CHARGES, AND LICENSING FEE. EXCLUDED ARE SUCH ITEMS AS LAND AND OWNERS CHARGES.
(4) CONTINGENCY IS BASED ON 30% OF DIRECT COSTS PLUS INDIRECT COSTS MINUS LICENSING
FEE.
-------
TABLE V
THERMAL DENOx ANNUAL OPERATING COST
FOR SAMPLE INDUSTRIAL BOILER
AMMONIA
(2)
ITEM
(3)
ELECTRIC POWER
STEAM(4)
MAINTENANCE MATERIAL AND LABOR^
ANNUAL CONSUMPTION
165 TONS
136 MWH
4290 TONS
(D
NOTES:
(1) ASSUMES 65% LOAD FACTOR
(2) INCLUDES AMMONIA FOR DIRECT INJECTION PLUS HYDROGEN PRODUCTION
(3) INCLUDES POWER REQUIREMENT FOR AMMONIA VAPORIZER AND DISSOCIATOR
(4) LOW PRESSURE STEAM (15 PSIG MINIMUM) FOR CARRIER
(5) ASSUMED TO BE 7% OF DIRECT INVESTMENT COST
UNIT GAS
ANNUAL COST
$170/T0N
$27,965
$50/MWH
6,800
$12/TON
51,480
11,760
TOTAL
$98,005
-------
UTILITY BOILER ENVIRONMENTAL ASSESSMENT — THE EPRI APPROACH
By:
Monta W. Zengerle
Electric Power Research Institute
P.O. Box 10412
Palo Alto, California 94303
471
-------
ABSTRACT
EPRI's environmental assessment program for air emissions comprises
physical and chemical characterization, ecological and human health research
and an integrated analysis of costs, benefits and risks associated with vari-
ous generating technology and emission control approaches. Physical and
chemical characterization is approached from both a regional and localized
aspect and includes primary and secondary pollutants. Regional transport
research began with the Sulfate Regional Experiment (SURE) in the Northeastern
U.S. and continues with visibility and acid deposition research in the East
and West.
Localized plume distribution is being studied on a site-specific basis
beginning with simple and continuing with more complex terrain. Both efforts
include extensive field measurement programs designed to evaluate or develop
modeling techniques for predicting utility contributions to ground-level
concentrations or deposition.
Ecological research currently emphasizes the potential effects of acid
deposition and includes watershed, aquatic, forest, crop, and grassland
research. Current studies focus on biogeochemlcal processes which influence
resultant soil and water acidity and nutrient balance.
Health effects research concentrates on determining human health effects
of airborne utility emissions using animal studies, human clinical studies and
epidemiology.
The ultimate objective of these research efforts Is the evaluation of
relative risk of generation mixes and emission control strategies.
472
-------
UTILITY BOILER ENVIRONMENTAL ASSESSMENT - THE EPRI APPROACH
INTRODUCTION
The purpose of my presentation is to describe NOx research results,
research in progress and research planned by the Environmental Assessment
Department of EPRI.
The Environmental Assessment Department is part of the Energy Analysis
and Environment Division, one of six technical divisions at EPRI*
The primary objectives of the EAD are to identify, measure and determine
the relative importance of emissions from electricity generation transmission
and use* The Department comprises four program areas: Environmental Physics
and Chemistry, Ecological Studies, Environmental and Occupational Health, and
Environmental Risk and Issues Analysis. The research I will discuss today is
planned and executed under the auspices of one or more of these programs. It
is apparent that we are organized along disciplinary lines, and you will see
from my discussion that research projects tend to fall within one of these
four categories.
My objectives in this presentation are:
1) to illustrate through examples of research results, research in progress
and research planned what we see as the most important areas of study with
reference to N0X,
2) to open our research program to investigators involved in similar or
allied pursuits in hopes of encouraging an exchange among our staffs, and
3) to invite your comments on the directions we have established and the
questions we are asking through research.
473
-------
My presentation is divided into six major areas: N0x emissions, trans-
formation and transport, human health effects, ecological effects, materials
damage and visibility degradation (see Figure 1). Discussion under each of
these headings will be organized, with minor variations, as follows: results
to date, research in progress, and planned research.
N0X EMISSIONS
Results to Date
Our first NO^ emissions project began in 1975 prompted by published
reports of elevated ozone concentrations in power plant plumes. The objec-
tives of this initial research were to determine whether ozone was generated
in power plant plumes and to develop a qualitative understanding of dominant
influences on the potential for such ozone generation. Investigative method-
ology included reactive plume modeling and measurements of plumes at four
southwestern and western power plants representing gas- and coal-fired boilers
in dry or humid climates (1,2,3).
Subsequently, interest turned to the potential of N0x to contribute to
the formation of NO2 and nitrates on a regional scale. This regional research
will be discussed under transformation and transport.
The next area of interest was the formation of nitrates and nitric acid
as potential contributors to visibility degradation and acidic precipita-
tion. A modeling approach was obviously required to evaluate the impact of an
individual or group of power plants on air quality at some distance from the
source. In order to accurately model NO^ reactions, we needed to develop an
understanding of key reactions and reaction rates. Chamber research was begun
in 1978 in this pursuit and continues to date.
Plume measurements have shown ozone formation in the West and Southwest
under the conditions studied to be a rare occurrence. The chamber studies
indicated that oxidation of N0x to HNO^ and NO^ in gas phase reactions depends
on the presence of hydrocarbons.
474
-------
On-going Research
Current research related to N0X emissions Includes validation of the
chamber studies by aircraft sampling in a power plant plume and a more general
plume model validation project (A). In this project, oxides of nitrogen are
of major importance. Ground level pollutant characteristics predicted by
plume models are being investigated in addition to chemical and physical
transformations which will require validation of reactive plume models* We
plan eventually to Include a moderately complex and a complex site in addition
to these current studies at a flat terrain site.
TRANSFORMATION fir TRANSPORT
Results to Date
As mentioned above, EPRI's attention turned early to regional considera-
tions with the Sulfate Regional Experiment (SURE). I have nothing to report
now on results because final reports are in preparation. However, I will
summarize the objectives and approach of the SURE (5).
SURE began in 1977 with the primary objective of defining the relation-
ship between emitted primary pollutants (e.g., SO^) and regional, ambient
concentrations of secondary pollutants (e.g., sulfates). Subobjectives
included:
o the establishment of a regional air quality data base through mea-
surements made at ground level and aloft, and
o the determination of the location and magnitude of emissions during
air quality measurement periods.
It was hoped that with this approach it would be possible to derive
quantitative methodology for relating emissions from power plants to regional
air quality as measured by S0£ and particulate sulfate. In addition to mea-
suring S02 and sulfates, nine ground-based stations measured continuously for
nineteen months NO/NOx, 0^, temperature, dewpoint and suspended particles.
Aircraft which operated six days during the central month during each season
(seven "intensives" during the nineteen month field program) measured S02» 0^,
475
-------
NO/NOx» bgCat and condensation nuclei. SURE field sampling is now complete
and the final report Is being written.
In early 1978, the SURE program was supplemented with a precipitation
chemistry component consisting of nine stations sited in close proximity to
the nine SURE continuous monitors. The primary objective of this research was
to relate rain pH to air quality or emission parameters already included in
the SURE.
Also in 1978, outdoor air quality studies were supplemented with charac-
terization of the indoor environment as a function of outdoor pollution lev-
els, structure type, geographic locality and other variables including pres-
ence of smokers and type of heating and cooking fuel (6).
Planned Research
We are currently in a major planning exercise regarding regional air
quality studies* This process consists of a "Delphi-type" survey of industry
representatives, academicians and agency personnel which asks for a ranking of
research areas, forcing functions and research aspects. SO^ and NO^ rank high
in importance and have been combined for planning purposes. The primary
forcing functions or motivating factors for research are the potential for
oxides of nitrogen to contribute to human health, ecological or esthetic
effects and materials damage. The most important research aspects are chemi-
cal transformation, regional transport, wet and dry deposition, synthesizing
models and atmospheric dilution.
A workshop was held in April 1978 to review effects of trace nitrogen
compounds on human health and welfare, identify mechanisms of nitrogen com-
pound formation and transport, assess the present state of knowledge of these
phenomena and recommend and prioritize critical areas of research (7). Parti-
cipants agreed research was needed in all aspects and most Importantly in
removal (from atmosphere) mechanisms, transformation in plumes and ammonia
flux.
476
-------
HUMAN HEALTH EFFECTS
Identified Research Needs
A planning study accomplished In 1978 (8) evaluated existing Information
on the health effects of nitrogen oxides, Identified the need for better
methodology and the existence of gaps In knowledge and recommended a research
program appropriate for EFRI and responsive to the need for Information^ At
the top of the list of needed research was the development of more precise and
accurate methods for monitoring NC^/NOj, peroxacyl nitrates (PAN) and N-nitro-
samines. Additional research needs include:
o characterization of NOj/NO^ In respirable suspended particles,
o animal Inhalation toxicology of
- NO2/NO3 aerosols
- N-nitro8amines
- PAN
o human clinical studies for threshold limit values and respiratory
function response for
- PAN
- NO2/NO3 aerosols, and
o human epidemiology for
- no/no2
- no2/no5
- PAN
- N-nitrosamlnes.
Results to Date
In response to the need for better monitoring methodology, research was
begun to determine the rates and amounts of artifact nitrate formation on
standard fiberglass filter media; to devise, if possible, correction factors
which might be applied to historical nitrate concentrations; to test other
media; and, to recommend best available sampling and analysis techniques.
The research clearly demonstrated large and erratic artifact formation on
untreated glass fiber filters (9). Anywhere from 10 to 100 percent of the
477
-------
reported nitrate was attributed to on-filter reactions, with a most likely
value of about 20 percent. Unfortunately, it is not likely that historical
records can be corrected because although artifact formation varied according
to experimental conditions, it also varied strongly and in an unpredictable
way on the filter handling procedures used prior to exposure.
Most health-related NO^ research is still underway, so results are lim-
ited but do not Indicate that exposure to ambient outdoor levels of NO2 and
nitrate is harmful to health. There is some evidence that elevated N(>2
indoors may be linked to increased respiratory disease in children.
On-going Research
Current research in animal Inhalation toxicology involves an increased
range and duration of exposure concentrations and interactions of N02 with
SC^t O3 and SO2 and O3 together.
Planned Research
The five-year plan includes animal inhalation toxicology studies of
chronic and acute exposures to mixed gases, human clinical studies of acute
effects of NO2 alone and epidemiological studies of chronic effects of indoor
and outdoor exposures to ambient concentrations.
ECOLOGICAL STUDIES
Results to Date
EPRI's ecological research on effects of oxides of nitrogen focuses
primarily on the contribution of nitrate througih dry and wet deposition. The
majority of research underway in this area has begun recently. One study from
which there are Interim results is the Integrated Lake/Watershed Study (ILWAS)
which is being conducted at three Adirondack (New York) Lakes. In this pro-
ject, water chemistry is determined from the incoming rainfall above the
canopy and as the water passes through the watershed/lake system.
Rainfall chemistry analysis has shown:
o NO 3 to vary less than sulfate over the year,
478
-------
o NO^ concentration is about one-fourth that of SO^ in the summer and
about equal In winter, and
o a relatively low SO^/NOg ratio for snow when compared to rain«
On-going Studies
The field program for the Integrated Lake/Watershed study is scheduled to
end in December 1981 and will be followed by a year of data analysis and model
development and refinement* An interim report will be published In the fall
of 1980.
Other on-going research falls under four categories: forests and surface
waters, agricultural crops, aquatic biota and economic assessment* Current
research In forests is focused on a microcosm approach to identifying key
parameters in forests sensitive to acidic precipitation.
Research on effects on agricultural crops and aquatic biota has only
recently begun. Crop research consists primarily of experimental field stud-
ies with some use of laboratory microcosms* In aquatic studies, research will
focus on the effects of acidification and potentially resultant toxic metal
concentrations on decomposition processes*
Planned Research
Future research In the lake/watershed approach will focus on application
and validation (and additional development, If necessary) of the ILWAS model
using watersheds located in other geographic locations and exhibiting other
geological, climatological and blotic regimes*
In forest studies, it is hoped that expansion of microcosm research to
field studies will Increase our understanding of acidification processes and
forest ecosystem responses*
Interest In grasslands is fostered primarily by concern about possible
effects of sulfur deposition, primarily In the form of SOj. Although NOj is
not considered a regional problem, the deposition of secondary products,
nitrates, must be considered in the overall investigation of the Impact of
power plant emissions* The approach will be to look at nutrient cycling,
yield and productivity*
479
-------
EPRI's crop effects research will focus on acid rain rather than gaseous
pollutants as has most previous research* Emphasis will be on ecological
processes leading to changes In fertility, yield and quality of agricultural
soils and plant species. The ultimate objective Is to develop mitigation and
management methodology where the need Is shown by research.
For aquatic systems, the on-going research on effects of acidification of
decomposition mentioned above will be expanded to Include effects on other
processes hopefully leading to predictive capability and results vhlch will
refine some aspects o£ the lake/watershed modeling. Studies of effects on
specific species or communities will be carefully chosen to complement
research being conducted by EPA, universities and federal and state resource
management agencies. Mltigatlve strategies (e.g., liming and fertilization)
will be investigated.
MATERIALS DAMAGE
The approach to date has been to refine damage estimates for various
surfaces and their distribution throughout the U.S. The primary pollutant of
Interest has been SO2 rather than N0X» Urban Boston was used as a prototypi-
cal site. Results should be published within the next year. Currently, no
additional work is underway. In the future, such estimates would be used in
overall risk/cost/benefit analyses of alternative generation and emission
control technologies.
VISIBILITY
On-going Research
EPRI's visibility research has taken three fundamental approaches:
o development and intercomparlson of quantitative instrumental methods
for measuring visibility
o development of regional data bases on visibility variability, causes
of visibility impairment, and the development of reliable regional
visibility models, and
o the assessment of social and economic costs generated by visibility
impairment and the remedial costs at various levels of protection.
480
-------
EPRI's first step was to establish one visibility station In the South-
vest and one in the Northeast to serve the first purpose mentioned above,
i.e., methodology lntercomparlson and, through fine particle measurements,
begin building a base of information for understanding the relationship
betveen ambient air quality and visibility*
Following these initial tests, additional measurement stations were
established (two in the Northeast and five In the Southwest) to acquire accu-
rate and reliable information on the regional distribution and variability of
visibility impairment.
Concurrently, EPRI has funded the development and testing of an automated
multi-wavelength telephotometer which is now in the field testing stage (10).
The third fundamental approach mentioned above was the development of
social and economic benefit/cost comparisons. Among the Important objectives
are:
o evaluation and development of methodology for estimating air pollu-
tion damage
o determination of the degree to which anthropogenic sources, particu-
larly power plants, are associated with visibility impairment
o estimation of costs associated with reducing power plant emissions,
and
o identifying and quantifying economic and social benefits from
improved visibility.
Planned Research
EPRI plans to expand the Western visibility network during 1981. Data
from these regional networks will provide data initially for setting baselines
and for empirical studies of the causes of visibility impairment. Eventually,
such data will be used In the development of regional models which could be
used to examine the potential for specific sources to Impact visibility in
defined geographic areas*
481
-------
Social and economic assessments will hopefully lead to a better under-
standing of source relationships, social costs and benefits of visibility
impairment and Improvement and the Impact of emission controls on visibility
in specific geographic areas.
SUMMARY
I have discussed the EPRI Environmental Assessment Department's program
of N0x research. This program includes six major areas: NO^ emissions,
transformation and transport, human health effects, ecological effects, mater-
ials damage and visibility degradation. My primary objective was to open
channels of exchange between our staff and contractors and other researchers
doing similar or complementary research. We believe it is very important to
coordinate such research activities to best utilize limited resources whatever
their source. We invite your comments and contacts and hope that you might
share with us your results, current research and plans for the future.
482
-------
REFERENCES
1. Ogren, J. A., D. L. Blumenthal, W. H. White* T. W. Tesche, M. A. Locke
and M. K. Liu. Determination of the Feasibility of Ozone Formation in
Power Plant Plumes. EPRI EA-307, Electric Power Research Institute, Palo
Alto, California, 1976.
2. Ogren, J. A., D. L. Blumenthal and A. H. Vanderpol. Oxidant Measurements
in Western Power Plant Plumes. EPRI EA-421, Electric Power Research
Institute, Palo Alto, California, 1977.
3. Davis, Douglas D. OH Radical Measurements: Impact on Power Plant Plume
Chemistry. EPRI EA-465, Electric Power Research Institute, Palo Alto,
California, 1977*
4. Hllst, Glenn R. Plume Model Validation. EPRI EA-917-SY, Electric Power
Research Institute, Palo Alto, California, 1978.
5. Mueller, P. K. and 6. M. Hidy. Implementation and Coordination of the
Sulfate Regional Experiment (SURE) and Related Research Programs. EPRI
EA-1066, Electric Power Research Institute, Palo Alto, California, 1979.
6. Morse, Sallle S. and Demetrlos J. Moschandreas. Indoor-Outdoor Pollution
Levels: A Bibliography. EPRI EA-1025, Electric Power Research Insti-
tute, Palo Alto, California, 1979.
/
7. Spicer, C. W. Workshop on Atmospheric Pollution by Trace Nitrogen Com-
pounds. EPRI EA-986-SY, Electric Power Research Institute, Palo Alto,
California, 1979.
8. Colucci, Anthony V. and William S. Simmons. Nitrogen Oxides: Current
Status of Knowledge. EPRI EA-668, Electric Power Research Institute,
Palo Alto, California, 1978.
9. Meserole, F. B., B. F. Jones, L. A. Rohlak, W. C. Hawn, K. R. Williams,
T. P. Parsons. Nitrogen Oxide Interferences in the Measurement of Atmos-
pheric Particulate Nitrates. EPRI EA-1031Volumes 1 and 2, Electric
Power Research Institute, Palo Alto, California, 1979*
483
-------
Viezee, W. and W. E. Evans. Development and Evaluation of a Prototype
Automated Telephotometer System. EPRI EA-1434, Electric Power Research
Institute, Palo Alto, California, 1980.
484
-------
Visibility
degradation
Ecological
effects
Human health
effects
Transformation
and
transport
Materials
damage
NOx
emissions
Figure 1. NOx Major Issues
-------
SINGLE-CYLINDER TESTS OF EMISSION CONTROL
METHODS FOR LARGE-BORE STATIONARY ENGINES
By:
Robert P. Wilson, Jr.
Arthur D. Little, Inc.
Cambridge, Massachusetts 02140
486
-------
ABSTRACT
The research work presented herein was undertaken in order to
develop combustion modifications which substantially reduce NO^ emis-
sions of large-bore engines, without significantly increasing fuel
consumption of carbonaceous emissions. The scope of the project covers
NO control technology for diesel and spark ignition engines, bore sizes
ranging from 8 to 20", and both 2 and 4 cycle charging methods. The
current status of the project permits us to report the results of 40%
of the Phase III experimental tests. In Phases I and II, a compendium
of 3* emission control methods was prepared, and an evaluation procedure
was used to screen down the list to the 12 methods which are now being
tested in Phase III.
Cooper Energy Services utilized a 20" bore, 330 rpm single cylinder
engine to test the effect of unmixedness (modified fuel gas injection)
and conventional "tuning" methods (timing, equivalence ratio, spark
location, gas valve location, and piston shape). The principal finding
was that N0x emissions are more sensitive to air-fuel ratio than any
other variable, giving a factor of five reduction as the equivalence
ratio was leaned out from $ ss .76 to a .62. The practical implication
is that spark gas engine emissions are limited primarily by turbocharger
efficiency and the combustion lean limit. Rate of heat release analysis
confirmed that a reduction in fuel-air ratio produces longer ignition
delay and lower flame speed. Gas valve modifications degraded NO^; pis-
ton shape had more effect on N0^ than either gas valve or spark plug
location.
487
-------
Fairbanks Morse conducted tests of pilot injection and increased
rate of injection on an 11" bore model PA-6 engine at 1,000 rpm. At
9 g/bhp-hr the baseline NQx is characteristic of this class of engines.
Retarded timing increases BSFC 1.1% and decreases N0x 4% per degree
crank angle. Exhaust temperature was found to limit the NO^ reduction
achievable with either pilot injection or injection rate at full load;
however, at part load NO^ reductions of 20% were found for both methods
with some BSFC improvement. Analysis of derived heat release profiles
show that the "spike" observed for high speed diesels does not appear
for the PA-6 engine.
This paper has been prepared under Contract No. 68-02-2664 by
Arthur D. Little, Inc., under the sponsorship of the U. S. Environmental
Protection Agency, covering work completed December 1978 through
February 1980.
488
-------
ACKNOWLEDGEMENTS
John H. Wasser, the EPA Technical Project Officer, has contributed
substantially to the project results obtained to date, through planning,
guidance, and insights derived from EPA in-house engine tests. 1 also
wish to acknowledge several Individuals at the Fairbanks Morse Division
of Colt Industries (Charles Newton, Eugene Kasel, and Dennis Bachelder)
and at Cooper Energy Services (Paul Danyluk, Fred Schaub, and Mel
Helmich) who have been responsible for the single cylinder engine tests
and the translation of NO -control concepts into practical engine hard-
ware. Key contributions have been made by several members of the
project team at Arthur D. Little, including Phil Gott, Larry Richardson,
John Mendillo, Bill Raymond, Don Hurter, and Ken Menzies. Finally, I
wish to acknowledge the vork of Maureen Donovan in the preparation of
this manuscript.
489
-------
SECTION 1
INTRODUCTION
Large-bore stationary engines are a significant source of NO
x
emissions in the United States. Primarily used in pipeline transmission
and compression of natural gas, oil drilling, and electric power genera-
tion, this class of engines accounts for 1.5% of U. S. fuel usage, as
shown in Table I. Covering a power range from 100 to 750 hp per cylin-
der, and a speed range from 300 to 1200 rpm, this engine class includes
both diesel and spark ignition types. Most large-bore engines are
turbocharged in order to reduce fuel consumption.
Emissions control efforts for large-bore engines are limited to
NO^, since the emissions of carbon-containing species (CO, HC, soot) are
relatively low and the conventional engine fuels nearly sulfur-free. As
shown in Table II, the N0^ emission rate of these engines is signifi-
cantly higher than any other major combustion device for a given amount
of fuel, and is almost a factor of six greater than the coal-fired
utility boiler (which is the stationary source of greatest concern).
This remarkably high N0x emission rate is caused by four factors:
• Compression preheating to 800°K increases flame tempera-
ture;
• Low heat loss during combustion maintains high flame
temperature;
• Air-fuel ratio in the 18-20 range which maximizes NO ;
x
and
• Low rpm results in 10-20 msec at peak temperature.
490
-------
In certain portions of the combustion chamber, the flame temperature is
so high that the NO^ level reaches equilibrium (8,000-12,000 ppm) and
then begins to decrease by the reverse of the Zeldovich mechanism as the
gas cools by expansion.
The outlook for emission standards for large-bore engines is uncer-
tain at this writing because of fast-changing control technology. In
July 1979, the EPA proposed NO^ emission ceilings corresponding to 40%
reduction from the typical level, as shown by Figure 1. These standards
as proposed would come into effect in 1982. However, while these stan-
dards were being formulated and reviewed, examples of new control
technology (e.g., torch-ignition for spark gas engines and catalyzed
reduction of NO^ by ammonia) began to be investigated by several engine
manufacturers. Recently, the California Air Resources Board issued a
model rule based on catalyst technology which calls for a 90% N0x reduc-
tion by as early as 1983, contingent on the successful demonstration of
control technology. Apart from the question of emission standards,
engine manufacturers are experiencing incentives for low N0^ emissions
due to PSD requirements. As shown in Figure 2, the installation of a
5,000-hp engine station will require an NO^ emission rate of 5 g/hp-hr
(which is equivalent to about a 60% reduction from uncontrolled levels)
in order to avoid a lengthy PSD review process.
The promise of recent developments reflects two basic premises
concerning the NO^ level of large-bore engines:
(1) There is substantial potential for HO reduction.
X
(2) The actual levels of NO^ which are feasible can only
emerge after extensive experimental and field test
work.
The EPA program described below Is aimed at this experimental test
work.
491
-------
SECTION 2
PROGRAM DESCRIPTION
The objective of the present program is to develop a technical
foundation for substantial reduction of N0x emissions of representative
large bore engines, without degradation of fuel economy. As the program
has evolved, it now appears reasonable to target for 50-60% NO^ reduc-
tion with less than 3% penalty in fuel consumption. Table III shows
the four major phases of work, starting with the identification of
various emission-control methods and ending with field tests on full
scale engines. Currently, Phase III is about 40% complete, and the
results of the first four methods of N0^ control are reported herein:
Spark Diesel
• Piston shape, gas valve, • Pilot injection
and spark location
• High injection
• Altered gas valve
The N0x control methods which are being tested for spark-ignition
engines fall into four categories, as follows:
(1) Reduced temperature
(2) Leaner combustion
(3) Stratified combustion
(4) Exhaust gas treatment
Figure 3 illustrates the mechanisms by which NO^ production is altered
in Categories (1)—(3) above. The contours labelled 1, 10, 100, and
1000 ppm/msec define a steep surface of increasing NO production rate.
Normally, a hot gas element follows the path labelled "conventional,"
starting at about 2400°K flame temperature and increasing to 2600°K or
more due to compression. This produces N0x at a rate well above
492
-------
1000 ppm per millisecond as illustrated by the contours on Figure 3.
The "reduced temperature" methods simply shift this path to the left
(about 100°K lower) where the NO^ production rate is much lower. The
"lean combustion" methods force the combustion to begin and proceed at
lower fuel-air ratio ( a; 0.7) where the NO production rate is reduced.
The "stratified combustion" method creates a lean and rich zone,
thereby avoiding the $ «¦ 0.8-0.9 regime (where NO^ production is at a
maximum). This brief discussion provides some background on the methods
which are being tested in Phase 111 for spark gas engines.
For diesel-engine N0x control, the approach which has guided our
Phase 111 plans is to retard the fuel injection process and compensate
for the delayed burning by:
• Reduced ignition delay, and/or
• Increased mixing rate.
Normally, retarded timing causes unacceptable penalties in fuel consump-
tion, and often results in excessive exhaust temperature and soot
levels. Therefore, it is essential for NO control to make the latter
x
stages of the combustion process more vigorous in order to allow the
early stages to be retarded and less intense.
Table IV presents the 12 NO^-control methods which were selected
for testing in Phase III. The process of selection was based on the
projected N0x reduction, the projected cost of control (per horsepower),
and the feasibility. The relative ranking of the concepts by two of
these criteria is illustrated in Figure 4, in which the threshold for
Phase III selection is shown as a dashed line starting at 20% N0^
reduction, and requiring greater NOx reductions above $30/hp ten-year
cost impact.
493
-------
SECTION 3
SPARK IGNITION ENGINE TESTS
A. EXPERIMENTAL ARRANGEMENTS
The tests were conducted using a Cooper Model Z330 single-cylinder,
two-stroke engine of 20" bore and 20" stroke dimensions, 681-hp rated
power at 130 bmep and 330 rpm. Gas fuel is injected directly into the
cylinder during the compression stroke, and the resulting mixture is
ignited by two spark plugs nominally firing at 7° before top center.
Figure 5 shows the air and fuel supply piping and controls in schematic;
the air manifold pressure and temperature could be controlled indepen-
dently. Figure 6 is a photograph illustrating the general layout of
the experimental facility. The large cylinder is the plenum in the air
supply used to dampen air pressure fluctuations.
Emissions samples were taken from the following three points, as
shown in Figure 7:
(1) Exhaust stack (diluted with scavenging air);
(2) Exhaust port (undiluted—see Figure 8); and
(3) Cylinder contents during combustion ("Cox valve").
Samples were analyzed for N0/N0x, THC, CO, C02» and 0£ by conventional
methods (chemiluminescent, FID, NDIR, and paramagnetic analyzers,
respectively). The fuel-air ratio of the trapped gas is essentially
unknown for a scavenged 2-stroke engine. This problem was solved by
mounting a poppet valve at the exhaust port as shown in Figure 8. The
valve was actuated by the flow of high pressure exhaust gas (undiluted
by scavenging air).
494
-------
Natural gas was used throughout the experiments; the composition
varied slightly as shown in Table V.
In order to provide additional diagnostic information about the
combustion process, the cylinder pressure traces were averaged for
several hundred consecutive cycles. The time-resolved rate of heat
release in the cylinder was calculated from each averaged pressure
trace by a thermodynamic computer program.
B. BASELINE DATA AND THE EFFECT OF FUEL-AIR RATIO
The N0x emissions of the engine in its normal configuration were
obtained for several spark timings and air-fuel ratios in order to
establish a baseline or reference for further tests. All runs were
made at rated power and speed (130 bmep and 330 rpm). The results,
presented in Figure 3, show that the air-fuel ratio affects N0^ markedly,
reducing NO a factor of five (from 25 to 5 g/bhp-hr) as the mixture is
X
leaned out from a fuel-air equivalence ratio of 0.76 to 0.62. Retarded
spark timing also reduces NO , giving about a 25% NO change for a
X X
timing shift of 4° crank angle (6% per degree).
The steepness of the NO^ vs $ graph, which is characteristic of
all premlxed combustion systems, suggests that the most fruitful
approach for low NO^ is to lean out the mixture. Several limits are
encountered, however, as the mixture is shifted to lower fuel/air
ratio:
• Misfire Limit: Positive, reproducible Ignition becomes
more difficult to achieve very lean mixtures. The flame
may begin to propagate away from the spark plug, but
falter or travel at low flame speed. The first sign of
misfire is increased variation in peak cylinder pressure.
Instrumentation on the Z330 single-cylinder engine auto-
matically calculates the standard deviation in peak
pressure (o), which is normally In the 60-90 psi range
for satisfactory ignition. In the 90-120 psi range, some
misfire is first noticed but engine operation may be
495
-------
marginally feasible. Above about 120-130 psi, misfire
is definitely unacceptable. For the baseline engine,
Figure 9 shows a- = 90 psi reached at about $ =» 0.62 to
0.66.
• Late or Incomplete Combustion: Flame propagation is
slower for very lean mixtures. This leads to penalties
in fuel consumption and CO/HC emissions due to pockets
of mixture which are partially unburned. Notice in
Figure 9 that the bsfc increases as the fuel-air equiva-
lence ratio is extended below about (J) = 0.65.
• Turbocharger Capacity Limit: In any shift to higher air-
fuel ratio, in order to maintain power, the fuel flow
cannot be reduced. Instead, the air pressure must be
increased; but the ability of the engine to supply itself
with high pressure air is limited. For a typical spark
gas engine, turbocharger efficiency is limited to about
64%, which ultimately limits the fuel-air ratio.
The emphasis of the EPA experimental program is on finding means to
overcome the misfire limit.
C. EFFECT OF PISTON SHAPE, GAS VALVE LOCATION, AND SPARK PLUG
LOCATION
Prior to the tests of specific N0x control methods, an attempt was
made to find out if the baseline engine configuration was optimum with
respect to the choice of piston and the location of gas valve and spark
plugs. Manufacturers often vary these components in order to "tune"
an engine for best fuel consumption and emissions. The matrix for
these "conventional" techniques is given in Table VI, which shows
systematic timing variations for each combination tested.
Three piston shapes (shown in Figure 10) were tested, and the
results are given in Figure 11. Apparently, substitution of the
"tee-pee" piston increased fuel consumption from 6,750 to 6,950 Btu/
bhp-hr (a 3% penalty). The "Mexican Hat" piston and the flat top
496
-------
(baseline) piston give almost identical performance, except that the
Mexican Hat piston may allow slightly leaner operation without misfire
(see Figure 11, top set of curves). If 110-psi pressure variation
were to be taken as the misfire limit, the baseline piston would be
limited to about 8 g/bhp-hr NO at cj) = 0.66 compared to the Mexican Hat
at 6 g/bhp-hr NO at $ ¦ 0.63.
X
The locations of the spark plug pairs and the gas valve are shown
in Figure 12. Variations in the location of these components made very
little difference in NO^ or bsfc with proper timing adjustments, as
shown in Figure 13. The center spark (open symbols in Figure 13)
allowed leaner operation without misfire (a below 120 psi).
In summary, the baseline engine configuration with flat top piston,
center spark plugs, and center gas valve appears to be relatively well
"tuned" with respect to NO^ and bsfc. This configuration was used as
the baseline standard for the remainder of the tests.
D. EFFECT OF GAS VALVE DESIGN
In a gas-injected engine, the mixture at the time of ignition may
contain incompletely mixed pockets. The number, size, and composition
of these pockets depends in part on the initial fuel gas dispersion
achieved by the gas valve. The standard gas injector is an outward
opening pintle valve of 1-1/8 in throat diameter (.405 sq-in annular
throat area). This design was altered, as shown in Figure 14, in order
to explore the effect of fuel gas dispersion (local stratification) on
NO^ emissions and fuel consumption.
The results are shown in Figure 15. It is interesting to note the
change in slope of the NO^ vs $ curves as the gas valve throat diameter
increases from 1.13" (o, •,a) to 1.31" (•) and then to 2.13" (O). The
flatter curves for a larger gas valve can be explained In terms of
differences in the extent of mixing at the time of combustion: It is
known that poorly-mixed fuel-air mixtures produce a constant amount of
N0x per unit fuel, no matter how much extra air is present. This is
because the extra air does not mix with isolated fuel pockets and
497
-------
therefore does not affect the flame temperature. In Figure 15 the very
large gas valve presumably resulted in this type of fuel-pocket combus-
tion and a flat NO vs 6 curve. By contrast, the well-mixed combustion
x
produced by a small gas valve is very responsive to any additional air
(lover <}>) because the additional air lowers the flame temperature.
As a result, the NO level can be increased but not decreased by
x
changing the gas valve design on the Z330 engine. The existing gas
valve appears to be nearly optimized in that it produces a very homo-
geneous mixture. Reducing the gas valve area from the baseline (.405
sq. in) may slightly reduce misfire (see the top set of curves in
Figure 15) and thereby extend the lean limit of operation.
E. ANALYSIS OF THE RATE OF HEAT RELEASE
For each run, the rate of heat release in the cylinder was calcu-
lated from a smoothed pressure trace. The rate of heat release is
determined by and therefore gives information about the ignition delay,
the rate of flame area growth in the early stages of combustion, and
the rate of burning in the critical late stage of combustion as residual
pockets of flammable mixture are reached by the flame and consumed.
Figure 16 illustrates typical rate-of-heat-release curves for two runs
made at 5° and 9° spark timing. Note that the two curves have essen-
tially the same triangular shape, but are shifted by about 4° (as
expected). The combustion duration is just under 40° crank angle. The
rate of heat release is sharply peaked, with the maximum heat release
occurring at the point of greatest flame area and rate of flame travel.
This probably occurs just as the flame (visualized as an irregular
hemisphere propagating toward the piston) first reaches the wall. The
front of the triangle is steep, the slope proportional to the average
flame speed. The back of the triangle is dependent on the chamber
geometry, which affects the flame area reduction. The "tail" represents
the consumption of residual unburned gas and is critical to the fuel
consumption performance.
498
-------
Recall that the fuel-air ratio, . As seen in Figure 17,
successive reductions in <|> cause four changes:
• Triangle starts later (longer ignition delay);
• More gradual front side of the triangle and peak reached
later (slower flame speed);
• Lower peak (lower maximum burning rate); and
9 Longer "tail" (longer combustion duration)
Further analysis of the shape of the rate of heat release is being con-
ducted. Apparently the NO^ level can be correlated with the slope of
the front side of the triangle (flame speed). As further data becomes
available, we will attempt to construct a function f (c*,0) of the same
form as the rate of heat release and correlate the parameters a, {5 with
operating conditions of the engine.
F. FORTHCOMING SPARK IGNITION TESTS
In the forthcoming tests, lean-burn methods such as torch-ignition
and multiple spark will be tested. These methods show promise of up to
70% NOx reduction, not only based on the baseline data on the Z330
single-cylinder engine, but also based on data presented by Helmich
(1979) and Chrisman (1980), as shown in Table VII. The research needs
for torch ignition which will be addressed by the EPA project include
the following:
• Methods of metering the amount of fuel in the torch chamber
in order to optimize torch fuel-air ratio.
• Systematic study of the effects of torch shape and direction,
by changing nozzle diameter, nozzle shape, nozzle length,
and orientation of the axis.
In addition ammonia/catalyst systems for exhaust gas treatment will be
tested. The research needs for these systems include the following
items:
499
-------
• Control systems for catalyst temperature
• NH^ carryover detection
• Platinum vs base metal
• Catalyst life (extended tests)
• Optimum space velocity
• Corrosion from sulfates (which form from the reaction of SO^
traces with NH^
500
-------
SECTION 4
DIESEL ENGINE TESTS
A. EXPERIMENTAL ARRANGEMENTS
The tests of diesel engine methods were carried out on an 11" bore
four-stroke turbocharged single-cylinder engine, the Model PA-6 of
S.E.M.T. design. The characteristics of this engine are as follows:
• Displacement, 1090 in3
• Compression ratio, 11.8/1
• Rated power, 350 hp* at 1000 RPM
• Brake mean effective pressure, 270 psi
• Injection: 20-mm plunger, 8 x .50 mm holes
• Piston shape: Mexican hat
• Standard timing: 23.5 btdc
Figure 18 gives a cutaway view of the entine. Manifold air temperature
and pressure can be controlled independently as in the Cooper facility
described above. The emissions and cylinder pressure instrumentation is
also comparable, with the exception that only exhaust sampling is done
for the diesel tests.
B. BASELINE CHARACTERISTICS AND THE EFFECT OF TIMING
The N0x emissions and fuel consumption of the PA-6 are strongly
influenced by the timing of the start of fuel injection, as shown in
Figure 19. The N0x can be reduced from 9 to about 6 g/bhp-hr by retarding
injection from 18 to 11° btdc (4.5% per degree). However, exhaust tem-
perature is excessive and there is a substantial fuel penalty. Attendant
to the N0x reduction is a fuel penalty of about 1.1% per degree of re-
tardation on the average.
Equivalent to 400 hp on multicylinder engine.
501
Arthur DLittletlnc
-------
These characteristics are representative of large-bore diesel
engines and the engine seems to be appropriate as a test facility. The
scatter in NO^ data for replicate runs is +0.5 g/bhp-hr or about 15%.
C. PILOT INJECTION
Pilot injection was tested in an attempt to compensate for the un~
desirable side effects of retarded timing. The pilot quantity was varied
from 4 to 12% of the total fuel per cycle, and four pilot advances were
tested (40, 60, 80, and 120° btdc). Figures 20 and 21 show the results
for full load and part load, respectively. The solid line represents
the baseline (no pilot) case. At full load, pilot injection did not
improve the BSFC-NOx characteristics. Retarded timing beyond 19.5® btdc
was not possible because of exhaust temperature limitations.
However, at part load, certain configurations of pilot injection
improved BSFC by up to 6% at fixed timing. Apparently pilot quantity
should be minimized unless timing is retarded. With retarded timing at
part load (Figure 21), N0X may be reduced by about 20%, Results were
inconsistent, however.
In summary, pilot injection did not give significant N0X reductions
except in special cases. The pilot system was not matched to the injec-
tor and this may have limited its potential NO effect.
D. INCREASED INJECTION RATE
The intent of increased injection rate, like pilot injection, is tp
allow retarded timing without undesirable side effects such as smoke
excessive fuel consumption, or high exhaust temperature. Figure 22
illustrates the difference in fuel injection schedule between the standard
injector (20 mm plunger, 8 x ,50 mm nozzle) and a high»-rate injector
(24 mm plunger, 9 x .55 nozzle). The larger injector has a higher average
fuel pressure and a shorter injection duration, Eight combinations of
plunger and nozzle were tested, as shown in Table VIII.
The results for full load (350 hp) and part load (234 hp) are pre-
sented in Figures 23 and 24, respectively. It is clear that unless timing
502
-------
is retarded, the increased injection rate acts to increase NO instead
x
of reducing it. With retarded timing, increased injection rate is bene-
ficial but the NO^ response is mixed:
(a) Full load - no improvement in NOx-bsfc.
(b) Part load - up to 20% NO^ reduction at fixed bsfc; and 30%
NO reduction if a bsfc penalty of 3% is permis-
sible.
The best injection system appeared to be the combination of the
24 mm plunger and 8 x 0.6 mm nozzle.
REFERENCES
1. Helmich, M. J., "Clean Burn Engine," paper presented at the AGA
Transmission Conference, New Orleans, Louisiana, May 1979.
2. Chrisman, B., "Exhaust Emissions Regulations and Control Technology,"
paper presented at the 27th Annual Meeting of the Gas Compressor
Institute, Liberal, Kansas, April 1980.
503
-------
1.25
1.00 _
Relative
NOx Emission
Levels
.75
.50
.25-
Proposed
EPA NSPS
(40% Reduction)
Advanced
Low-NO
Engines
(Torch, Catalyst)
Phases
I & II
Concepts
_J
Phase III
R&D
I
Phase
IV
Field
Tests
California
Model Rule
M90% Reduction)
-L
1977 1978 1979 1980
I EPA Program —
1981
1982
1983
1984 1985 1986
Figure 1. Scenario of the development of
large-bore engine emissions.
-------
16000
14000
12000
10000
8000
o
X
250 Ton/Yr. Limit
6000
4000
Uncontrolled Engines
2000
25
10
20
15
5
0
NOx (gm/hp-hr)
Figure 2. Low NOx allows larger engine installation
without PSD review process.
-------
1.4
1.3
1.2
1.1
en
O
CT>
Fuel/Air
Equivalence
Ratio, 0
0.9
0.8
0.7
0.6
0.5
Stratified Combustion (rich portion)
<§>
Leaner
Combustion
/ I Reduced Temperature
Conventional
iStratified Combustion (lean portion)
2,000 2,100 2,200 2,300 2,400 2,500 2,600 2,700 2,800
Temperature,°K
Figure 3. Basic categories of Nitric-Oxide suppression
for spark ignition engines.
-------
40
30
Projected
Feasible
N°x
Reduction (%)
20
10
T
O Emulsion
• Torch
EGR
O*
Dual Fuel
Refrigeration • /
/
^ ^Divided 0
Spark • oHi^hBate Chamber
• High Energy / Prechamber O
Spark /
Degraded Premix / Circumferential O
fji Injection
Turbulence
t
Modified
Shape
Threshold for
Phase 111
Selection
Pilot
O
Open Chamber
Stratified
Legend:
• Spark Gas
O Diesel
_i
_L
-L
10 20 30 40 50
Projected 10-Year Cost ($/HP)
60
Figure 4. Cost effectiveness of emission control methods
-------
Combustion Air
Excess
Flowmeter
Fuel
(Natural Gas)
Side
Valve
Flow Meter
Control
Valve
Center
Valve
Plenum
Exhaust Orifice
Exhaust
Gas
Heat
Exchanger
Dynamometer
Inlet Plenum
Exhaust Plenum
Figure 5. Plan view schematic laboratory single-cylinder engine Model
Z330 (20" bore x 20" stroke).
-------
Figure 6. General arrangement of Z330 single cylinder engine.
509
-------
DESICCANT
SPAN
SPAN
SPAN
^FLOW
METER
FLOW METER—
PR. REG.
SAMPLE
WASTE
PUMP
PUMP
SPAN
SPAN
—CONDENSER
4UU
r lheated
CAPILLAR Y11
LP5— |,
CONDENSATE
TRAP
figure 7. Analytical instrumentation.
BY PASS
PUMP
-------
Cylinder
Walt
cn
SAMPLE VALVE
Y-107
Piston Travel
Gas To
Analysis
Figure 8.
Sample valve for extracting cylinder blowdown
constituents from exhaust port.
-------
cn
ro
c
.2 •
tS 3
"> «
• £
Q£
¦o j*
!*
x
o 6.
U. £
CO GO
CO-
4-*
CD
120
100
80
60
- Misfire Onset
b\>-
330 RPM
130 BMEP, 681 BHP
rr
—D-
6800
6700
7°T er—
B v
-J •
6600
¦
«
25
Timing:
5° BTDC
20 -
_ 15 .
Z CD
- 10 -
5 -
Nominal Baseline
Condition
1 L
T6
j L
J L
7 .8
Trapped Equivalence Ratio (0)
I I i
Figure 9. Baseline characteristics Cooper Z330 engine.
-------
FLAT TOP MEXICAN HAT
Figure 10. Piston crown shapes.
-------
s 3
Misfire
Onset
Baseline
Teepee^
BSFC Limit
Teepee
Mexican Hat
b 6900
Base hne
m = 6700
tn
25
20
~ 15
2 XI
5 10
. Mexican Hat
/ Baseline
-
/ (Flat Top)
-
/jY Teepee
1
1 1 1
.7 .8
Trapped Equivalence Ratio (0)
Figure 11. Effect of piston shape.
Z330 Engine
330 RPM
130BMEP
Center Spark
Center Gas Valve
(1 1/8")
5° Timing
-------
SPARK PLUGS
GAS VALVES
Figure 12. Spark plug and gas valve locations.
-------
140
£
w 120
£100
.5
°S 80
tn 60
Z330 Engine
330 RPM
¦ \i NN. \
. 130BMEP
Flat Top Piston
Misfire
Onset
—« •
o Q-
u. -E
co 55
CO ^
7000
6900
6800
6700
S 6600 L
25
20
a
si
m
O!
15
-? 10
.6
-
-
rr^xBaseline
- i
•
•
Configuration
© Baseline (Center)
~ Side Gas Valve
~ Side Spark
Timing
7°
7°
11°
Side Spark & Gas Valve 11°
X
.7 .8
Trapped Equivalence Ratio ()
.9
Figure 13. Effect of gas valve and spark location.
-------
REDUCED
AREA
(.336-in.^ throat)
1'i STD.
(.405-in.^ throat)
1 ' ¦ Increased Area
16
2^
^ 8
(.704-in.^ throat)
(2.454-in.^ throat)
Pigure 14. Gas valves.
-------
C 2!
O 3
'% g}
"I*
Q -*
. «o
-ri <»
5 Q-
V)
150
130
110
90
70
50
330 RPM
130BMEP, 681 HP
• yw rW
Mexican Hat Piston
^ Center Spark & Gas
7° Timing
Misfire
^ — —JS
^5^-Baseline
Onset
.1
1
_ 6900
$ 6800
O Q-
& 55 6700
co "a
£ 6600
~ 6500
Baseline
20
a
Largest
D)
Throat/Stem Diameter
o1.13"/.50" (Baseline)
a 2.137.50"
° 1.31"/. 50"
• 1.137.52"
A 1.137.5®"
x
.7
.8
.6
Trapped Equivalence Ratio (0)
Figure 15. Effect of gas valve modifications.
518
-------
Z330 Engine
130BMEP
0= .70
5° BTDC
-20
TDC 20
Crank Angte (Deg.)
Figure 16. Effect of Timing
519
-------
7330 Engine
330 RPM
130BMEP
Baseline
5° Timing
621 (Misfire
Onset)
-20 TDC 20 40
Crank Angle (Deg.)
Figure 17. Effect of Fuel-Air Ratio
520
-------
Figure 18. Section across a cylinder of the PA-6 engine
521
-------
X
CL
X
m
o
o
LL
CO
OQ
180
170 -
160 L
200
190
180
Multi Single
10
x
a.
x
OQ
a
o
z
8
4 .
Slope « 1.5%/
DEG. C.A.
J L
J L
Baseline 9 G/BHP-HR
o
Slope « 4%/
DEG. C.A
JL
-L
Exhaust
Temperature
Limit
I.
18 16 14 12
Timing (DEG. BTDC)
10
8
PA—6 Engine
350 HP
20 mm Plunger
8 x 0.5 mm Nozzle
1000 RPM
Figure 19. Diesel Baseline Characteristics
522
-------
11
10
9
8
7
Baseline
PA-6 Engine
350 HP
1000 RPM
6
170
180
190
200
210
BSFC (g/Bhp-hr)
Figure 20. Effect of Pilot Injection
(Full Load)
523
-------
11 -
10 -
x:
Q.
J2
CO
^ 8
X
O
8%@ Q
100° -*
- 4% @40°-
4% @ 80°
5%@ 100°-*00
Baseline
7 -
6 .
s
4% @ 40°
PA—6 Engine
1000 RPM
234 HP
JL
o6>
j.
X
170 180 190 200
BSFC (g/Bhp-hr)
210
Figure 21. Effect of Pilot Injection
(2/3 Load)
524
-------
A/E£DL£ UfT
~U£L P££-55UZE
24mm PWA/GEZ
9x. 55mm mozzlE
eomm PLUMSc^
3x.SO mm NOZZLE (5A5EUNE)
-20 TDC t20
Figure 22. Effect of Plunger/Nozzle on Injection Rate
-------
10
oi
no
CT>
- 6
£.
a
!
*x
8 S
PA-6
350 HP
1000 RPM
(^) Baseline
(8 x^Omm, 20mm)
o
Increased Rate
Baseline
Runs Identified by Engine Hour
_L
J.
180 190 200 210
BSFC (g/Bhp-hr)
Figure 23. Effect of Rate of Injection
220
230
-------
PA—6
234 HP
1000 RPM
Q Baseline
(8 x .50mm,
284
280
20mm)
297
\l274j_
8x0.60,
24mm
Other
| | Increased
Rate
Numbers indicate
Engine hour
Baseline
8 x .60mm
24mm
Xr
180
190 200
BSPC (g/Bhp-hr)
210
220
230
Figure 24. Effect of Kate of Lnjection
-------
TABLE I: STATIONARY LARGE BORE ENGINE POPULATION
„ , Estimated Annual
"8 Application Fuel Use
^ <10" Btu)
Spark Gas Pipeline Transmission 3.4
Gas Gathering, recompression
and storage 1.7
Diesel Deep oil well drilling rigs and oil
transport 3.0
Baseload electricity generators for
municipal utilities 1.7
Standby generating sets for nuclear
and hospitals 0.7
Industrial power and water/sewage
pumping 0.6
TOTAL 11.0
(1.5% of U. S. Fuel Use)
528
-------
TABLE II: N0X EMISSION RATE OF LARGE BORE ENGINES
COMPARED TO OTHER COMBUSTION DEVICES
n . Typical NOx Emission
V ce (Ib/MM Btu)
Large Bore Engines 4.0
Automotive SI Engines 2.0
Coal Fired Utility Boilers 0.7
Industrial & Commercial Boilers 0.4
(Oil Fired)
Industrial Furnaces 0.3
(Gas Fired)
Residential Furance and 0.1
Water Heater
529
-------
TABLE III: MAJOR PHASES OF WORK
Share of Project
Resources
PHASE I
(4 months)
Identify emission-control methods
(34 identified)
3%
PHASE II
(11 months)
Evaluate the methods to select the 12
most promising based on:
• predicted NO^ reduction
• retrofit feasibility
• BSFC effect
• cost
14%
PHASE III
(24 months)
Operate single-cylinder 20"-bore spark
and ll"-bore diesel engines to test the
12 promising emission control methods
66%
PHASE IV
(9 months)
Field tests on multicylinder engines
17%
530
-------
TABLE IV: PHASE III TESTING
SPARK IGNITION
Conventional
Spark Location
Gas Valve Location
Piston Shape
Lean Burn
Multiple Spark
Torch Ignition
High Energy Spark
Feedback Control
Stratified Charge
Degraded Premix
External
Charge Refrigeration
NH3/Catalyst
DIESEL
Retarded Timing
Emulsion
High Injection Rate
Pilot Injection
External
EGR
NH3/Catalyst
531
-------
TABLE V: FUEL COMPOSITION
(NATURAL GAS)
March 1979
July 1979
October 1979
January 1980
Methane
95.70%
96.00%
97.96%
95.07%
Ethane
3.20
3.00
1.07
1.82
Propane
.17
.14
.21
.27
Butanes
.03
.04
.08
.12
C,. and up
.03
.04
.06
.08
Nitrigen
.35
.37
.20
2.18
Carbon Dioxide
.47
.40
.42
.45
Heating Value
(Btu/scf)dry
1016
1033
1024
1011
532
-------
TABLE VI! TEST MATRIX FOR CONVENTIONAL TECHNIQUES
CENTER
SPARK
SIDE SPARK
Center Gas
Side Gas
Center Gas
Side Gas
Timing
9 7 5 3°
9 7 5 3°
13 11 9 7°
13 11 9 7°
Mexican Hat Piston
- • • -
- • • •
- • • -
- • • -
Teepee Piston
• • • •
• • • -
• • - -
• • • -
Flat Top Piston
- • • -
- • • -
- • • -
• • •
Each dot represents 3 to 5 runs, varying F/A.
533
-------
TABLE VII
PRELIMINARY DATA ON LEAN-BURN ENGINES
Source
Engine
Configuration
NOx
(g/bhp-hr)
Fuel Penalty
Cooper Bessemer
(EPA Program)
Z330
20" Bore
1-Cylinder
I ncreased
Air Pressure*
{* = .72 .62)
16.2 -~ 4.9
(70%)
1-2%
Cooper Bessemer
(Helmich, 1979)
GMVH
14" Bore
12-Cylinder
• Jet-Cell
• Modified TC
• Retarded Timing
10.2 -»• 3.0
(70%)
2.6%
Cooper Superior
(Chrisman, 1980)
SGT
10" Bore
2-Cylinder
• Pre-Chamber
• Retarded Timing
• 0 - .67 -*¦ .54
13.0 -~ 4.0
(70%)
Unspec.
*No Turbocharger Limitation
-------
ABSTRACT
This paper is based upon work sponsored by EPA to evaluate combustion
modification technology applied to industrial process equipment. The test
program described herein was aimed at the development of combustion modifications
for reducing NO^ emissions from process heaters. In particular, staged
combustion air and lowered excess air were applied separately and in combination
to a natural draft vertical cylindrical crude heater and the effects on heater
efficiency and NO^ emissions were studied.
At a crude throughput of approximately 59 percent of the heater capacity,
reductions in NO^ emissions of over 50 percent from baseline emission levels were
observed when firing refinery gas fuel and using the combined modifications of
staged combustion air and lowered excess air. Burner and heater performance
actually improved slightly with the application of these modifications. An
increase in heater efficiency of over two percent was observed for low-NO^-firing
as compared to baseline conditions.
The sane modifications were tried firing residual oil simultaneously with
the gas fuel. Some reduction in NO^ emission was achieved, however the magnitude
of the reduction was smaller than that obtained for gas fuel only.
The cost effectiveness in dollars per unit mass of N0x removed is
calculated and the feasibility of the staged air/low excess air modification for
retrofit application to natural draft process heaters is discussed.
537
-------
SECTION 1
INTRODUCTION
At the Third EPA Stationary Source Combustion Symposium, KVB reported on
subscale process heater combustion modification tests. The report summarized
the effects on N0X emissions of several types of modifications, including
staged combustion air, flue gas recirculation, lowered excess air, altered
injection geometry, and low N0X burner installation. The work showed that
staged combustion air appears to be the most cost effective combustion modifi-
cation for process heaters* Both staged air and flue gas recirculation
produced N0X emission reductions in excess of 60 percent below baseline
emission levels.
COMBUSTION MODIFICATIONS TO A FULL SCALE PROCESS HEATER
The present report summarizes the testing of a full scale natural draft
refinery process heater firing natural gas, refinery gas, and No. 6 oil
fuels. A staged air injection system was designed by KVB and installed by the
refinery for these tests. The system is capable of supplying up to 50 percent
of the stoichiometric air requirement. The staged air is injected by means of
lances inserted through the heater floor as it was in one of the subscale test
configurations. The process heater was tested with and without the staged air
modification and over a range of loads and stack excess oxygen concentra-
tions. Stack gas emissions and heater efficiencies were measured for all
conditions*
Lowered excess air and staged combustion air modifications were applied
separately and in combination. NOx emission reductions of up to 52 percent
below baseline levels were achieved when firing gas fuel with no short term
538
-------
ill effects on the heater. Heater efficiency was increased at the low NO^
conditions due to the lowering of excess air. Staging the combustion air did
not appear to significantly alter efficiency; the same level of lowered excess
air was attainable both with and without staging.
In the most favorable situations the combination of staged air and
lowered excess air costs $64/Mg N0X reduction* Staged combustion air applied
separately is expected to be considerably more expensive-approximately $2600-
$3000 per Mg NOx reduction.
OBJECTIVE AND SCOPE
The objective of the program is to develop advanced combustion
modification concepts requiring minor hardware modifications that could be
used by operators and/or manufacturers of selected industrial process equip-
ment to control emissions* The development is aimed at equipment on which the
modifications will be most widely applicable and of the most significance in
mitigating the impact of stationary source emissions on the environment* The
program involves investigation not only of emissions but also multimedia
impacts and control cost effectiveness*
The program includes both subscale and full-scale testing* Subscale
testing is a necessary part of development of new hardware to ensure accept-
able performance, which is a vital aspect of emissions control* Full-scale
testing is also necessary on more than one process design configuration (e.g.,
forced draft and natural draft) before equipment manufacturers and the process
industry can employ a given emission control technology*
At the conclusion of the study, a final engineering report will be
prepared summarising the accomplishments of the subscale and full-scale deson-
stration tests* A series of guideline manuals will be prepared to acquaint
equipment manufacturers with the most promising emission control methods that
have been demonstrated and to offer technical guidance that can be directly
applied in their process equipment design*
539
-------
SECTION 2
TEST HEATER AND EMISSIONS SAMPLING
TEST UNIT DESCRIPTION
The test unit was a natural draft/ vertical cylindrical crude oil process
heater which is used to supply a partially vaporized charge to a crude oil
3
distillation column* A maximum load of 108 m /h (16,250 bbl/d) may be sent
through the heater in two passes* A sketch of this crude oil heater is
presented in Figure 1*
The maximum firing rate of the heater is 16.1 Mw thermal input (55 x 10®
Btu/hr). It is fired by six John zink DBA-22* natural draft burners. The
burners are combination gas/oil burners rated at a maximum of 2.68 Mw (9.14 x
10® Btu/hr) each with a turndown ratio of 3:1. Although combination gas/oil
burners are used, some gas must always be fired because the unit is base
loaded at constant oil firing rate and an automatic tenperature controller
adjusts the gas fuel flow to maintain crude oil outlet temperature.
Three parameters may be controlled in the heater: excess oxygen, (by
furnace draft), firing rate, and load. The stack damper is the main control
for the pressure drop across the furnace. The pressure drop nay also be
controlled by opening or closing the secondary air registers which are in the
base of the heater. Each burner has a set of primary and secondary registers
~Mention of trade names does not
Protection Agency.
constitute endorsement by the Environmental
540
-------
which can be adjusted independently of the other burners• The stack damper
and register adjustments establish the excess oxygen. The load is controlled
by pumps and valves on the inlet and outlet of the heater.
STAGED AIR SYSTEM DESCRIPTION
This application of staged combustion utilizes air lances in the firebox
to supply air to the flame zone a given distance above the base of the
flame. Figure 2 presents a schematic of this system.
The system consists of twenty-four vertical 316 stainless steel pipes
3.18'cm (1-1/4 in.) diameter arranged four per burner at 90° apart. A 45°
elbow is placed on each pipe to provide better mixing across the flame. A fan
supplies air to the lances through a manifold and flexible tubing. The lances
may be varied in height up to 1.2 m (four feet) from the base of the
burners. Extensions for the lances allowed staging heights up to 2.4 m (eight
feet) for oil firing tests.
EMISSIONS SAMPLING INSTRUMENTATION
Monitoring of the required gaseous and particulate emissions was
performed with an EPA furnished mobile laboratory. The laboratory's monitor-
ing capabilities are presented in Table 1. A schematic of the continuous
monitoring system is presented in Figure 3• A detailed description of a
similar mobile emissions laboratory has been presented in a previous report
(Ref . J_) .
Continuous gaseous emissions analyzers provide the capability for
measurement of O2t 002# 00, SOj, NO, NOx, and HC (as methane) in the flue
gas. Particulate total mass and sizing as well as S03 measurements are non-
continuous. SO3 measurement is by a controlled condensation technique using a
Goksoyr-Ross type coil*
1. Hunter, S.C., et al. "Application of (bmbustion Modifications to
Industrial Combustion Equipment," EPA Report 600/7-79-015a, NTIS Order
No. PB 294214, January, 1979.
541
-------
The 02, CO2, and CO concentrations are measured on a dry basis. S02, HC,
and NO2 measurements are made through a heat-traced sample line and are deter-
mined on a wet basis. NO may be sampled either wet or dry.
542
-------
SECTION 3
DISCUSSION OF RESULTS
LOAD AND EXCESS OXYGEN VARIATIONS
A series of tests was conducted to evaluate the performance of the
process heater with regard to N0X emission and efficiency over varying load
and excess oxygen conditions. These tests were first made while firing a
gaseous fuel mixture consisting of natural gas and refinery fuel gas and then
while firing residual oil simultaneously with the gas mixture*
The NO emissions as a function of excess oxygen are shown in Figures 4
and 5. The highest NO emission generally occurred at the intermediate load
condition, 70% of rated capacity. NOx emission increased with increasing
excess oxygen up to approximately 4-6 percent 0£ for all fuel combinations.
In the 4-6 percent 02 range, NO emission was relatively insensitive to excess
02 level and, at higher excess 02, NO emissions decreased.
The nitrogen content of the gaseous fuels was negligible whereas the
nitrogen content of the oil fuel was 0*8 percent by weight, hence, the
difference in absolute NO emission between Figures 4 and 5. Several ratios of
oil/gas are reported in Figure 5. These ratios indicate the approximate fuel
split by percentage of total heat input* Changes in the fuel split occurred
because a) gas fuel composition and heating value could change in a short time
period and b) total heat input required to maintain the load could change due
to changes in the composition of the crude charge to the heater.
At the intermediate load, the baseline excess oxygen was 4 percent which
was regarded by the plant as a normal operating level* The excess oxygen
variation tests indicated that continuous operation of the heater at 2 percent
543
-------
excess 02 was possible without additional operator supervision. Below this
level the heater draft tended to become unstable. Fluctuations in the draft
caused occasional smoking of the unit and resulted in positive stack or
convection section pressures which are dangerous in a natural draft heater
because they cam cause flashback. Flashback occurs when a flame encounters a
back pressure which forces it downward and out the bottom of the heater
through the air registers.
STAGED COMBUSTION AIR TESTING
The next phase of testing at the process heater site involved the
evaluation of the staged combustion air system. This evaluation included the
variation of three important parameters: 1) burner equivalence ratio,
2) excess oxygen level, and 3) staged air insertion height.
For the staged combustion air tests, due to changes in plant operation,
it was necessary to fire a different refinery gas from that used in load and
excess oxygen variations. This gas, called "adsorber gas," contained a
greater percentage of higher hydrocarbons than did the "fuel gas" or natural
gas and, therefore, a higher heating value. Baseline NO emission with
adsorber gas was about 8 percent higher (8 ppm) compared with fuel gas.
The NO emission is graphed as a function of burner equivalence
(A/F)
ratio, B ¦ . . urner , The staging height for
Stoichiometric
this test series was 1.2 m (4 ft.) and the load was 64m /h (9600 bbl/d),
60 percent of rated capacity. At each overall excess 02 level 4>B was
decreased in steps to its minimum value which was determined by the
limitations of the staged combustion air fan. Figure 6 shows that at
4 percent 02 the minimum 4>B (maximum staging) obtained was 0.74 which
decreased NO emissions 35 percent below the baseline of 105 ppm, dry at
3 percent 02« At 2 percent 02 and minimum $B of 0.65 the NO concentration
dropped to 51 ppm, dry, corrected to 3 percent 02» This represented a
reduction of 51 percent below the 4 percent 02 baseline condition and a 43
percent below the 2 percent 02 (non-staged) level. Table 2 summarizes the N0X
544
-------
reductions and efficiency changes for the various combustion modifications to
the process heater while firing adsorber gas.
Figure 7 shows similar trends in NO emissions for varying $B and excess
O2 at a higher load. The same degree of staging could not be achieved at the
intsrmediate load because of fan capacity limitations, however, the curves in
Figure 7 indicate that roughly the same N0X reductions would have been
obtained had the minimum values of Figure 6 been reached.
The results of another test series in which overall excess oxygen and
burner equivalence ratio were kept constant while varying staged air insertion
height are shown in Figure 8. (The staged air height is defined as the height
above the heater floor at which the staged combustion air is injected. This
is approximately equal to the height above the burner gas tips and oil gun.)
There was little, if any, effect of staging height on NO emissions. Since the
burner tile top was about 0.23 m (0.75 ft) above the floor of the furnace and
since the staged air pipes were located on a diameter outside that of the
burner tile injection heights of less than about one foot above the heater
floor resulted in impingement of the staged air on the burner tile. Thus, the
minimum staging height was approximately 0*3 m (one foot).
The results of staging the combustion air while firing the No. 6
oil/adsorber gas mixture at intermediate load are shown in Figure 9 and 10.
While lowering the excess air reduced NOx by a percentage similar to that
obtained firing gas fuel only, the use of staged combustion when firing oil
with gas did not produce as large a percentage decrease in NOx emissions as it
did when firing adsorber gas only. The absolute amount of NOx reduction,
however, was about the same for combined fuel firing as it was for gas
alone. For example, for the case of staged combustion air combined with
lowered excess air firing oil and gas. Figure 9 shows that the NO level
dropped from a baseline of 219 ppm, dry corrected to 3 percent » to 166 ppm,
dry at 3 percent O2 for a drop of 53 ppm. For the same conditions firing
adsorber gas only at the intermediate load the drop was 46 ppm.
This behavior indicates that fuel nitrogen conversion to NOx was
reponsible for a large fraction of the observed emissions when firing the
oil/gas mixture, it appears then that approximately half of the baseline NO„
545
-------
emission firing oil with gas is due to fuel nitrogen and half is due to
thermal N0X since NO emissions firing this mixture were about twice those
observed firing a nitrogen-free refinery gas. The estimated fuel nitrogen
conversion efficiency/ based on the ratio of oil to gas in the fuel and the
oil fuel nitrogen content (~0*8%), is approximately 19 percent.
Figure 10 shows that NOy emission decreased slightly as staging height
increased firing oil with gas. Very little decrease was observed at heights
greater than 1.2 m (4 feet).
During all testing with the staged combustion system operating, careful
observation of the flame and furnace draft was made. There did not appear to
be any problems with coking of the process tubes and at no time was there any
emission of carbon monoxide or unburned hydrocarbons even at 2 percent stack
excess oxygen. Biere were certain instances in which the draft in the convec-
tion became slightly positive at the low Oj condition with maximum staging,
however, flashback was never observed. The long term effect of this change in
furnace draft in terms of maintenance or operational costs is not yet known*
COMPARISON OF PRESENT RESULTS WITH SUBSCALE RESULTS
Table 3 summarizes the NOy reductions obtained at the present full scale
process heater with the emission reductions observed at the subscale test
site. For gaseous fuels the trends are the same for both subscale and full
scale tests except that the percentages are somewhat lower in the present
data. For tests with oil fuels lowered excess air alone was more effective in
reducing NOx emission than was staged combustion alone - unlike the trend
observed in the subscale test where staging the combustion gave a signifi-
cantly greater NOx reduction than low excess air. Also, the percentages are
much lower for the full scale heater them they were for the subscale heater
firing No. 6 oil alone*
The reasons for the lower percentage NOx reductions firing gas fuel only
in the full scale heater are not altogether clear* Furnace bulk gas and wall
temperatures were much lower in the full scale heater than they were in the
subscale unit (~700°F vs. 1700#F). Residence times, however, were probably
longer in the full scale heater* ®ie baseline NO emission with the same
546
-------
burner type in the subscale heater firing natural gas was 131 ppm, dry,
corrected to 3 percent (>2 compared with the present 105 ppm, which is
reasonable in light of the temperature differences mentioned. There may be a
decrease in the effectiveness of combustion modifications in reducing thermal
NOx as temperatures are decreased* The mixing of the combustion air and fuel
may have been different enough to cause the changes in NOx reductions from
subscale to full scale* The precise location, angle of injection, and spray
angle associated with the staged air system are probably also important
parameters in the NOx reduction process.
Unfortunately, direct comparisons of the oil-firing results are not valid
since the full scale unit was firing a considerable amount of gas along with
the residual oil. In the subscale tests no gas was fired with the oil.
Although the fuel nitrogen in the subscale tests was much lower (~0.3%) than
it was in the full scale tests, the apparent fuel nitrogen conversion in the
former tests was about twice that of the latter. The fractions of thermal and
fuel N0X are thought to be about the same in either case-about 50-50. Further
work is planned to determine whether or not the performance of these combus-
tion modifications can be improved for oil-firing applications.
547
-------
SECTION 4
COST ANALYSIS
In this paper we will concentrate on the cost analysis of lowered excess
air and staged combustion air applied separately and in combination to a
process heater with a thermal input capacity of 16.1 MW (55 x 10® Btu/h). We
will consider only the gaseous fuel application. Costs for oil-fired units
will be determined pending the results of further experimentation.
INITIAL CAPITAL COSTS
There is no initial capital cost associated with the lowered excess air
(LEA) modification at the levels of O2 used in the present tests* Such costs
would be incurred if operation at stack oxygen contents of less than 2 percent
were desired since an automatic oxygen control and analyzer would then be
required.
The initial capital costs of a staged air system such as that of Figure 2
are given in Table 4. The installed costs include direct labor and overhead
charges at contract labor rates. Required equipment and materials include a
high pressure blower, a damper, piping, fittings, valves, stainless steel
lances for insertion into the firebox, and temperature and flow measurement
probes. Shipping costs are also included.
The annual operating costs for combustion modifications to a full scale
heater are given in Table 5. The fuel savings were based on the efficiency
increases for the various modifications shown in Table 2 and a market price of
natural gas of $2.50/106 Btu. (The latter assumption may not be valid in a
548
-------
case where the refinery uses an off-gas which is not readily marketable and
would otherwise be flared. In that case, there would be no cost savings
associated with the increased efficiency) .
In determining the total annualized costs of combustion modifications
several assumptions regarding state and federal taxes, insurance, depreciation
method, and financing constraints have been made. These are listed in Table 6
and the total annualized costs for the process heater modifications are given
in Table 7 .
Note that there are two columns for the staged combustion air (SCA)
modification only. The one labelled "4% 02" applies to the SCA modification
at a normal operating condition which results in 4 percent O2 at the stack •
The fuel savings and annualized costs are calculated relative to those of a
heater having a 4 percent 02 condition without staging. For the column label-
led "2% 02" the fuel savings and annualized costs are calculated relative to
those of a heater having a normal operating condition of 2 percent 02 at the
stack with no staging. For those refinery heaters which normally operate at
~2% 02 the column "SCA Only-2% 02" applies. For those which normally operate
at ~4% 02 the column "SCA Only-4% 02" applied. The "SCA + LEA" column shows
fuel savings and annualized costs relative to a 4 percent 02 normal operating
condition without staging.
One may ask why all process heaters do not operate at 2 percent 02 rather
than 4 percent 02 since it appears to be cost effective to do so. Refinery
gas varies significantly in composition and thus requires a varying amount of
combustion air. To avoid the possibility of air deficiency and to minimize
the need for continuous operator attention, many process heaters are operated
at a higher excess oxygen than would be necessary with a more consistent
fuel. With the current need to conserve energy, this practice is being
revised by installing oxygen controllers and increasing operator attention.
Table 8 gives the annual N0X emission reductions and cost effectiveness
of each combustion modification. The annual N0X emission reduction is calcu-
lated using the values given in Table 2 for percent N0X reduction and assuming
continual operation at 70 percent capacity for 80 percent of the year. The
549
-------
cost effectiveness is simply the total annualized costs divided by the annual
N0X reduction capability.
Table 8 shows that for a gas-fired heater which normally operates at
4 percent excess oxygen the combination of staged combustion air and lowered
excess air is an economically and environmentally attractive modification.
Lowered excess air alone is very economical but is only 1/3 to 1/4 as effec-
tive in reducing N0X. For a heater which normally operates at 2 percent Oj
the costs of staged combustion air would not be greatly different from those
arising from the use of SCA alone at 4 percent 02« This cost is, however,
much greater than the cost of SCA + LEA.
The question of which modification is most cost effective in a given
instance will depend on each individual plant situation and each individual
heater* Each plant must determine the following:
1) the credit, if any, it should allow for fuel savings,
2) the normal excess oxygen in the heater, and
3) whether or not a lower excess oxygen can be obtained without
additional cost•
550
-------
SECTION 5
SUMMARY
Combustion modifications proved effective on a natural draft process
heater rated at 108 m /h (16,250 barrels/day) of crude capacity when firing
refinery gas. In addition, heater performance was measured for other fuel
mixtures including natural gas and No. 6 oil.
While firing refinery gas a maximum NOy emission reduction of 51 percent
was observed, below a baseline emission concentration of 105 ppm, dry
corrected to 3 percent 02* An increase in efficiency of 2.37 percent was
observed for this condition and, in general, efficiency was improved by the
application of combustion modifications*
Staged combustion air gave a significant NOx reduction when firing
refinery gas regardless of excess oxygen level. Staging height did not have a
major effect on NOx emissions over the range of 0.3 to 2*4 meters.
The N0X emissions occurring for the various modification techniques
, followed the trends observed at the subscale level, however, the percentage
reductions were slightly less. In absolute magnitude, the N0X emission reduc-
tions occurring for gas-firing were approximately equal to those occurring
when No. 6 oil was included in the fuel. The percentage reductions were less,
however, for oil apparently because of the fuel nitrogen conversion.
551
-------
A cost analysis cowering initial capital and annual operating costs,
total annualized costs, and cost effectiveness of each of the combustion
modifications was made. The results indicate that staged combustion air may
be an economically and environmentally desirable modification, however, the
cost effectiveness for any single application will ultimately depend upon
several factors unique to that situation.
552
-------
LOCATION 7 - NATURAL DRAFT REFINERY PROCESS HEATER
c
* E
I v
c' CO
i
c
in
i
i
8
E
0
-------
STAfiF f) AIR SYSTFMl
HEATER
xiblc tubing
see oc
3 PVC BALL VALVE
Itf* 8HCLJ
(TIHlTltRFLY WUȣ
PI TOT A U TUBE
0F4SADDLE tPVC)TO CONNECT FP\CT0 4*3 P\C PUSHING
1 3/T 10. FLEXIBLE TUBING
Tr M «"R€DUCCR
X I v4 ~il J :W
riTOTftii-Tuec
M J BALL VAlVi
¦3 Pvc
DETAIL A
SUMS VIEW
TYPICAL
MTA1L •
TOP VIEW
TYPICAL
Figure 2. Flow Schematic of Staged Combustion Air System for a Natural
Draft Process Heater.
-------
Hot
Sanple Ory Staple Lines
Line (Typical See-Up Six Lineali
Heated Line
Sawple
Puaips
(3)
Condenser
,, 16
ij (tot/Cold
T Switch
Plowieetars |(|
ZeronZjSpan
Refrigeration Condenser
ISasiple Pressure
|_Mani_£old___J
Vent
Xero
|Span
Span
CO
CO.
NC
SO.
MO.
Figure 3. Flue gas sampling and analyzing system.
-------
NO emissions at three different loads as a function of stack
excess oxygen for a natural draft process heater firing a
natural gas/refinery gas mixture.
110
7/1-3
100
7/3-2
7/1-1
L7/3-3
7/3-1
7/1-2
7/2-4
7/2-
7/2-2
Fuel: Natural Gas +
Fuel Gas
18.2 kg/s of crude
= (11,000 bbl/day)
7/1-4
7/3-4
(14,000 bbl/day)
(8,000 bbl/day)
Stack Excess Oxygen, %, dry
Figure 4.
-------
NO emissions at three different loads as a function of stack
excess oxygen for a natural draft process heater.
cn
tn
O
CO
<3
£
¦o
E
Cl
a.
300
250 -
200 -
150 -
100
50 -
Oil: NG
63:37
37:6
Oil: NG
Oil: NG = 46:54
Fuel: Residual Oil +
Natural/Fuel Gas
NG = 44:56
^18.2 kg/s
(J (11,000 bbl/day)
~23.1 kg/s
(14,000 bbl/day[high gas])
A 13.2 kg/s
(8,000 bbl/day)
£S23.1 kg/s
(14,000 bbl/day [high oil])
I I I
8
STACK EXCESS OXYGEN, %, DRY
Figure 5.
-------
No emissions as a function of burner equivalence ratio at two
excess oxygen levels with constant staged air insertion height.
120
100
90
80
70
60
50
I I |
Fuel: Adsorber Gas
Staging Height: 4' (1.2m)
Load: 9600 bbl/d (15.9 kg/s)
04% °2
~
2% O,
7/12-5
7/12-1
7/12-6
17/12-7
7/12-10
7/12-2
7/12-3
7/12-4
7/12-8
7/12-9
40
t
0.2
0.4
0.6
0.8
1.0
1.2
= (a/F) ACT/(A/F) STOIC.
B
Figure 6.
558
-------
NOemissions as afunctionof burner equivalence
ratio at two excess oxygen levels with constant
staged air insertion height.
i 1 r
Fuel: Adsorber Gas
Staging Height: 4 ft. (1.2m)
Load: 11,500 bbl/d (18.9 kg/s)
O 4% Qj
Q 2% Oz
T
T
.t
0.2
0.4
7/9-1
7/9-4
0.6
0.8
1.0
1.2
Figure 7.
559
-------
NO emissions as a function of
staging height at constant excess
oxygen and 6 .
B
100
7/13-4 7/13-3
7/13-5
7/13-2
-O
Fuel: Adsorber Gas
Load: 11,500 bb|/day(18.9 kg/s)
4% 02, 0_ = 1.00
B
i
2 3
Staging Height, ft.
Figure 8.
560
-------
NOemissions as a function of burner equivalence ratio
for a gas-oil fuel mixture.
220
Oz = 4%
7/18-8
Baseline
Load: 11,500 bbl/d (19.0 kg/s)
Excess 02: 2%
Staging Height: 4 ft (1.2m)
Fuel: 47% Adsorber Gas +
53% No. 6 Oil
210
200
190
180
/ 7/18-7
(No staging)
7/18-4
7/18-5
170
165
7/18-6
0
0.8
0.9
1.2
1.3
0.6
0.7
1.1
1.0
0B
Figure 9.
-------
NOemlssion as a function of staging height for gaa-oH
fuai mixture.
220
Load: 11,500 bbl/d (19.0 kg/s)
Excess 02: 4%
>v 7/15-6
21OV^ Baseline
7/15-5
7/15-2
Baseline
Fuel: *•'75% Adsorber Gas +
-25% No. 6 Oil
7/15-4
200
7-15-3
7-15-8
7/15-9
Q.
Q.
190
7/15-7
180
8
6
3
7
4
5
0
2
1
(0.30) (0.61) (0.91) (1.22) (1.52) (1.83) (2.13) (2.44)
Staging Height, ft (m)
Figure 10.
-------
ADDENDUM
SUBSCALE TESTS OF COMBUSTION MODIFICATION
FOR STEEL FURNACES
By:
R. L. Tidona, W. A. Carter
and S. C. Hunter
KVB, Inc.
Irvine, California 92714
(The main text of this paper was
published in Volume III, page 274.)
563
-------
COST CALCULATIONS FOR FLUE GAS RECIRCULATION
APPLIED TO STEEL FURNACES - CASE II
If one may assume complete recovery of the flue gas energy in the
recirculated flue gas such that the only losses in the steel furnace system
are the stack lossesr then there will be no fuel penalty associated with the
FGR modification. In this instance, the annual operating costs of 20 percent
FGR in a steel furnace will consist of only the fan electrical and system
maintenance cost elements. The annual operating costs for three heater sizes
assuming full recovery of the heat of the recirculated flue gas is shown in
Table I.
TABLE I. ANNUAL OPERATING COSTS FOR A STEEL
FURNACE HAVING 20 PERCENT FGR (1980 $)
Heater Size 2.9 MW 73.3 MW 147 MW
Fan Electricity $ 400 $3,600 $7,700
Maintenance 1,850 5,400 9,400
Total $2,250 $9,000 $17,100
The revised total annualized costs of FGR on steel furnaces are shown in
Table II.
TABLE II. TOTAL ANNUALIZED COSTS OF 20 PERCENT FGR ON
A STEEL FURNACE WITH WASTE HEAT RECOVERY (1980 $)
2.9 MW 73.3 MW 147 MW
$9,516 $30,186 $53,979
The annual N0X reduction capability of the FGR modification and the cost
effectiveness (equal to the total annualized cost divided by the annual nox
reduction) are given in T&ble III.
564
-------
TABLE III. ANNUAL NC>X EMISSION REDUCTION AND COST
EFFECTIVENESS OF FGR (1980 $)
2.9
MW
73.3
MW
147
MW
Heater Size
No. 2 Oil
NG
No. 2 Oil
NG
No. 2 Oil
NG
Annual NOx Emission
Reduction Capability
(Mg/y)
8.7
7*5
218
186
437
374
Cost Effectiveness
1094
1269
138
162
124
144
($/Mg)
Note that if the assumption of total waste heat recovery from the
recirculated flue gas is valid (as could be the case if the flue gas for FGR
cones from the outlet of a waste heat boiler or a regenerative heat
exchanger), then flue gas recirculation becomes more attractive than either
the steam or water injection modifications for larger heater sizes (73.3 MW
and up). For small heaters, however, it is not as cost effective as steam
injection with either No. 2 oil or natural gas fuels nor is it as cost
effective as water injection firing natural gas fuel*
The original Tables VII, VIII, and X in this report have been recently
revised and the updated versions sure shown below* The cost effectiveness of
flue gas recirculation with full heat recovery may be compared to the cost
effectiveness calculations in the revised Table X.
When the cost effectiveness values in Table X (Rev*) and Table III are
plotted on log-log paper against heater size, one finds that for No* 2 oil
fuel FGR with heat recovery capability has the lowest cost per Mg N0X
reduction for heaters larger than 13*5 MW (46 x 10® Btu/h). For natural gas
fuel, FGR with heat recovery becomes cheaper than water injection for heater
sizes greater than 8*5 MW (29 x 10® Btu/h) and it is cheaper than steam
injection at heater sizes in excess of 60 MW (205 x 10® Btu/h)*
565
-------
TABLE VII (REV.) TOTAL ANNUALIZED COSTS OF WATER OR STEAM INJECTION
Annual Operating Cost $3,100/1,600 $81,000/39,300 $159,500/83,000
(No. 2/Natural Gas)
State and Federal Taxes 385 2,090 3,520
(11% of IPC)
Insurance (0.5% of IFC) 18 95^ 160
Total Annual Expenses $3,503 $83,185 $163,180
(Mo. 2)
Total Annual Expenses $2,003 $41,485 $86,680
(Natural Gas)
INITIAL FIXED COSTS (IFC) 3,500 19,000 32,000
(WATER OR STEAM)
ROR=i=15%,n=12
Capital Recovery
Factor®.1845»CR
Annual Income
Tax Rate=50%
Investment Tax
Credit=10%=i
c
(1st year only)
Total Annual
Capital Factor*
(ACF)-.2773
Annual Capital
Charge (=IFCxACF) 971 5,269 8,875
TOTAL ANNUALIZED COSTS
(1980 DOLLARS)
No. 2 Oil 4,474 88,454 172,055
Natural Gas 2,974 46,754 95,555
1 La
*ACF = CR + T (CR - ~)
n n
where CR = capital recovery factor
1-(1+i)~n
and T = 1.0 (for debt/equity ratio of 0)
566
-------
TABLE VIII (REV.) TOTAL ANNUALIZED COSTS OF FGR
Annual Operating Costs 37,630/19,350 894,000/437,000 1,787,100/873,100
(No. 2/Natural Gas)
State and Federal Taxes 2,035 5,940 10,340
(11% of IFC)
Insurance (0.5% of IFC) 100 270 470
Total Annual Expenses $39,765 $900,210 $1,797,910
(No. 2)
Total Annual Expenses $21,485 $443,210 $883,910
(Natural Gas)
INITIAL FIXED COSTS 18,500 54,000 94,000
ROR=i=15%,n=12
Capital Recovery
Factor*.1845
Annual Income Tax
Rate=t=50%
Investment Tax
Credit-i =10%
Q
(1st year only)
Total Annual
Capital Factor
-.2773
, Annual Capital
Charge 5,131 14,976 26,069
TOTAL ANNUALIZED COSTS
(1980 DOLLARS)
No. 2 Oil 44,896 915,186 1,823,979
Natural Gas 26,616 458,186 909,979
567
-------
TABLE X (REV.) COST EFFECTIVENESS OF COMBUSTION MODIFICATIONS
ON A STEEL FURNACE ($/103 Kg OF NOx REDUCTION)
INCLUDING ANNUAL INCREMENTAL FUEL COSTS
ui
cn
oo
Modi f icat ion
STEAM INJECTION
No. 2 Oil
NG
HATER INJECTION
No. 2 Oil
NG
FLUE GAS RECIRCULATION
No. 2 Oil
NG
Furnace Heat Input
2.9 MW (10x106 Btu/hr) 73.3 MW (250x106 Btu/hr) 147 MW (500x106 Btu/hr)
443
294
1,119
744
5,160
3,549
351
186
893
472
4,198
2,463
340
189
860
478
4,174
2,433
-------
LIST OF ATTENDEES
JOINT SYMPOSIUM ON STATIONARY COMBUSTION NOx CONTROL
Denver, Colorado
October 6-9, 1980
Abbasi, Hamid
Institute of Gas Technology
A201 W 36th Street
Chicago, IL 60632
Adams, Gregory M.
LA County Sanitation Districts
1955 Workman Mill Road
Whittier, CA 90601
Albertson, Walt
Union Oil Research
P.O. Box 76
Brea, CA 92621
Allen, John M.
Battelle Columbus Labs
505 King Avenue
Columbus, OH 43201
Antil, James A.
PPG Industries
One Gateway Center
Pittsburgh, PA 15222
Aoki, N.
I.H.I. Toyo Office
Toyo 5-30-13 Koto-Ku,
Tokyo 135, JAPAN
Asay, Blaine
Brigham Young University
327 CB
Provo, UT 84602
Aure, Tyrone
LA Dept. of Water Power
111 North Hope Street
Los Angeles, CA 90051
Axtman, William H.
American Boiler Manufacturers Assn.
Suite 700, AM Building
1500 Wilson Blvd.
Arlington, VA 22209
569
Baird, R. K.
Union Oil Company of California
Los Angeles Refinery
P. 0. Box 758
Wilmington, CA 90744
Barkley, Joe B.
Tenn. Valley Authority
1120 Chestnut St. Towers II
Chattanooga, TN 37402
Barrow, E. T.
Ministry of the Environment
880 Bay Street
Toronto, Ontario
CANADA M5S 1Z8
Barsin, Joseph A. C.
Babcock & Wilcox Co.
20 S. Van Buren Ave.
Barberton, OH 44203
Baublis, Daniel C.
Babcock & Wilcox
1 California St. #1100
San Francisco, CA 94111
Baur, Fred
Met-Pro Corp, Systems Div.
P.O. Box 144
Harleysville, PA 19438
Beer, Janos M.
Massachusetts Inst, of Tech.
Cambridge, MA 02139
Bell, Colin
CEA Combustion Limited
East Street
Portchester, Hampshire,
ENGLAND P016 9RD
Bell, Doug
Environmental Protection Agency
6209 Summerfield Drive
Durham, NC 27712
-------
Bemis, Gerry
California Energy Comm.
1111 Howe Ave. (MS-30)
Sacramento, CA 95825
Benson, Chas. E.
Exxon Res. & Eng.
P.O. Box 101
Florham Park, NJ 07932
Bland, Verle V.
KVB, Inc.
3131 Briarpark, Suite 250
Houston, TX 77042
Bley, Bruce C.
Combustion Engineering
1000 Prospect Hill Road
Windsor, CT 06095
Bojko, Rita
Institute of Gas Technology
3424 South State Street
Chicago, IL 60616
Bowen, Joshua S.
Environmental Protection Agency
Combustion Research Branch (MD-65)
Research Triangle Park, NC 27711
Bradford, Willis P.
Amax Environmental Srvs., Inc.
4704 Harlan Street
Oenver, CO 80212
Bradley, P.D.
So. California Gas. Co.
810 S. Flower Street
Los Angeles, CA 90017
Bradshaw, R.
University of Utah
Salt Lake City, UT
Brady, Hugh
American Gas Assn.
1515 Wilson Blvd.
Arlington, VA 22209
Brandt, E. F.
Hydro-Sonic Systems
807 Campbell Centre
11/8150 N. Central Expwy.
Dallas, TX 75206
Bray, Chuck
Occidental Oil Shale, Inc.
P.O. Box 2687
Grand Junction, CO 81502
Breen, Bernard P.
Research-Cottrell
Energy Technology & Projects
18004 Skypark Blvd., Suite 150
Irvine, CA 92714
Broer, W. T.
N.V. Nederlandse Gasunie
P.O. Box 19
Groningen, NETHERLANDS 9700 MA
Brower, Frank M.
The Dow Chemical Co.
2030 Dow Center
Midland, MI 48640
Bruce, Steven R.
TOSCO Corporation
10100 Santa Monica Blvd.
Los Angeles, CA 90067
Bumstead, Ron
Northeast Utilities Service Co.
P.O. Box 270
Hartford, CT 06101
Burke, Jack
Radian Corporation
P.O. Box 9948
Austin, TX 78766
Burns, Eugene A.
Systems, Science & Software
P.O. Box 1620
La Jolla, CA 92038
Campobenedetto, Edward J.
Bab cock & Wilcox Co.
20 S. Van Buren Avenue
Barberton, OH 44203
570
-------
Candelaria, Robert B.
Sr. Environmental Engineer
P.O. Box W
Page, AZ 86040
Caretto, L.S.
California Air Resources Bd.
13935 Chandler Blvd. (H)
Van Nuys, CA 91401
Carmine, Benjamin C. Ill
Houston Lighting & Power Co.
P.O. Box 1700
Houston, TX 77001
Carter, Wallace A.
KVB, Inc.
18006 Skypark Blvd.
P.O. Box 19518
Irvine, CA 92714
Castaldini, Carlo
Acurex Corporation
485 Clyde Avenue
Mountain View, CA 94042
Chang, Charles S.
TOSCO Corporation
10100 Santa Monica Blvd.
Los Angeles, CA 90067
Chapman, Kirk S.
Coen Company
1510 Rollins Road
Burlingame, CA 94010
Chen, S. L.
Energy & Environmental Research
8001 Irvine Blvd.
Santa Ana, CA 92705
Child, Huntley
Western Fuelsav-r Corp.
P.O. Box 20432
Billings, MT 59104
Christiano, John P.
National Park Service
NPS-AIR
P.O. Box 25287
Denver, CO 80225
571
Chu, Hung
Los Angeles Department of Water & Power
111 North Hope Street
Los Angeles, CA 90051
Cichanowicz, Ed
Electric Power Research Institute
3412 Hillview Avenue
Palo Alto, CA 94303
Clark, James M., Ill
Western Precipitation Division
Joy Manufacturing Co.
P.O. Box 2744
Los Angeles, CA 90051
Clark, Wyman
Energy & Environmental Research
8001 Irvine Blvd.
Santa Ana, CA 92705
Claudin, Shelley J.
Caterpillar Tractor Co.
100 N.E. Adams
Tech Center-E
Peoria, IL 61629
Cleveland, Joseph J.
GTE Products Corporation
Box 70
Towanda, PA 18848
Coe, E. L.
Joy Industrial Equipment
4565 Colorado Blvd.
P.O. Box 2744 Term Annex
Los Angeles, CA 90051
Cofield, W. W.
Transco Companies, Inc.
P.O. Box 1396
Houston, TX 77001
Collette, Bob
Combustion Engineering, Inc.
1000 Prospect Hill Road
Windsor, CT 06095
Courtney, A1
Commonwealth Edison Co.
P.O. Box 767
Chicago, IL 60690
-------
Cress, W. R.
Allegheny Power Service Corp.
Cabin Hill Dr,
Ggeensburg, PA >5601
Crowell, Barbara A.
Weyerhaeuser Co.
Weyerhaeuser Tech. Ctr.
WTC 1B36
Tacoma, WA 98477
Cvicker, John S. Dept. 177
Foster Wheeler Energy Corp.
9 Peach Tree Hill Road
Livingston, NJ 07039
D'Allessandro, Alfred
Johnson Matthey, Inc.
1401 King Road
Westchester, PA 19380
Damon, James
Stearns-Roger
P.O. Box 5888
Denver, CO 80217
Davies, Ted
North American Mfg. Co.
4455 E. 71st Street
Cleveland, OH 44105
Davis, Sam
Chevron
8 Duffy Court
Pleasant Hill, CA 94523
Dawson, Charlene A.
Conoco, Inc.
1000 So. Pine
Ponca City, OK 74601
De Voe, J. M.
Allied Chemical
P.O. Box 1139R
Morris town, NJ 07960
De Zubay, E.
Westinghouse R&D
1310 Beulah Road
Pittsburgh, PA 15235
Delacy, John M.
Coen Company
1510 Rollins Road
Burlingame, CA 94010
Demi an, Atef
Chemico Air Pollution Control
2101 Tompkins Ave. #C-2
Albany, GA 31705
Destefano, James
PPG Industries
One Gateway Ctr.
Pittsburgh, PA 15222
Dhawan, Arun K.
EID, Air Quality Bureau
State of New Mexico
P.O. Box 968 Crown Bldg.
Santa Fe, NM 87503
Doty, Jane
Chevron Research Co.
576 Standard Ave.
Richmond, CA 94802
Downey, F. Kent
AMAX Environmental Services,
4704 Harlan Street
Denver, CO 80212
Drissel, Geoffrey
Stearns-Roger
P.O. Box 5888
Denver, CO 80217
Duran, Sam
Getty Oil Co.
Rt-1, Box 197X
Bakersfield, CA 93308
Dykema, Owen W.
Rockwell International
8900 DeSoto
Canoga Park, CA 91304
Ebel, Robert H.
American Cyanamid
1937 West Main Street
Stamford, CT 06904
572
-------
Echter, Dana
Arco Coal Co.
4236 Jellison Street
Wheatridge, CO 80033
Eldridge, John W.
University of Mass.
Ch. E. Dept.
Amherst, MA 01002
Ellison, Wm.
NUS Corporation
4 Research Place
Rockville, MD 20850
England, Glenn
Energy & Environmental Research
8001 Irvine Blvd.
Santa Ana, CA 92705
Etter, Roger S.
Atlantic Richfield Co.
400 E. Sibley Blvd.
Harvey, IL 60430
Evans, Michael
Acurex Corporation
485 Clyde Avenue
Mt. View, CA 94042
Evers, Theo
Netherlands Embassy
4200 Linnean Ave., NW
Washington, DC 20008
Faist, Suzan M.
Mobil Oil Corporation
Billingsport Road
Paulsboro, NJ 08066
Farmer, R. C.
Science Applications, Inc.
21133 Victory Blvd., Suite 216
Canoga Park, CA 91303
Farrar, Mike A.
Chevron Research Co.
576 Standard Avenue
Richmond, CA 94802
Ference, Robert A.
Climax Molybdenum Co.
1600 Huron Parkway
Ann Arbor, MI 48106
Finn, Dennis P.
Babcock & Wilcox
777 S. Wadsworth Blvd.
Lakewood, CO 80226
Fir ley, Janet
International Coal Refining Co.
P.O. Box 2752
All en town, PA 18001
Fischer, Jack
Arqonne National Laboratory
9700 S. Cass Ave.
Argonne, IL 60439
Fleming, Edward S.
General Elec. Co., Rm L9505
P.O. Box 8555
Philadelphia, PA 19101
Folsom, Blair
Energy & Environmental Research
8001 Irvine Blvd.
Santa Ana, CA 92705
Freeberg, Clayton R.
Chevron Research Co.
576 Standard Avenue
Richmond, CA 94802
Freedman, Steven
Department of Energy
Germantown, MD 20545
Freel, John
P & M Coal Company
Englewood, CO 80111
Frey, Donald J.
Combustion Engineering, Inc.
10000 Prospect Hill Road
Windsor, CT 06095
Gage, Stephen
Environmental Protection Agency
401 M Street, SW (RD-672)
Washington, DC 20460
Gallaher, David
Standard Oil of California
555 Market Street
San Francisco, CA 94112
-------
Galluzzo, N.G.
Black & Veatch Consulting Engrs.
P.O. Box 8405
Kansas City, MO 64114
Garelick, Barry
Woodward-Clyde Consultants
3 Embarcadero Center, Suite 900
San Francisco, CA 94111
Gasperecz, Greg
Louisiana Air Quality Div.
P.O. Box 44066
Baton Rouge, LA 70804
Gay, Richard L.
Rockwell International
8900 DeSoto Avenue
Canoga Park, CA 91304
Geren, P. M.
Air Correction Div.
Universal Oil Products, Inc.
101 Merritt-7
Norwalk, CT 06856
Giammar, Robert D.
Battelle-Columbus Labs
505 King Avenue
Columbus, OH 43201
Giovanni, Dan
Electric Power Research Institute
3412 Hillview Avenue
Palo Alto, CA 94303
Gladden, John R.
Caterpillar Tractor Co.
100 N.E. Adams Street
Peoria, IL 61629
Glazer, Jerome L.
Air Products & Chemicals, Inc.
P.O. Box 538
Allentown, PA 18105
Goalwin, Daniel
Bay Area AQMD
939 Ellis Street
San Francisco, CA 94109
Goodley, Alan
California Air Resources Bd.
P.O. Box 2815
Sacramento, CA 95812
Gow, Roland
Colorado Interstate Gas Co.
P.O. Box 1087
Colorado Springs, CO 80944
Green, George
Public Service Company of Colorado
550 15th Street
Denver, CO 80202
Greene, Jack H.
Environmental Protection Agency
IERL-RTP (MD-60)
Research Triangle Park, NC 27711
Greenfield, Stan P.
TRW DSSG
One Space Park
Bldg. 01, Rm 1220
Redondo Beach, CA 90278
Grewal, Lakhmir
S.J.C. Air Pollution Control Dist.
P.O. Box 2009
1601 E. Hazelton Ave.
Stockton, CA 95201
Griffith, Robert E.
Peadoby Engineering-Combustion
Products Division
39 Maple Tree Avenue
Stamford, CT 06906
Grimm, R. Paul
Stearns-Roger
P.O. Box 5888
Denver, CO 80217
Gulley, Pam
PNM (Public Service Co. of NM)
P.O. Box 2267
Albuquerque, NM 87103
Hall, Robert E.
Environmental Protection Agency
Combustion Research Branch (MD-65)
Research Triangle Park, NC 27711
Hangebrauck, Robert P.
Environmental Protection Agency
Energy Assessment & Control Div.
(MD-61)
Research Triangle Park, NC 27711
-------
Harris, Leonard A.
Kerr-McGee Chem. Corp.
Box 367
Trona, CA 93562
Higginbotham, E. B.
Acurex Corporation
485 Clyde Avenue
Mt. View, CA 94042
Hillenbrand, Louis
Battelle Columbus Labs
531 Brookside Drive
Columbus, OH 43209
Hinkamp, James B.
Ethyl Corp. Chem. Res.
1600 West Eight Mile Road
Ferndale, MI 48220
Hinrichs, James M.
San Diego Gas & Elec. Co.
1348 Sampson St.
San Diego, CA 92113
Holtz, Don
Engineering Science
125 W. Huntington Dr.
Arcadia, CA 91106
Hood, Kenneth T.
NCASI of the Paper Industry
Engr. Exp. St. (OSU)
Corvallis, OR 97331
Horwitz, Judy
UOP
10 UOP Plaza
Des Plaines, IL 60616
Howell, Brooks, M.
Combustion Engineering, Inc.
1000 Prospect Hill Road
Windsor, CT 06095
Hsieh, C.K.
Southern California Edison
P.O. Box 800
Rosemead, CA 91770
Huang, Hann S.
Argonne National Labs
9700 S. Cass Avenue
Argonne, IL 60439
Hubble, B. R.
Argonne National Laboratory
Bldg. 362
9700 South Cass Avenue
Argonne, IL 60439
Hubickey, W. D.
Process Combustion Corporation
1675 Washington Road
Pittsburgh, PA 15228
Hunter, S. C.
KVB, Inc.
P.O. Box 19518
Irvine, CA 92714
Hurst, Boyd E.
Exxon Res. & Engr. Co.
P.O. Box 101
Florham Park, NJ 07932
Isenberg, Jerrold
Joy Mfg. Co.
4565 Colorado Blvd.
Los Angeles, CA 90039
Jastrzebski, Richard
Consolidated Edison
4 Irving Place
New York, NY 10003
Johansson, Eddy
Swedish State Power Board
Racksta, 162 8F Vallingby
Stockholm, SWEDEN
Johnson, Neil H.
Detroit Stoker Co.
1510 E. First St.
Monroe, MI 48161
Johnson, Roger
York Research Consultants
938 Quail Street
Denver, CO 80215
Johnson, Stephen A.
Research & Development Div.
Bab cock & Wilcox
1562 Beeson Street
Alliance, OH 44601
575
-------
Jones, Gary D.
Radian Corp.
P.O. Box 9948
Austin, TX 78766
Jones, Michael H.
Environmental Protection Agency (MD-
Strategies and Air Standards Div.
Research Triangle Park, NC 27711
Jones, Randall A.
Standard Oil (Ohio)
Midland Bldg Loc 438 CB
Cleveland, OH 44115
Jordan, Richard J.
FLUOR E&C, Houston
P.O. Box 35000
Houston, TX 77035
Joubert, James I.
U. S. Dept. of Energy/PETC
P.O. Box 10940
Pittsburgh, PA 15236
Kajibata, Yoshihiro
Kawasaki Heavy Industries, Ltd.
1-1, Kawasaki-cho
Akashi-shi, JAPAN
Kasischke, Martin W.
Hydro Sonic Systems
807 Campbell Centre II
8150 North Central Expressway
Dallas, TX 75206
Kau, Ed
Energy & Environmental Research
8001 Irvine Blvd
Santa Ana, CA 92705
Kawamura, Tomozuchi
Mitsubishi Heavy Industries, Ltd.
852-5 Hata-cho
Chiba, JAPAN
Kawashima, K.
Sakai Chemical Co., Ltd.
c/o Sakai Trading New York, Inc.
417 Fifth Avenue
New York, NY 10016
576
Keith, George
Babcock & Wilcox
Iron gate Exec. Plaza I
Suite 200
777 S. Wadsworth Blvd.
Lakewood, CO 80226
Keller, James
Stearns-Roger
P.O. Box 5888
Denver, CO 80217
Kelly, John
Acurex, Corporation
485 Clyde Avenue
Mountain View, CA 94042
Kemp, Fred
Power Systems Division
United Technologies Corp.
P.O. Box 109
South Windsor, CT 06074
Kerho, Stephen E.
KVB, Inc.
18006 Skypark Blvd.
Irvine, CA 92714
Kesselring, John P.
Acurex Corporation
485 Clyde Avenue
Mountain View, CA 94042
Kikuchi, K
Sakai Chemical Co., Ltd.
c/o Sakai Trading New York, Inc.
417 Fifth Avenue
New York, NY 10016
Kliegel, James R.
KVB, Inc.
18006 Skypark Blvd.
Irvine, CA 92714
Knowles, Joan
Crown Zellerbach
904 NW Drake Street
C amas, WA 98607
Kobayashi, Sho
Union Carbide Corporation
Old Sawmill Road
Tarrytown, NY 10591
-------
Koda, Hiromasa
Kawasaki Heavy Industries, Ltd.
1900 Avenue of the Stars, Room 1165
Los Angeles, CA 90067
Kojima, Hideo
c/o Hitachi America, Ltd.
437 Madison Avenue
New York, NY 10022
Kothari, Vijay
Department of Energy
Morgantown Energy Technology Center
P.O. Box 880
Morgantown, WV 26505
Kramlich, John
Energy & Environmental Research
8001 Irvine Blvd.
Santa Ana, CA 92705
Kress, P. Joseph
Ball Corporation
1509 S. Macedonia Ave.
Muncie, IN 47302
Kressl, Frank
Gotaverken Angteknik AB
Box 8734
S-402 75
Gateborg, SWEDEN
Kuroda, H.
Babcock Hitachi, Kure Works
609 Takara cho
Kure City, Hiroshima
JAPAN
LaRue, Albert D.
Babcock & Wilcox
20 S. Van Buren
Barberton, OH 44203
Lachapelle, David G.
Environmental Protection Agency
Industrial Environmental Research Lab
Combustion Research Branch, MD-65
Research Triangle Park, NC 27711
Lange, Howard
KVB, Inc.
P.O. Box 19518
Irvine, CA 92714
Latchem, Ken
Gifford-Hill Cement
P.O. Box 520
Midlothian, TX 76065
Layman, George 0.
Gulf Power Company
P.O. Box 1151
Pensacola, FL 32510
Leavitt, Julian J.
American Cyanamid Co.
Berdan Avenue
Wayne, NJ 07470
Lee, David C.
Clark Co. Health District
Air Pollution Control Div.
625 Shadow Lane
Las Vegas, NV 89106
Leivo, Charles C.
Dresser Industries
2408 Timberloch-Building C
The Woodlands, TX 77380
Leo, Paul
Aerospace Corporation
P.O. Box 92957
Los Angeles, CA 90274
Leppa, Kalevi
Ekono, Inc.
410 Bellevue Way, SE
Bel levue, WA 98004
Lester, Thomas W.
Environmental Protection Agency
Industrial Environmental Research Lab
Combustion Research Branch (MD-65)
Research Triangle Park, NC 27711
Levy, Arthur
Battelle Columbus Labs
505 King Avenue
Columbus, OH 43201
Lew, Henry G.
Westinghouse Electric Corp.
P.O. Box 251
Concordville, PA 19331
-------
Lewis, Julius P.
Mitre Corporation
1820 Dolley Madison Blvd.
McLean, VA 22102
Lim, Charles T.
Norton Co.
P.O. Box 350
Akron, OH 44309
Lim, Kenneth
Acurex Corporation
485 Clyde Avenue
Mountain View, CA 94042
Linke, William
American Cyanamid
1937 W. Main Street
Stamford, CT 06904
Lipfert, F. W.
Brookhaven National Labs
Upton, NY 11973
Lipscomb, Jay
Mostardi-Platt, Inc.
1077 Entry Drive
Bensenville, IL 60106
Lisauskas, Robert A.
Riley Stoker Corp.
P.O. Box 547
Worcester, MA 01613
Loblich, Hans
Consultant
Krietkamd 38
2 Hamburg 65
W. GERMANY
Luken, Ralph A.
Environmental Protection Agency
401 M Street, S.W.
Washington, DC 20460
Mace, Fred
Texaco, Inc.
2101 E. Pacific Coast Hwy
Wilmington, CA 90748
Maloney, K. L.
KVB, Inc.
18006 Skypark Blvd.
Irvine, CA 92714
Manny, E. H.
Exxon Research & Engineering
P.O. Box 101
Florham Park, NJ 07932
Mansour, M. N.
KVB, Inc.
18006 Skypark Blvd
Irvine, CA 92714
Marshal 1, John J.
Riley Stoker Corp.
P.O. Box 547
Worcester, MA 01613
Martin G. Blair
Environmental Protection Agency
Industrial Environmental Research Lab
MD-65
Research Triangle Park, NC 27711
Mason, Howard
Acurex Corporation
485 Clyde Avenue
Mountain View, CA 94042
Massoudi, M.S.
Teknekron Research, Inc.
2118 Milvia Street
Berkeley, CA 94704
Maxwell, J. D.
Tennessee Valley Authority
Energy Demonstration & Technology
501 CEB
Muscle Shoals, AL 35660
McElroy, Michael
Electric Power Research Institute
3412 Hillview Avenue
Palo Alto, CA 94303
McRanie, Richard
Southern Company Service
P.O. Box 2625
Birmingham, AL 35206
Mehl, Carl A.
Mobil Oil Corporation
3700 West 190th Street
Torrance, CA 90 509
-------
Mehta, Arun K.
Combustion Engineering, Inc.
1000 Prospect Hill Road
Windsor, CT 06095
Meier, John G.
Solar Turbines International
2200 Pacific Hwy
P.O. Box 80966
San Diego, CA 92138
Mellor, A. M.
Purdue University
School of Mechanical Engineering
West Lafayette, IN 47907
Merril 1, Austin
A. H. Merrill & Associates
24 California Street
San Francisco, CA 94111
Michelfelder, Sigfrid
Steinmueller
Postfach 1949/1960
D-5270 Gummersbach 1
W. GERMANY
Monacelli, John E.
Babcock & Wilcox
1385 Girard Street
Akron, OH 44301
Moore, Berkley L.
Environmental Protection Agency
2200 Churchill Road
Springfield, IL 62706
Morii, A.
Mitsubishi Heavy Industries
5-34-6 Shiba Minato-ku
Tokyo, JAPAN
Morrison, Geoffrey F.
IEA Coal Research
14/15 Lower Grosvenor Place
London SW1W0EX
ENGLAND
Morsing, Per
Niro Atomizer A/S
Gladsaxevej 305
Soeborg, DENMARK DK-2860
Morton, William
E. Keeler Co.
238 West Street
Williamsport, PA 17701
Mouri, Konihiko
Electric Power Development Co.
1-8-2 Marunouchi Chiyoda Ku
JAPAN
Mozes, Miriam S.
Ontario Hydro
Room KR230
800 Kipling Avenue
Toronto, Ontario
CANADA M8Z5S4
Middleton, Daryl J.
Ford Motor Company
Glass Division
25500 West Outer Drive
Lincoln Park, MI 48146
Miller, Michael J.
Electric Power Research Institute
3412 Hillview Avenue
Palo Alto, CA 94303
Miron, R.L.
Shell Oil Company
196 S. Fir Street
Ventura, CA 93001
Misa, Winston E.
Procon, Inc.
9650 Flair Drive
El Monte, CA 91731
Mobley, J. David
Environmental Protection Agency
Industrial Environmental Research Lab
MD-61
Research Triangle Park, NC 27711
Mulder, W.
Ministry of Health & Environmental Protecl
Kiggelaerstraat 15
2596 TL DEN HAAG
NETHERLANDS
Munro, James
University of Utah
Salt Lake City, UT
-------
uzio, L. J.
VB, Inc.
8006 Skypark Blvd.
rvine, CA 92714
laulty, Dave
>tearns-Roger
'.0. Box 5888
)enver, CO 80217
^eblett. Lei and
Setty Oil Company
}oute 1, Box 197-X
Bakersfield, CA 93308
Nein, Alex
Purdue University
School of Mechanical Engineering
West Lafayette, IN 47907
Nesbitt, Gregory S.
Brown Roveri Turbomachinery, Inc.
711 Anderson Ave., North
St. Cloud, MN 56301
Nickerson, Greg
Combustion Engineering
1000 Prospect Hill Road
Windsor, CA 06070
Nicol, S. K.
Broken Hill Prop. Co., Ltd.
Chrysler Blvd.
405 Lexington Avenue
41st Floor
New York, NY 10174
Nordheim, Mark W.
Chevron USA
324 W. El Segundo Blvd.
El Segundo, CA 90245
Norton, Dennis
Portland General Electric
121 S. W. Salmon Street
Portland, OR 97204
Novitsky, Walter M.
PA. Power & Light Co.
Two North Ninth Street
Allentown, PA 18101
580
Nuila, Carlos
Dow Chemical
P.O. Bxo 1398
Pittsburg, CA 94565
Nurick, W. H.
Energy & Environmental Research
8001 Irvine Blvd.
Santa Ana, CA 92705
Oglesby, Scott
NCASI
P. 0. Box 14483
Gaines, FL 32604
Orloff, H. D.
Ethyl Corporation
1600 West Eight Mile Road
Ferndale, MI 48220
Overduin, Cornelis L.
Southern California Edison Co.
2244 Walnut Grove Ave.
Rosemead, CA 91770
Oxley, Joseph H.
Battelle Columbus Labs
505 King Avenue
Columbus, OH 43214
Packham, H.
Jacksonville Electric Authority
P.O. Box 53015
Jacksonville, FL 32201
Palomba, Joseph, Jr.
Air Quality Control Commission
Colorado Dept. of Health
4210 East Uth Avenue
Denver, CO 80220
Palomino, G. E.
Salt River Project
P.O. Box 1980
Phoenix, AZ 85283
Parks, Robert M.
Radian Corporation
Durham, NC 27705
Perlsweig, Michael
Department of Energy
Office of Coal Utilization SE-22
Germantown Office MS E-178
Washington, DC 20545
-------
Pershing, David M.
University of Utah
Salt Lake City, UT 84112
Pietruszkiewic, Jon
Bechtel Corporation
50 Beale Street
San Francisco, CA 94119
Nog, John
Colorado Department of Health
4210 E. 11th Avenue
Denver, CO 80220
Pohl, John H.
Sandia National Laboratories
Combustion Research Division 8353
Livermore, CA 94550
Pohlenz, Jack B.
UOP
10 UOP Plaza
Des Plaines, IL 60616
Ponder, Made H.
Environmental Protection Agency
Industrial Environmental Research Lab
MD-62
Research Triangle Park, NC 27711
Potterton, S. T.
Babcock & Wilcox
4282 Strausser St., N. W.
North Canton, OH 44720
Preston, James
Tenneco, Inc.
P.O. Box 2511
Houston, TX 77001
Prokopuk, R.
EMR, CCRW
555 Booth Street
Ottawa, Ontario, CANADA K14 0G1
Protheroe, D.
Rolls Royce (Canada)
P.O. Box 1000
Montreal A.M.F.
Montreal, Quebec, CANADA H4Y1B7
Pruce, Leslie
Power Magazine
1221 Avenue of the Americas
New York, NY 10020
Purcell, Steven
University of Utah
Salt Lake City, UT 84112
Rabin, Irwin A.
IAR Technology, Inc.
130 Sandringham South
Moraga, CA 94556
Radak, Les
Southern Calif. Edison
P.O. Box 800
Rosemead, CA 91770
Rarick, Tom
Environmental Protection Agency
215 Fremont Street
San Francisco, CA 94105
Rawdon, A.H.
Riley Stoker Corp.
P.O. Box 547
Worcester, MA 01613
Rees, Dee P.
Utah Power & Light
Research and Development Dept.
P.O. Box 899
Salt Lake City, Utah 84110
Richter, Wolfgang
Energy & Environmental Research
8001 Irvine Blvd.
Santa Ana, CA 92680
Robinson, Jerry
KVB, Inc.
18006 Skypark Blvd.
Irvine, CA 92714
Roffe, Gerald
BASL
Merrick & Stewart Avenues
Westbury, NY 11590
Rollbuhler, R. J.
Lewis Research Center of NASA
Mail Stop 86-5
21000 Brookpark Road
Cleveland, OH 44135
-------
Ross, Laurence W.
Consultant
925 Adams Street
Denver, CO 80206
Rovesti, William C.
Electric Power Research Institute
3412 Hillview Ave.
Palo Alto, CA 94303
Rudnicki, Mark I.
Aerojet Energy Conversion Co.
P.O. Box 13222
Sacramento, CA 95813
Rullman, Don
UOP Air Correction Division
101 Merritt-7
Norwalk, CT 06856
Sage, W.L.
Stearns-Roger
P.O. Box 5888
Denver, CO 80217
Saiki, Hiroshi
Cleaver Brooks, Div. of Aqua Chem, Inc.
5100 North 33rd Street
Milwaukee, WI 53201
Sako, Frank
FMC Corporation
1185 Coleman Avenue
P.O. Box 580
Santa Clara, CA 95052
Sannes, Carl
Northern States Power
143 West Pleasant Lake Road
St. Paul, MN 55110
Sarofim, Adel
Massachusetts Institute of Technology
Rm 66-466
Cambridge, MA 02139
Scheck, Robert
Stearns-Roger
P.O. Box 5888
Denver, CO 80217
Schleckser, Charles E.
Exxon Research & Engineering
P.O. Box 101
Florham Park, NJ 07932
Schmidt, George A.
Englehard Industries
2655 U. S. Route 22
Union, NJ 07083
Schultz, Thomas
Midland-Ross Corporation
Technical Center
900 N. Westwood
Toledo, OH 43696
Schuster, A. G.
Northern States Power Co.
100 North Barstow Street
Eau Claire, WI 54701
Schuster, Herbert
Deutsche Babcock AG
Duisburgerstr 375
D4200 Oberhausen
W. GERMANY
Seckington, Blair
Ontario Hydro
700 University Ave.
Toronto, Ontario, CANADA M56 1X6
Seebold, James G.
Standard Oil Company of Calif.
P.O. Box 3069
San Francisco, CA 94119
Seeker, Randall
Energy & Environmental Research
8001 Irvine Blvd. ¦
Santa Ana, CA 92705
Semerjian, Hratch G.
National Bureau of Standards
Building 221, Room B252
Washington, DC 20234
Sengoku, Tadamasa
Mitsubishi Heavy Industries
SHIN-TAMACHI Bldg 34-6
Shiba 5-Chome, Minatoku
Tokyo, JAPAN
582
-------
Sephton, Harold
Indian Head-Madera Glass
Consultant
Madera, CA 93637
Sheehan, J.P.
Stearns-Roger
P.O. Box 5888
Denver, CO 80217
Shenk, Richard
Weyerhauser
Tacoma, WA 98477
Shigehiro, Katsuya
Hitachi America, Ltd.
437 Madison Avenue
New York, NY 10022
Shui, Ven H.
Avco Everett Research Lab, Inc.
2385 Revere Beach Parkway
Everett, MA 02149
Sigal, Lorene L.
Oak Ridge National Laboratory
P.O. Box X
Oak Ridge, TN 37830
Simmons, Gloyd
Montana Energy
P.O. Box 3809
Butte, MT 59701
Simpson, James H.
North American Mfg. Co.
6836 Cranbrook Drive
Brecksville, OH 44141
Skidmore, Terry
Arizona Public Service
P.O. Box 21666
Phoenix, AZ 85036
Slack, Archie V.
SAS Corporation
Sheffield, AL 35660
Slaughter, Michael
University of Utah
Salt Lake City, UT 84112
583
Smith, Lowell L.
KVB, Inc.
3131 Briarpark Drive, Suite 250
Houston, TX 77043
Snow, Eric
Shell Oil Company
P.O. Box 576
Houston, TX 77001
Sommer, Todd M.
Babcock & Wilcox
20 S. Van Buren Avenue
Barberton, OH 44203
Sonnichsen, T. W.
KVB, Inc.
18006 Skypark Blvd.
Irvine, CA 92714
Speronello, Barry
Engelhard Minerals & Chemical Co.
Menlo Park
Edison, NJ 08817
St. Pierre, M. F.
Shell Oil Company
196 S. Fir Street
Ventura, CA 93001
Starley, Gregory
University of Utah
Salt Lake City, UT 84112
Statnick, Robert M.
Environmental Protection Agency
Energy Processes Division (RD-681)
401 M Street, S.W.
Washington, DC 20460
Stenby, Edward W.
Stearns-Roger
P.O. Box 5888
Denver, CO 80217
Stevens, Clark G.
W. R. Grace & Co.
3400 First International Building
Dallas, TX 75270
Stief-Tauch, H. P.
Commission of European Communities
200 Rue de la Loi
Brussels, BELGIUM 1049
-------
Stier, John
Anheuser-Busch Companies, Inc.
727 North 1st Street
St. Louis, MO 63102
Stockdale, Robert
Texaco, Inc.
2101 E. Pacific Coast Hwy
Wilmington, CA 90748
Su, Benjamin Y.
United Engineers & Constructors
100 Summer Street
Boston, MA 02110
Su, Y. P.
Brown & Root, Inc.
P.O. Box 3
Houston, TX 77001
Sweet land, D. B.
CEA Combustion Limited
East Street
Portches te r
Hampshire, ENGLAND P016 9RD
Sybert, Louis
Bechtel National, Inc.
Fifty Beal Street, P.O. Box 3965
San Francisco, CA 94119
Takagi, K.
Sakai Trading New York, Inc.
417 Fifth Avenue
New York, NY 10016
Takahashi, Henry
Hitachi America, Ltd.
437 Madison Avenue
New York, NY 10022
Tamony, Andree
Dow Chemical
Loveridge Road
Pittsburg, CA 94565
Tanaka, Shingo
Hitachi Shipbuilding & Eng. Co., Ltd.
1-1-1, Hitotsubashi
Chiyoda-ku
Tokyo, JAPAN 100
Teixeira, Donald P.
Electric Power Research Institute
3412 Hillview Avenue
Palo Alto, CA 94022
Thompson, A. J.
Southern California Gas Co.
810 S. Flower Street
Los Angeles, CA 90017
Thompson, Richard E.
KVB, Inc.
18006 Skypark Blvd.
Irvine, CA 92714
Thompson, W. E.
Research Triangle Institute
P.O. Box 12194
Research Triangle Park, NC 27709
Tidona, Robert J.
KVB, Inc.
18006 Skypark Blvd, Sox 19518
Irvine, CA 92714
Todo, Yoshinori
Mitsubishi Heavy Industries
Akunoura, Nagasaki
Nagasaki, JAPAN
Travis, Stephen R.
Arizona Public Service Company
P.O. Box 21666
2216 W. Peoria Avenue
Phoenix, AZ 85036
Trayser, David A.
Battelle Columbus Labs
505 King Avenue
Columbus, OH 43201
Truett, Bruce
The Mitre Corporation
1820 Dolley Madison Avenue
McLean, VA 22102
Tso, Arthur
Mobil Research & Development Corp.
P.O. Box 1026
Princeton, NJ 08540
584
-------
Tubbs, Kevin
Exxon Chemical Co.
P.O. Box 271
Florham Park, NJ 07932
Van Der Aa, Randall
Public Service Company of New Mexico
P.O. Box 227
Waterflow, NM 87421
Van Oostveen, A.
Ministry of Health & Environ. Protection
P.O. Box 439
2260 AK Leidschendam
NETHERLANDS
Van der Kooij, Jan
KEMA
Utrechtseweg 310
Arnhem
NETHERLANDS
Varga, G. M.
Exxon Research & Engineering Co.
P.O. Box 101
Florham Park, NJ 07932
Vasquez, Abe A.
Colorado Department of Health
4210 E. 11th Avenue
Denver, CO 80220
Vatsky, Joel
Foster Wheeler Energy Corp.
9 Peach Tree Hill Road
Livingston, NJ 07039
VerShaw, James T.
The Trane Company
3600 Pammel Creek Road
La Crosse, WI 54601
Vogel, Chester A.
Environmental Protection Agency
Industrial Environmental Research Lab
MD-65
Research Triangle Park, NC 27711
Von Kleinsmid, William
Southern California Edison
P.O. Box 800
Rosemead, CA 91770
Vranos, Alexander
United Technologies Research Center
Silver Lane
E. Hartford, CT 06108
Waibel, Richard T.
Institute of Gas Technology
4201 W. 36th Street
Chicago, IL 60632
Warfe, W. A.
Department of the Environment
Place Vincent Massey
12th Floor
Air Pollution Control Directorate
Ottawa, Ontario, CANADA K1A1C8
Wehr, Allan G.
Mississippi State University
P 0 Rnx AW
Mississippi State, MS 39762
Weisel, Kenneth A.
San Diego Gas & Electric
P.O. Box 1831
San Diego, CA 92112
Wendt, Jost 0. L.
University of Arizona
Department of Chemical Engineering
Tucson, AZ 85721
White, James H.
Coen Company
1510 Rollins Road
Burlingame, CA 94010
Wiener, Richard
Env i rotech/C hemi co
349 E. 49th Street
New York, NY 10017
Williams, Roger
Environmental Protection Agency
1860 Lincoln Street
Denver, CO 80295
Willson, Ernest J., Jr.
United Technologies Corporation
10 Farm Springs
Farmington, CT 06032
585
-------
Wilson, Robert P., Jr.
Arthur D. Little, Inc.
Acorn Park
Cambridge, MA 02140
Winkler, Philip W.
Envirotech/Chemido
1 Penn Plaza
New York, NY 10001
Winter, Robb
University of Utah
Department of Chemical Engineering
Salt Lake City, UT 84112
Wipf, Edward H.
Koppers
P.O. Box 21649
Denver, CO 80221
Wood, Jim
Salt River Project
P.O. Box 1018
Saint Johns, AZ 85936
Wright, Dennis L.
Texas Electric Service Co.
P.O. Box 8368
Fort Worth, TX 76112
Wu, Muh-cheng Milton
Conoco Coal Development Co.
Research Division
Library, PA 15129
Young, Wil1iam N
United Gas Pipe Line Company
P.O. Box 1478
Houston, TX 77001
Yu, Kar Y.
TRW-Env ironmental Eng. Oiv.
One Space Park
Redondo Beach, CA 90278
Yugami, Hiroshi
Electric Power Development Co.
1-8-2 Marunouchi Chiyoda Ku
JAPAN
Zengerle, Monta W.
Electric Power Research Institute
3412 Hillview Avenue
Palo Alto, CA 94303
Zwiacher, Wayne E.
South Coast Air Quality Mgmt. Dist.
9150 Flair Drive
El Monte, CA 91731
Wuerer, Josef
Spectron Development Laboratories, Inc.
3303 Harbor Blvd. Suite G3
Costa Mesa, CA 92626
Wyzga, Ronald
Electric Power Research Institute
3412 Hillview Ave.
Palo Alto, CA 94303
Yang, R. J.
KVB, Inc.
18006 Skypark Blvd.
Irvine, CA 92714
Yokoyama, Naruo
Mitsubishi Heavy Industries
4-Kanonshin-Machi
Hiroshima, JAPAN
-------
TECHNICAL REPORT DATA
(Please read Inunctions on the reverse before completing)
1. REPORT NO.
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Proceedings of the Joint Symposium on Stationary
Combustion NOx Control. Vol. 5. Addendum
5. REPORT DATE
6. PERFORMING ORGANIZATION CODE
7. AUTHORIS)
Symposium Cochairmen: Robert E. Hall (EPA) and
J.E. Cichanowicz (EPRI)
8. PERFORMING ORGANIZATION REPORT NO.
IERL-RTP-1087
9. PERFORMING ORGANIZATION NAME AND ADDRESS
See Block 12.
10. PROGRAM ELEMENT NO.
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; 10/6-9/80
14. SPONSORING AGENCY CODE
EPA/600/13
15. supplementary notes ePA-600/7-79-050a through -050e describe the previous sympo-
sium.
i6 abstract The proceecjingS document the approximately 50 presentations made during
the symposium, October 6-9, 1980, in Denver, CO. The symposium was sponsored
by the Combustion Research Branch of EPA's Industrial Environmental Research
Laboratory, Research Triangle Park, NC, and the Electric Power Research Institute
(EPRI), Palo Alto, CA. Main topics included utility boiler field tests; NOx flue gas
treatment; advanced combustion processes; environmental assessments; industrial,
commercial, and residential combustion sources; and fundamental combustion re-
search. This volume contains papers that were not received in time for inclusion
in the four volumes distributed during the symposium.
KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
b. IDENTIFIERS/OPEN ENDED TERMS
c. COSATi Field/Group
Pollution Flue Gases
Combust^ Engines
Nitrogen Oxides
Boilers
Tests
Assessments
Pollution Control
Stationary Sources
Environmental Assess-
ment
13B
2 IB 2 IK
07B
13A
14B
lUpW*«Tl^kiriON STATEMENT
j*Reteasto to Public
IB. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
591
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
2220.1 (»-73J 587
------- |