Test Design for Environmental Technology Verification (ETV) of Add-on NOX Control
Utilizing Ozone Injection
Douglas VanOsdell and Jack R. Fanner, Research Triangle Institute
P. O. Box 12194, Research Triangle Park, NC 27709
Andrew Trenholm and Craig Clapsaddle, Midwest Research Institute
5520 Dillard Road, Gary, NC 27511 -9232
Theodore G. Brna, Air Pollution Prevention and Control Division, U.S. Environmental
Protection Agency, MD-4, Research Triangle Park, NC 27711
ABSTRACT
The U.S. Environmental Protection Agency (EPA) established the Environmental Technology
Verification (ETV) program to enhance domestic and international market acceptance of new or
improved commercial-ready technologies through independent testing which incorporates
rigorous quality assurance (QA) procedures. Through ETV, potential purchasers and permitters
are provided an independent and credible assessment of what they are buying and permitting.
Operated by the Research Triangle Institute, the Air Pollution Control Technology (APCT)
program is one of 12 ETV pilot programs. Add-on nitrogen oxides (NOX) air pollution control
technologies are among those being verified, and Low-Temperature Oxidation (LTO) is the first
NOX control technology planned for verification by the APCT program,
LTO uses ozone injection into the gas phase of a combustion or chemical process waste gas
stream to oxidize NOX molecules to higher order NOX, which can be removed by wet or semi-dry
scrubbing. Ozone is produced on-site and on-demand from stored oxygen. As a low-temperature
process, it can be installed as an end-of-pipe system. The LTO technology was jointly developed
and patented by Cannon Technology and BOC Gases; LoTOx™ is the trademark of BOC Gases
for large-scale LTO applications.
An ETV test of the LoTOx™ system, as installed on a natural-gas-fired industrial boiler in
southern California, is planned for this year. The primary performance measure being verified is
the NOX emission concentration. Other performance and cost factors are also to be measured,
including inlet NOK concentration, temperature, and flue gas flow rate. Test design and planning
for an ETV test present special challenges. This paper provides an overview of the APCT
program and describes the test planning requirements of the ETV program, as applied to the LTO
technology.
INTRODUCTION
The U. S. Environmental Protection Agency (EPA) has designed the ETV program to verify the
performance of innovative and improved technical solutions to problems that threaten human
health or the environment (EPA, 1997). Managed by EPA's Office of Research and
Development, the ETV program was created to substantially accelerate the entrance of new
environmental technologies into the domestic and international marketplaces. The ETV program
verifies commercial-ready, private sector technologies through 12 pilot programs. (Its web site
is at http://www.epa.gov/etv/.)
-------
The Air Pollution Control Technology (APCT) program (web site at http://etv.rti.org/apct/ ), a
partnership between the Research Triangle Institute (RTI) and the EPA, is one of the 12 pilot
programs. Its primary purpose is to verify the performance of selected commercial-ready air
pollution control technologies using objective and quality-assured data resulting in the
publication of verification statements for the technologies. The verifications will provide state
and local permitters, vendors, and potential purchasers independent and credible verifications of
control technology performance and enhance their marketability. The products of the ETV
process are: generic verification protocols, test data obtained following test/quality assurance
(QA) plans complying with those protocols, verification reports, and verification statements.
The verification statements are brief summaries that describe the technologies and their verified
performance.
APCT Program Operation
The overall project flow for the APCT program is shown in Figure 1. The top part shows the
technology selection process. The APCT program relies upon its Stakeholders Advisory
Committee (SAC), consisting of individuals who have broad knowledge of air pollution issues,
for guidance in many aspects of program operation. SAC activities include recommending
APCT program priorities and technologies to verify. Manufacturers/vendors submit the
technologies to verify and interact with the APCT program and SAC as important technology .
areas are identified and technologies are chosen for verification.
;; tmptementationPIan&'»
Quality Management Piart
Business & • Stakeholders Advisory:» Set Priorities ana Evaluate
,ltelteU^
'
t
,'tasAesi
Meeting(s)
Select Technologies
-Technical Panel(TP}%
The lower (shaded) part of Figure 1
shows the verification testing flow.
Tests are performed following the
guidance of generic verification
protocols (GVPs) developed with
input from technology-area technical
panels (TPs). TPs may include SAC
members, but primarily consist of
other stakeholders who have a
specific interest and knowledge in the
technology area and pertinent to the
verification. A separate site-specific
test/QA plan is prepared for each
technology within a technology area.
Test results are reported as
verification reports and brief
verification statements.
From the viewpoint of the APCT
program, verifying several innovative
and competitive technologies within a
Figure 1. APCT ETV How Chart single technology area is an
advantage because a single GVP may suffice for all. However, since the ETV process is
intended to test innovative technologies and help move them into the environmental marketplace,
TP Meet and Re view Protocol
Generic Verification
ytfntout (sypjAii
'ft r '
/«••*,
process'Evatuation/MocJificatiort
-------
it may be best to test single technologies in some cases.
ETV for Add-on NOX Air Pollution Control
NOX pollution has both direct and indirect human health impacts and contributes to fine .
paiticulate matter in the air. These impacts are widely known and accepted (62 FR 36957, 63 FR
49443). NOX standards and requirements for control have been developed under the Clean Air
Act Amendments of 1990 (EPA, 1991) as Titles I (attainment and maintenance), IV (acid
deposition), and V (permits), serving as regulatory drivers for significant decreases in NOX
emissions.
NO, emissions largely originate in combustion processes, including boilers and fuel-fired
industrial processes. NOX emissions can be reduced by modifying the combustion process or by
removing or converting the NOX in the combustion gases before discharge. Selective catalytic
reduction (SCR) and selective non-catalytic reduction (SNCR) are processes in commercial use
to remove NOX from combustion stack gas. Hybrids of SCR and SNCR are also being
implemented. Less applied and developed technologies include selective injection reduction
utilizing titanium dioxide (TiO2) and ultraviolet (UV) light and various scrubbing and sorption
technologies. One new add-on NOX technology utilizes ozone injection and low-temperature
oxidation of NOX to dinitrogen pentoxide (N2O5), which is soluble in water and can be scrubbed
from the gas stream. Some of the technologies have potential application for combined control
of NOX with sulfur oxides (SO,), carbon monoxide (CO), particles, or their combinations.
Add-on NOX control technologies were recommended to and endorsed by the SAC as a high
priority technology area for the APCT program. For simplicity, the add-on NOX controls
verification focused initially on natural-gas-fired boilers and the ability of the technologies to
control only NOX.
Identification of Add-on NO^ Control Technologies for Verification. Technologies to be tested
in the Add-on NOX Control ETV program must be commercialized and are identified primarily
from those that manufacturers/vendors have offered directly to the APCT EXV program. The
vendor must be willing to participate in the program and must have an installed system or pilot
unit that is available for testing.
As with the other ETV APCT verifications, a verification test under the Add-on NOX Control
ETV program is partially supported by the manufacturer/vendor. In addition, the
manufacturer/vendor is expected to provide the equipment to be tested in a ready-to-run state,
fully installed at the test site, checked out, and with required operating staff and support.
Ozone-injection based technologies were among those first submitted to the APCT program for
verification, and these became the initial focus of the Add-on NOX Control ETV program. The
two technologies submitted are those promoted by BOC /Cannon (LoTOx™/ LTO) and by
Comply Technologies (Comply 2000), The LTO process is commercialized, the vendor is
willing to participate, and verification of the LTO process will proceed first, LTO uses ozone
injection into the gas phase of a combustion or chemical process waste gas stream to oxidize NOX
molecules to higher order NOX, which can be removed by wet or semi-dry scrubbing. Ozone is
-------
produced on-site and on-demand from stored oxygen. As a low-temperature process, it can be
installed as an end-of-pipe system. The LTO technology was jointly developed and patented by
Cannon Technology and BOC Gases; LoTOx™ is the trademark of BOC Gases for large-scale
LTO applications. A verification test of the LTO system as installed on a narural-gas-fired
industrial boiler in southern California is planned for this year.
Other add-on NOX controls (in addition to ozone injection technologies) in various stages of
development have also been submitted to the APCT program. This paper focuses on experience
with the ozone-injection NOX control verification effort.
Generic Verification Protocol
Within the APCT program, numerous technology areas are being verified, and overall are
specified for each. The Generic Verification Protocol (GVP) sets the objectives, general
approaches, and data quality objectives for verification testing within a technology area. In
particular, the GVP sets the data quality objectives that must be met by verification tests within a
technology area. Specific test/QA plans, meeting the requirements of the GVP, are then prepared
for each technology verification. Because of the breadth of technologies that control NOX
emissions, a GVP was prepared for the ozone-injection add-on NOX control technology area. The
draft GVP for ozone-injection add-on NOX control technologies (RTI, 2000) has been posted at
http://etv.rti.org/apct/pdf/addonNOxGVP.pdf. The GVP will remain a draft until it has been
used to verify a technology, its usefulness reviewed, and it has been modified if appropriate.
Verification Test Design
After the test site is identified and approvals are obtained from all interested parties (including
the manufacturer/vendor and the site owner), the test/QA plan for the actual verification tests can
be completed. Prior to beginning testing, a contractual agreement establishing the expectations
of all parties will be completed. Cost share will be required from the vendor/manufacturer. As is
required of any test plan, the plans address all data that will be gathered, a description of the
technology to be tested, project organization and responsibilities, QA objectives, testing and
analytical procedures, instrument calibration, data reduction, reporting, quality control (QC)
checks, audits, and calculations. Each is approved by the APCT program QA Officer and EPA.
Statement of Technology Performance. A verification test is not a test against an emission
standard, but is intended to verify performance claims made by the manufacturer/vendor.
Therefore the verification test/QA plan must contain a statement by the manufacturer/vendor of
performance capabilities to be evaluated in the verification testing. The statement must be
specific and verifiable (by statistical analyses of the data, when appropriate). Statements should
also be made regarding the applications of the equipment, the known limitations of the
equipment, and the advantages the equipment provides over other existing equipment. An
example of an acceptable statement of performance is:
"This ozone injection NOX control technology is capable of achieving _%
removal of NOX when operated at....[specify process operating conditions]."
-------
Technology Description. The verification test/QA plan must adequately describe the technology
to be verified. Name plate data, such as the technology name, model number, serial number, and
manufacturer's name and address might be sufficient. But for custom-designed technologies,
parameters that scale with throughput such as capacity or output rate, utility requirements, and
performance guarantees might be required to adequately describe a technology. Warning and
caution statements can also be provided in the description.
Other descriptive information the vendor may provide can address the logistical, human, and
economic resources necessary to operate the technology, such as installation requirements,
operator qualifications and training required, maintenance requirements, operating manpower
requirements, and technology life expectancy.
Verification Test Bounds of Applicability. Industrial processes have ranges over which they
operate well and compete strongly, and their use outside that range might not be recommended
by the manufacturer/vendor. A verification test must be designed to determine the performance
of an APCT in specified terms and of known quality and to define the applicability bounds of the
verification. Four major factors to consider in the test design are:
1. the scale of the control device,
2. the range of technology operation during the tests,
3. measurement methods, and
4. the number and length of measurements.
Each has an impact on the verification range of applicability. The possible options for scale of
the control device being verified are a full-scale installation, a pilot (transportable) device
operated on a slipstream at a full-scale facility, and a pilot device operated at a controlled
laboratory facility. A full-scale facility will provide a test that best matches real world
conditions but offers limited flexibility to test the device under as wide a range of conditions as a
vendor may request. A laboratory facility provides the most control of the pollutant source and
control device operating conditions, which allows the test to cover the broadest range of
conditions, but is the least representative of real world conditions among the scales discussed
here. A pilot device on a slipstream at a full-scale facility provides a compromise between the
two other approaches. Each test will be evaluated on a case-by-case basis as to the suitability of
various available test sites.
The range of technology operating conditions determines the breadth of applicability for the
verification statement. Key operating parameters, along with their expected range of values for
the desired applications, should be identified and included in the test design. Both measurement
methods and the number and length of measurements are discussed below as part of the
experimental design.
Two aspects of control device performance variability will not be addressed for reasons of cost
and practical difficulty. One is changes in performance over time. The verification will address
performance only during a one-time test. The second potential variability not covered is the
performance differences from one APCT device to another installation of the same device, which
can be controlled by the manufacturer's quality management program or performance guarantee.
-------
Each test conducted as part of the ETV for add-on NOX controls testing may be of a different
technology at another location with different flue gas and operating conditions. Thus the details
of the verification tests will be different, though the tests will include many elements common to
all verifications. For these reasons, separate test/QA plans will be required to attain at each test
site the data quality objectives (DQOs) specified in the GVP.
Ozone-injection Add-on NO, Test Parameters and Methods
Measurement parameters to consider in the verification tests fall into four categories:
1. Performance factors (e.g.,, direct emission measurements of inlet and outlet NOX),
2. Test conditions documentation (e.g., fuel type, fuel flow rate, air flow rate, percent of
rated capacity, combustion temperature, ozone injection rate, and ambient conditions).
3. Associated impacts (e.g., ozone slip, wastewater discharge), and
4. Associated resource usage (e.g., total energy usage, fuel usage).
Whenever possible, EPA or other standard measurement methods will be used,
Verification Test Reports and Statements. Both verification test reports and verification
statements will be prepared. The verification test report thoroughly documents the test results,
including testing and analytical procedures, raw test data, equipment calibration data and
procedures, and QA results. The report will also explain and document any necessary deviations
from the test/QA plan. The verification statement is a concise summary of the verification report
that describes the performance of the add-on NOX control device. The verification statement will
be prepared according to a standardized format that has been reviewed and approved by EPA. To
the extent possible, the format will be consistent with the data requirements of permitting
agencies, thereby enhancing reciprocity among agencies.
The remainder of this paper describes the experimental design that is to be used in the
verification testing. Several assumptions were made to allow a specific example to be presented.
The same approach will be used to develop the design for each verification test.
Experimental Design for Add-on NO, Verification.
NOX Emission Concentration POO. The critical measurement for add-on NOK control
technologies utilizing ozone injection has been identified to be control device NOX emission
concentration. For the NOX emission concentration, the test/QA plan will include measurements
sufficient to allow determination of the technology's overall NOX emission within ± 10% of the
mean emission concentration above 5 ppmv, ±25% below 5 and above 2 ppmv, and ± 50%
below 2 ppmv. The DQO is to be computed as the half-width of the 95% confidence interval of
the mean, divided by the mean, or, equivalently, as the product of the standard error of the mean
and the appropriate Students-t value divided by the mean. All measurements apply within the
performance envelope (process temperature, flow, inlet NOX concentration) being verified. The
NOX emission concentration will be measured using EPA Method 7E (40 CFR 60, Appendix A),
which is the reference standard for NOK emissions, and thus each measurement is taken to be
without bias.
-------
These stringent requirements arise out of a desire to compare various technologies through their
verification statements. In the absence of statistical design and QA, verification test results can
be taken to represent only a snapshot of performance. The DQOs above were thought by the
technical panel to be sufficient to allow users to compare different technologies and accomplish
the objectives of ETV.
Unfortunately, data were not found that demonstrate that the critical NOX DQO specified in the
paragraph above can be attained with a test program of modest duration. In all likelihood, the
NOX DQO can be met at some level of testing, but other constraints on the ETV process require
that the test cost be commensurate with the benefit derived from the verification. For this
reason, the DQOs specified in the GVP must be considered tentative until field data are available
to allow evaluation of this approach.
Should the verification test be conducted and the critical NOX DQO not be met due to excessive
data variability, the verification partner and testing organization will present the data to the
vendor and discuss the relative merit of various options. The two primary options will be either
to continue the test to obtain additional data, with resulting increases in cost to all parties, or to
terminate the test and report the data obtained.
Specific DQOs will be included in the test/QA plan for all measurements addressing technology
performance and the test conditions. The quality of measurements for associated environmental
. impacts and operating resources will be addressed through numeric DQOs when possible or
through discussions when numeric estimates are not possible. Specific measurement DQOs may
vary between different test/QA plans written to conform to the above critical NOX DQO.
While not critical, accurate measurements of test conditions such as flow rate, temperature, and
inlet NOX are important because they set the operational boundaries within which the technology
verification applies. Plant instruments may be used to make the measurements provided they are
found to be adequate and have current calibrations. Parallel calibrated instrumentation will be
used whenever practical. Measurement DQOs will be set after inspection of the test site and as
specified in the test/QA plan. The potential for measurement bias will be evaluated by inspection
and experience, QC procedures and technical assessments will evaluate measurement bias
during verification testing for those measurement parameters where the potential for bias has
been identified.
Statistical Experimental Design. The basic design will be to test the technology by performing
runs under different operating conditions. Three parameters were identified that can be
controlled and are expected to affect the performance of the ozone injection NOX control device:
the inlet NOX concentration, the reaction chamber temperature, and the reaction's residence time
(a surrogate measure is gas flow rate). Tests will be conducted under various combinations of
these three parameters.
The experimental design that is recommended is a 2 x 2 x 2 factorial design using each of the
three parameters at two levels (low and high). Each complete replication of the factorial design
requires eight test runs. At least 2 replications are recommended, giving a minimum of 16 test
runs, preferably randomized in order of performance. This would require about a week of field
7
-------
testing, which is thought to be a reasonable effort for a single verification.
This factorial design allows for statistical significance tests to determine whether the
performance measure varies significantly with any of the three parameters. Further, provided
that at least 2 replicates for a total of 16 runs are done, the significance of interactions can also be
tested. If the performance does not change significantly with a parameter, then the results are
valid for the range of that parameter. If the performance does vary significantly with some
parameter, then the statement of the results of the test must include information indicating the
dependence of the performance on (he operating parameter.
The confidence interval for the estimated performance measure, removal efficiency, depends on
several things: the variability of the NOX measurement, the desired level of confidence, the
number of runs, and the removal efficiency. Figure 2 illustrates how the half-width of the
confidence interval (i.e., accuracy of the removal efficiency determination) varies with the
number of test runs for three confidence levels. To construct the figure, a hypothetical data set
was prepared assuming that the removal efficiency was approximately 90% and that the error in
the NOX measurement was about 5%. (Note, the error in the NOX measurement could be as high
as 25% for outlet concentrations below 5 ppm [Hung and Campbell, 1998].) Since the removal
efficiency depends on a ratio of two NOX measurements, an approximate confidence interval for a
ratio was used.
Figure 2 shows the half-widths of confidence intervals for three confidence coefficients. The
upper line corresponds to a confidence coefficient of 99%, the middle line to 95%, and the lower
line to 90%. Assuming two complete replications of the design, an expected 95% confidence
interval for the removal efficiency (of about 90%) can be estimated from the figure to be about
3%. Then the estimated 95% confidence interval for the efficiency is 90% NOX removal ± 3.0%,
or from 87 to 93% NOX removal.
Application of Experimental Design Process to a Specific Example.
The experimental design process described above will now be applied to a specific set of
performance claims. Assume that a manufacturer/vendor presents the following set of
performance claims:
"This ozone injection NO, control technology is capable of achieving a NOX
emission concentration of 1 ppmv when operated at an inlet NOX concentration of
55 ppmv, ozone injection temperature of 290°F, and flow rate of 20,000 cfm; the
controlled NOX emissions are below 10 ppmv when operated at an inlet NO^
concentration of 300 ppmv, 310°F, and 30,000 cfm."
The performance of the technology can then be verified within the context of these claims by
setting up the experimental matrix with the indicated high and low values of inlet NOX
concentration, temperature, and flow rate for the particular test site where the technology is
installed. In addition to the statements above, the performance claims would also include a
description of the combustion source and fuel, combustor operation, and other operating features.
The test/QA plan would then consist of a detailed schedule of tests and data required to conduct
-------
-------
Table 1. Hypothetical Field Test Data Set
Run Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Mean HI
Mean LO
IN NO,,
ppmv
55.1 (LO)
54.7 (LO)
53.8(LO)
• 55,3 (LO)
287 (HI)
290 (HI)
292 (HI)
292 (HI)
52.8 (LO)
51.6(LO)
53.1 (LO)
52.5 (LO)
291 (HI)
294 (HI)
289 (HI)
290 (HI)
290.6
53.6
TEMP,
°F
285 (LO)
290 (LO)
320 (HI)
325(HI)
283 (LO)
287 (LO)
303 (HI)
306 (HI)
289 (LO)
291 (LO)
3 12 (HI)
3 15 (HI)
290 (LO)
292 (LO)
306 (HI)
309 (HI)
312
288
FLOW,
cfm
21,000(LO)
29,000 (HI)
20,000 (LO)
29,500 (HI)
20,500 (LO)
30,000 (HI)
20,000 (LO)
28,500 (HI)
20,400 (LO)
30,100 (HI)
20,800 (LO)
29,600 (HI)
20,200 (LO)
30,200 (HI)
21,100(LO)
3 1,000 (HI)
29,738
20,500
OUT NO,,
ppmv
1.22
1.32
1.11
1.26
8.70
10.2
9.10
9.70
1.12
0.98
1.21
1.26
7.80
9.20
8.10
9.50
The next step in the statistical analysis was to repeat the analysis of variance including only the
significant factors: IN NOX, FLOW, and their interaction. This step confirmed that all the
factors remain significant, with P-values below 0.05. It also showed that only the IN NOX
parameter was significant at the 99% level (P-value below 0.01). At this point a decision must
be made whether a significance below 99% is important in reporting the results; i.e., should the
effects of FLOW and IN NOX * FLOW be included in the verification results. Otherwise, the
verification performance results would be reported in terms of only IN NOX.
For this example, it was assumed that only the effect of IN NOX was significant enough to be of
interest. Variation in the performance results due to the parameter FLOW or the interaction will
not be accounted for separately, but as part of the variation related to the parameter IN NOX. Two
approaches can be used for the final analysis. One is to fit a model with IN NOX as the only
parameter. This result is shown in Table 2. The estimated performance results are presented
separately for the low and high IN NOX levels. Confidence intervals for the OUT NOX level can
be calculated by taking the mean for each IN NOX level and adding and subtracting the t-value for
14 degrees of freedom and a 95% confidence interval times the standard error indicated in the
table. The t-value for this example is 2,145; it can be found in standard statistical texts. The
verification claim in this case would be that, for temperatures between 288 and 312 °F and flow
rates between 20,500 and 29,738 cfm, the outlet NOX concentration was 1.185 ± 0.438 ppmv at
an inlet NOX concentration of 54 ppmv and 9.038 ± 0,515 ppmv at an inlet NOX concentration of
10
-------
Table 2. IN NO, Model
Parameter or Interaction - Level
IN NOX - Low
IN NOX - High
OUT NOX Least Square Mean,
ppmv
1.185
9.038
Standard Error,
ppmv
0.204 .
0.240
290 ppmv. These results meet the NOX emission concentration measurement DQO stated earlier.
The second approach to estimating the performance of the control device is to perform a
linear regression of OUT NOX on IN NOX. The result is an equation of the form
OUT NOX = a + b (IN NOX)
that can be used to predict the OUT NOX as a function of IN NOX. For this example, the estimated
value of a, the intercept, is -0.593 ppmv while that of b, the slope, is 0.033. Thus, the predicted
equation is
OUT NOX = -0.593 + 0.033 (IN NOX)
This linear equation would be applicable over the tested range of temperature, flow rate, and inlet
NO,, values, as shown in Table 1. Outside that range, it might be useful, but such use will not
have been verified.
In some cases, the difference in OUT NOX values for the low and high IN NOX levels may
be too small to be of practical importance. For example, if the OUT NOX differed by only
1 ppmv between the low and high IN NOX levels, then one would not likely make a distinction in
performance based on the IN NOX level as in the example above. For such a case, the overall
mean OUT NOX would be calculated and reported along with the appropriate confidence interval.
Additional hypothetical data sets, which include increased levels of variability in all four
experimental parameters - inlet NOX, outlet NOX, temperature, and flow rate - have also been
investigated. They generally support the expectation that a 2x2x2 test matrix will provide NOX
emission concentration data meeting the DQOs stated above. However, field process variability
cannot be accurately taken into account. For this reason, confirming that acceptable levels of
process variability exist in the field will be an important goal of the verification test. While
preliminary calculations and cost estimates will be based on the expectation of using a fully
replicated 2x2x2 test matrix, in practice additional tests may be required to narrow the
confidence interval.
SUMMARY
The EPA is implementing the ETV program to verify the performance of innovative or improved
technologies which protect human health and the environment. The program, composed of 12
pilot programs focused on selected pollution areas, concerns commercially ready technologies
11
-------
and seeks to enhance markets for the verified environmental technologies, particularly those of
U. S, companies. The APCT program focuses on air pollution control technologies and includes
Add-on NOX Control technologies as an active verification program.
Ozone-injection NOK control technologies were chosen as the NOX control area of initial
emphasis, and the GVP has been written. The GVP requires a statistical experimental design
having the goal of providing highly reliable NOX control performance data. While the protocol
has not yet been applied in the field, the evaluation of hypothetical (but realistically variable)
results indicates that the proposed test design will meet the data quality requirements. Field
testing of add-on NOX control devices utilizing ozone injection has not yet begun, but is expected
to begin this year.
REFERENCES
EPA. 1990 Clean Air Act. West Publishing Company, St. Paul, MN. 1991. Available on web
at http://www.epa.gov/oar/caa/caaa.txt.
EPA. Environmental Technology Verification Program Verification Strategy. EPA/6GO/K-
96/003. (NTIS PB97-160006. http://www.epa.gov/etv/dload/etv strategv.pdf). Office of
Research and Development, U. S. Environmental Protection Agency, Washington, DC, 1997.
EPA. 40 CFR Part 60. Proposed Revision of Standards of Performance for Nitrogen Oxide
Emissions From New Fossil-Fuel Fired Steam Generating Units; Proposed Revisions to
Reporting Requirements for Standards of Performance for New Fossil-Fuel Fired Steam
Generating Units. Federal Register, Vol. 62, No. 131, p. 36957.
EPA. 40 CFR Part 60. Revision of Standards of Performance for Nitrogen Oxide Emissions
From New Fossil-Fuel Fired Steam Generating Units; Revisions to Reporting Requirements for
Standards of Performance for New Fossil-Fuel Fired Steam Generating Units. Federal Register,
Vol. 63, No. 179, p. 49443.
EPA. Appendix A - Test Methods to National Emission Standards for New Stationary Sources.
Code of Federal Regulations, Title 40, Part 60 (40 CFR Ch. 1 (7-1-98 Edition) Pt. 60, App. A),
http://www.epa.gov/epacfr40/chapt4.info/subch-C/40P0060/40P060XA.pdf.
Hung, Wilfred S. Y. and Alan Campbell, "Uncertainty in Gas Turbine NOX Measurements,"
ASME Paper No. 98-GT-75. Presented at the International Gas Turbine & Aeroengine Congress
& Exhibition, Stockholm, Sweden, June 2-5, 1998.
RTI. Generic Verification Protocol for Add-on NOX Control Technologies Utilizing Ozone
Injection For Stationary Combustion Sources. http://etv.rti.org/apct/pdf/addonNOxGVP.pdf
Research Triangle Park, NC. 2000.
SYSTAT, Version 8.0. SPSS Inc., 233 S. Wacker Drive, 11th floor, Chicago, IL. 1998.
12
-------
NRMRL-RTP-P-497
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing]
1. REPORT NO.
EPA/600/A-00/007
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
S. REPORT DATE
Test Design for Environmental Technology Verifica-
tion (ETV) of Add-on NCX Control Utilizing Ozone
Injection
6. PERFORMING ORGANIZATION CODE
7.AUTHOR(s!D( VanOsdell and J.Farmer (RTT), A.Tren-
holm and C. Clapsaddle (MRI), and T.Brna (EPA) -
8. PERFORMING ORGANIZATION REPORT NO,
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Research Triangle Institute, P.O. Box 12194, RTP,
NO 27709.
Midwest Research Institute, 5520 Dillard Rd., Gary,
NO 27511.
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
CR826152-01-2 (RTI)
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Air Pollution Prevention and Control Division
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Published paper; 1-3/00
14. SPONSORING AGENCY CODE
EPA/600/13
is.SUPPLEMENTARY NOTES APPCD project officer is Theodore G. B.rna, Mail Drop 4, 919/
541-2683. Presented at Forum 2000, Rosslyn, VA, 3/23-24/00 (sponsored by
Institute of Clean Air Companies).
16.ABSTRACT Tjie paper discusses the test design for environmental technology verifica-
tion (ETV) of add-on nitrogen oxides (NOx) control utilizing ozone injection. (NOTE:
ETV is an EPA-established program to enhance domestic and international market
acceptance of new or improved commercially ready technologies through independent
testing which incorporates rigorous quality assurance (QA) procedures. Through
ETV, potential purchasers and permitters are provided an independent and credible
assessment of what they are buying and permitting.) Operated by the Research Tri-
angle Institute, the Air Pollution Control Technology (APCT) program is one of 12
ETV pilot programs. Add-on NOx air pollution control technologies are among
those being verified, and the Low-temperature Oxidation (LTO) technology is the
first NOx control technology planned for verification by the APCT program. LTO
uses ozone injection into the gas phase of a combustion or chemical process waste
gas stream to oxidize NOx molecules to higher order NOx, which can be removed by
wet or semi-dry scrubbing. Ozone is produced on-site and on-demand from stored
oxygen. As a low-temperature process, it can be installed as an end-of-pipe system
The LTO technology was jointly developed and patented by Cannon Technology and
BOC Gases; LoTOx is BOC Gases' trademark for large-scale LTO applications.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.iDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
Pollution
Verifying
Ozone
Nitrogen Oxides
Oxidation
Boilers
Natural Gas
Combustion
Pollution Control
Stationary Sources
Environmental Technol-
ogy Verification
13B 21D
14B 2 IB
07B
07C
13A
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
20. SECURITY CLASS (This page}
Unclassified
21. NO. OF PAGES
-12-
22. PRICE
EPA Form 222O-1 (9-73)
------- |