EPA/60Q/A-95/067
1
Continuous Emissions Monitoring Demonstration Program
Dan B. Bums
Westinghouse Savannah River Company
Savannah River Technology Center
Aiken, SC 29802
Marta K. Richards
U.S. Environmental Protection Agency
Office of Research and Development
Cincinnati, OH 4526S
ABSTRACT
Continuous emissions monitoring of hazardous and mixed waste thermal treatment processes is
desired for verification of emission compliance, process control, and public safety perception.
Species of particular interest include trace metals and organic compounds resulting from
incomplete destruction. Continuous real-time monitoring of these pollutants would permit
measurement of real-time (actual) hazardous compound emissions and allow accurate (realistic)
human risk assessment from hazardous and mixed waste thermal treatment facility operation.
This paper describes a joint DOE/ EPA program developed to identify and demonstrate
emerging continuous emissions monitoring technologies ready for pilot-scale demonstration.
The demonstrations will include burning simulated waste spiked with hazardous metals and
organics in a pilot-scale rotary kiln incinerator while flue gas metals and organics concentrations
are continuously monitored. Simultaneous manual flue-gas sampling using EPA reference
method sampling trains and analytical procedures will provide a benchmark for the continuous
monitoring technologies. Both method accuracy and short-term system reliability will be
assessed. A program coordination committee consisting of representatives from DOE, EPA,
academia, end-users, and technology developers will provide technical support in demonstration
protocol development, technology selection criteria, and performance assessment.
INTRODUCTION
Continuous emissions monitoring of hazardous and mixed waste thermal treatment processes is
desirable for several reasons: verification of emission compliance, enabling responsive process
control, and increasing public confidence in thermal treatment process safety and regulatory
agency credibility. Pollutants of particular interest include trace metals and organic compounds
resulting from incomplete destruction of wastes. In particular, EPA plans to propose new rules
in 199S governing the emission of toxic metals from hazardous waste thermal treatment facilities,
including hazardous waste incinerators, cement and aggregate kilns, and smelting, melting, and
refining furnaces. Continuously monitoring of these species would permit real-time hazardous

-------
2
compound emissions measurement and realistic assessment of human risk associated with the
atmospheric release of these compounds resulting from the operation of hazardous and mixed
waste thermal treatment facilities.
Programs are currently being funded by the Department of Energy (DOE), the Environmental
Protection Agency (EPA), and private industry to develop technologies and equipment for
continuous emissions monitors (CEM)s. Continued advancement and future implementation of
these technologies require pilot- and full-scale demonstrations in realistic process environments.
The DOE and EPA have established a joint program to demonstrate emerging technologies for
continuously monitoring metals and organic compound emissions from pilot-scale thermal
treatment facilities. The objectives of this program include identifying promising emerging
technologies for continuous monitoring of hazardous compounds in emissions from thermal
treatment facilities. A demonstration protocol will be developed to evaluate each CEM
technology against defined criteria and EPA reference methods. A series of technology
demonstration tests will be performed at the EPA Incineration Research Facility (IRF), a
pilot-scale rotary kiln incinerator. This program is not intended to provide a definite assessment
of which CEM is best, but is intended to reveal potential advantages and disadvantages with each
technology and identify issues that could be encountered in a process environment. This joint
program will utilize a team of recognized experts in the CEM field to provide technical support
of all program activities.
TECHNOLOGY REQUIREMENTS
Continuous monitoring of an emissions source by the strict definition requires continuous and
red-time sampling, analysis, and reporting. Yet, the current EPA Office of Solid Waste (OSW)
definition of continuous emissions monitoring requires continuous process sampling, while the
analysis can be conducted in a batch operation. The batch analysis must be completed on-site
and be integral to the CEM. The CEM should provide a concentration value for the species of
interest at least once every three hours. The response time (the time interval between the start of
a step change in the system input and the time when the monitor output reaches 95% of the final
value) of the CEM should be less than three hours. For CEMs utilizing batch analyses, the delay
between the end of the sampling time and reporting of the sample analysis should be no greater
than one hour. Also, there should be no greater than a five-minute gap in sampling when the
sample collection media is changed.1 Thus, a CEM should be able to continuously sample facility
emissions and have as close to real-time reporting of effluent concentrations as possible.
In addition to requirements for sampling and data reporting, the CEM must have detection limits
low enough to assure ability to comply with the eventual regulatory limits for specific species of
interest. A multi-metals CEM should be capable of measuring, at a minimum, the total elemental
concentrations of two or more of the metals listed in Table 1. Although final detection limit
requirements for each metal have not been determined (it will be based on the future regulatory
emission limits for each metal or group of metals which is yet to be defined), a metals CEM
should be capable of measuring concentrations of each metal (in both the solid and vapor form)
at or approaching the detection limits listed in Table 1.

-------
Table 1
Multi-Metals CEM Detection Limits
Metal
Detection Limit

Oig/m3)
Antimony
5.0
Arsenic
5.0
Barium
2.5
Beryllium
0.25
Cadmium
2.5
J Chromium
2.5
I Cobalt
2.5
Lead
25
Manganese
0.5
Mercury
0.5
Nickel
2.5
Selenium
25
Silver
2.5
I Thallium
2.5
Organic CEMs should be capable of measuring stack gas concentrations of one or more of the
organic compounds listed in Table 2 at a detection limit approaching 1 ^ig/m3 for all compounds
except dioxins. Concentrations of interest for dioxins are in the ng/m3 range, or lower.

-------
4
Table 2
»anic Speclesfor CEMs
Benzene
Carbon Tetrachloride
Chlorobenzene
Chloroform
Dichlorobenzenes
Dichloroethanes
Dichloroethenes
Formaldehyde
Methylene Chloride
Polynuclear Aromatic Hydrocarbons
Tetrachloroethene
Toluene
Trichloroethanes
Trichloroethene
Vinyl Chloride
Dioxins
TEST FACILITY
All CEM demonstration tests will be conducted in the pilot-scale rotary kiln incineration system
(RKS) at EPA's Incineration Research Facility (IRF) located in Jefferson, Arkansas. A process
schematic of the RKS is shown in Figure 1. The RKS consists of a primary combustion chamber
(a 590 kW rotary kiln with 1.04 m ID and 2.26 m length), a transition section, and a fired
afterburner chamber (590 kW, 0.91 m ID, 3.05 m length). After exiting the afterburner, flue gas
flows through a quench section followed by the primary air pollution control system (APCS).
The primary APCS consists of a venturi scrubber followed by a packed-column scrubber.
Downstream of the primary APCS is a redundant APCS to ensure facility permit compliance.
The redundant APCS contains a demister, an activated-carbon adsorber, and a high-efficiency
particulate air (HEP A) filter.
Currently, it is anticipated that up to four CEMs would be tested simultaneously. The CEMs
would likely be located in the facility flue gas duct (0.36 m diameter, gas velocity approximately
6 m/s) between the primary and secondary APCS at the scrubber exit. At this location, the flue
gas is saturated, approximately 60'C, and can contain 100-200 mg/m3 particulates with a mean
particle size below 20 microns. Also, the flue gas can contain up to 5 ppm acid gas at this
location.

-------

QUENCH
NATURAL
GAS,
LIQUID
FEED
TRANSFER
DUCT
i
SECONDARY
BURNER
VEnTUhI
SCRUBBER
AFTERBURNER
EXTENSION
SOUDS
SCREW
FEEDER
AFTERBURNER
MAIN
BURNER
ID FAN
PACKED
COLUMN
SCRUBBER
ROTARY
KILN
NATURAL
GAS,	|
LIQUID FEED
SCRUBBER
LIQUOR
RECIRCULATION
44-
Oi
<0
o
UJ
<
ATMOSPHERE
CARBON BED HEPA
ADSORBER FILTER
t
STACK
ID FAN
ROTARY KILN
INCINERATOR
PRIMARY AIR POLLUTION
CONTROL SYSTEM
REDUNDANT AIR I
POLLUTION CONTROL |
SYSTEM
Figure 1. Schematic of the IRF rotary kiln incineration system.
U1

-------
6
Simulated waste will be incinerated during all CEM demonstration tests. Metals will be
introduced by two different methods. A simulated hazardous solid waste will be generated by
adding a concentrated aqueous metals solution (primarily soluble metal nitrates) to a clay-based
oil sorbent material (< 1 mm particle diameter). A mixture of toluene, chlorobenzene, and
tetrachloroethene will be added to the solid waste at a concentration of approximately 4 wt%.
These organics will provide a source of chlorine which is present in most hazardous waste
streams. This waste simulant will be continuously fed to the rotary kiln and burned at a nominal
exit gas temperature of 870*C. In addition to the metal and organics-bearing solid waste, an
aqueous metal solution will be injected directly into the kiln main burner, simulating the burning
of liquid hazardous waste containing trace toxic metals.
Organics CEMs will be tested by metering a mixture of approximately 10 selected volatile and 3
semi-volatile organic compounds into the flue gas. Hie specific organic spiking compounds mid
their respective flue gas concentrations will be selected based on the CEMs being tested.
TEST METHOD
The CEM demonstration will be designed and conducted to assess each instrument for the
following performance characteristics:
•	Relative Accuracy - The absolute mean difference between species concentrations in the flue
gas determined by the CEM and the value determined by the applicable reference method.
•	Calibration and Zero Drift - The difference in the CEM output from the established reference
values (including a blank) after a stated period of operation during which no unscheduled
maintenance, repair, or adjustments took place.
•	Response Time - The time interval between the start of a step change in the species
concentration and the time when the CEM output reaches 95% of the final value.
•	Instrument Robustness - A qualitative assessment of instrument applicability to a process
environment. The criteria includes set-up time, ease of operation, percent downtime,
maintenance requirements, calibration time, number of operators, etc.
The assessment of relative accuracy will consist of three separate runs at three concentration
levels of the species of interest. For all metals CEM tests (including the monitors that measure
only one specific metal), the gas stream will contain all the metals listed in Table 1. The target
metal concentrations in the flue gas for each run is shown in Table 3.

-------
7
TABLE 3
Target Metal Concentrations
Metal
Assumed
Detection Limit
(Hg/m3)
Run 1
Concentration
(|ig/m3)
Run2
Concentration
(jig/m3)
Run 3
Concentration
(ng/m3)
1 Antimony
5.0
10
40
400
| Arsenic
5.0
10
40
400
Barium
2.5
5.0
20
200
Beryllium
0.25
0.5
2.0
20
Cadmium
2.5
5.0
20
200
1 Chromium
2.5
5.0
20
200
I Cobalt
2.5
5.0
20
200
I Lead
25
50
200
2000
I Manganese
0.5
1.0
4.0
40
Mercury
0.5
1.0
4.0
40
Nickel
2.5
5.0
20
200
Selenium
25
50
200
2000
Silver
2.5
5.0
20
200 I
Thallium
2.5
5.0
20
200 1
The flue gas for the organics CEM tests will contain an assortment of volatile and semi-volatile
organic compounds. The specific organic spiking compounds will be selected based on the
CEMs being tested. The concentrations of each organic species in the flue gas stream will be
defined based on the capabilities of the CEMs being tested, yet the organics concentrations
should be in the range of 0.5,2.0, and 20 ng/m3for each of the three runs.
During each run, three manual Reference Method sampling procedures will be conducted. Each
Reference Method sampling event is expected to last approximately three hours. Thus, the CEM
will be exposed to nine hours of steady-state operation at each of the three target concentrations
while three independent Reference Method samples are being taken. The Reference Methods
planned for use are shown in Table 4.

-------
8
TABLE 4
Test Reference Methods
Pollutant
Reference Method
Multi-Metals
M29
Mercury
M101A
Volatile Organics
MO03O
Semi-Volatile Organics
MOO 10
PCDD/PCDF
M23
Calibration procedures, methods, and materials will be specified, supplied, and conducted by the
respective CEM developer/operator. Only the analyzer, not the sampling interface, requires
calibration with calibration and zero drift measurements. If, during the course of a run, or any
time during the demonstration period, automatic or manual adjustments are made to the CEMs
zero or calibration settings, a drift measurement must be taken. At minimum, zero and
calibration drift measurements will be taken at the beginning and end of each set of three runs.
The instrument response to a process step change will be measured by two separate methods.
The first method requires operation of the CEMs for a sufficient time period both before and
after spiked feed is introduced to the facility. The second method will utilize injection of a
spiked simulant directly into the flue gas duct. Since both of these methods also include some
lag time inherent to the test facility, it is not a true CEM response time measurement. These
response time tests are designed to demonstrate the instrument performance during a step change
in facility operation. The CEM response to both of these transients will be recorded. Each
GEM'S resolution and response time will determine the ability to follow and recorded these
events. This will also define how close these monitors are to "real time." There is no reference
method for measuring response time.
For EPA/DOE test records, during the course of testing, each CEM operator will maintain a test
log book documenting the time required for instrument set-up, shakedown, and calibration. All
maintenance activities will be recorded, along with all calibration and zero drift data. The
percent the instrument is off-line due to calibration or maintenance will also be recorded. In
addition to the CEM operator log, test personnel will be conducting general observations of the
ease of operation and the quantity of "hands on" attention each monitor requires. This type of
qualitative data will allow estimation of the technology robustness and if or how soon the
monitor will be ready for use in a commercial facility.
PROGRAM SCHEDULE
The request for interested CEM developers was published in the January 4, 1995 Commerce
Business Daily. The draft test plan and protocol was sent to all potential participants in early
April. The CEM developers are requested to respond with a commitment to participate by

-------
9
mid-May. The demonstrations are scheduled to begin in late June 1995 and be completed by the
first of September. The program report is expected to be complete in early November 1995.
SUMMARY
Continuous emissions monitoring of hazardous and mixed waste thermal treatment processes is
desired for verification of emission compliance. Species of particular interest include trace
metals and organic compounds resulting from incomplete destruction. Continuous real-time
monitoring of these pollutants would permit actual measurement of hazardous compound
emissions and allow accurate human risk assessment from hazardous and mixed waste thermal
treatment facility operation. A joint DOE/ EPA program has been developed to identify and
demonstrate emerging continuous emissions monitoring technologies for metals and organics
ready for pilot-scale demonstration. The demonstrations will take place at the EPA Incineration
Research Facility in Jefferson, Arkansas. Simulated waste spiked with hazardous metals and
organics will be burned in a rotary kiln incinerator while flue gas metals and organic
concentrations are continuously monitored. Simultaneous manual flue gas sampling will provide
a benchmark for the continuous monitoring technologies. The CEM demonstration will be
designed and conducted to assess each instrument for relative accuracy between the CEM and
the value determined by the applicable reference method. Calibration and zero drift will be
measured at least every twenty-four hours during testing. Response time to step changes in the
process will also be measured. Instrument robustness will be assessed for applicability to a
process environment. A program coordination committee consisting of representatives from
DOE, EPA, academia, end-users, and technology developers will provide technical support in
the determination of demonstration protocols, technology selection criteria, and CEM
performance assessment.
REFERENCES
1. "Draft Performance Specification - Specification and Test Procedures For Multi-Metals
Continuous Emission Monitoring Systems in Stationary Sources," March 22, 1995.

-------
TECHNICAL REPORT DATA
fPlease read Instructions on the reverse before completing)
—
1. REPORT NO. 2.
EPA/600/A-95/067
3. RtCiP
4. TITLE ANO SUBTITLE
Continuous Emissions Monitoring Demonstration
Program
5. REPORT DATE
6. PERFORMING ORGANIZATION CODE
7. AUTHQR(S) Dfln B> Burn$ an£j
Marta K. Richards
8. PERFORMING ORGANIZATION REPORT NO,
9. PERFORMING ORGANIZATION NAME ANO ADDRESS
Acurex Environmental Corporation
IRF
Jefferson, Arkansas 72079
10 PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-C4-0044
12, SPONSORING AGENCY NAME ANO AODRESS
RISK REDUCTION ENGINEERING I ABORATORY-CINCINNATI, OH
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
13. TYPE OF REPORT ANO PERIOD COVERED
PROCEEDINGS P*p.or
14. SPONSORING AGENCY CdOE
EPA/600/14
16. SUPPLEMENTARY NOTES
Incineration Conference May 8-12, 1995 Bellevue, Washington
Technical Project Officer: Marta K. Richards 513/569-7692
16. ABSTRACT
Continuous emissions monitoring of hazardous and mixed waste thermal treatment
processes is desired for verification of emission compliance, process control, and
public safety perception. Species of particular interest include trace metals and
organic compounds resulting from incomplete destruction. Continuous real-time monitor-
ing of these pollutants would permit measurement of real-time (actual) hazardous
compound emissions and allow accurate (realistic) human risk assessment from hazardous
and mixed waste thermal treatment facility operation. This paper decribes a joint
DOE/EPA program developed to identify and demonstrate emerging continuous emissions
monitoring technologies ready for pilot-scale demonstration. The demonstrations will
include burning simulated waste spiked with hazardous metals and organics in a pi tot-
scale rotary kiln incinerator while flue gas metals and organics concentrations are
continuously monitored. Simultaneous manual flue-gas sampling using EPA reference
method sampling trains and analytical procedures will provide a benchmark for the
continuous monitoring technologies. Both method accuracy and short-term system reli-
ability will be assessed. A program coordination committee consisting of representa-
tives from DOE, EPA, academia, end-users, and technology developers will provide
technical support in demonstration protocol development, technology selection criteria,
and performance assessment.
17. KEY WORDS AND DOCUMENT ANALYSIS
». DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COS ATI Field/Croup

Monitors
Emissions
Measurement
Flue Gas
Combustion
Organic Compounds
Metals
•
16. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19 SECURITY CLASS (This Report)
UNCLASSIFIED
21. NO. OF PAGES
10
20. SECURITY CLASS (This page)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (R»*. 4-77) previous coition ts obsolete

-------