Execution of Comprehensive Performance Test Using Particulate, HCl and
Metals Continuous Emissions Monitoring Systems

Paper #26

Keith Beach1, P.E., Thomas G. Busmann2, P.E., Chris E. McBride2, P.E., 
Richard H Lambert3, and Krag Petterson4

1Evonik Degussa Corporation, Tippecanoe Laboratories, 1650 Lilly Road,
Lafayette, IN   47909   USA

2Focus Environmental Inc., 4700 Papermill Drive, Knoxville, TN  37909
USA

3Eli Lilly and Company, Lilly Corporate Center, Indianapolis, IN  46285
USA

4Cooper Environmental LLC, 10180 SW Nimbus Ave. Suite J6, Portland, OR
97223 USA

ABSTRACT

Evonik Degussa Corporation, Tippecanoe Laboratories successfully
demonstrated compliance with the HWC MACT Comprehensive Performance Test
(CPT) requirements utilizing the continuous emission monitors for
particulate matter (PM), hydrochloric acid (HCl), and metals as the
compliance analyses for stack emissions.  It is the authors’ belief
that this is the first CPT conducted in the United States utilizing this
approach.

This paper will briefly review the history of the CEMs development on
the solid-liquid incinerator, provide an overview of the CPT plan and
objectives, discuss the performance of the CEMs prior to and during the
CPT, and present the performance test results.  This performance test
demonstration illustrates that the use of CEMs for performance test
demonstration is an efficient, effective, and viable approach for
compliance demonstration on hazardous waste incinerators.

INTRODUCTION

Evonik Degussa Corporation (Evonik) operates a Hazardous Waste
Incinerator that burns solids and liquid organic and aqueous waste at
their newly acquired facility, Tippecanoe Laboratories (Tippe), in
Lafayette, IN.  The site was acquired by Evonik in 2010 from Eli Lilly
and Company (Lilly) and has continued to manufacture pharmaceutical and
animal health compounds for Lilly.  

Selected incinerator operating parameters limits and emissions limits
are specified by the U. S. EPA Hazardous Waste Combustor Maximum
Achievable Control Technology (HWC MACT) rule in order to minimize
emissions of Hazardous Air Pollutants (HAPs). Since the early 1990s when
the MACT rules were initially being considered, EPA has often expressed
a desire to use Continuous Emissions Measurement Systems (CEMS)
technology as a means to directly demonstrate compliance by measuring
emissions directly, rather than use operating parameters to infer
compliance. Several years ago Tippe embarked on a program to evaluate
and apply available CEMS technologies for particulate matter and
hydrogen chloride, and to assist in developing a novel CEMS technology
for metals.

Evonik successfully completed a Comprehensive Performance Test (CPT) in
2010 utilizing CEMs for HCl, particulate and the HWC/MACT metals in lieu
of standard EPA test methods for those pollutants.  It is the authors’
belief that this is the first CPT conducted in the United States
utilizing this approach.

HISTORY AND BACKGROUND

A previous Lilly affiliate facility in Ireland employed particulate
matter (PM) CEMS and infrared multi-component CEMS under a different
regulatory structure for a several years as indicators of compliance. 
EPA included the requirement for the application of PM CEMS to hazardous
waste incinerators in the proposed HWC MACT rule in 1996.  Tippe, along
with a number of industry peers, demonstrated the limitations of these
technologies on certain types of hazardous waste incineration systems. 
EPA subsequently deferred the requirement to implement PM CEMS under the
HWC MACT. While demonstrating the viability of the PM CEMS, Tippe joined
forces with Cooper Environmental Services to develop practical
application of X-ray fluorescence as a multi-metals CEMS technology use
on a solid-liquid waste incinerator.

Tippe has been involved with the testing, evaluation and development of
CEMs beginning with installations on the liquids incinerators in the
1990s and subsequently on the solids and liquids incinerators in the
early 2000s.

This development of the use of these CEMS technologies met a couple key
Tippe concerns.  First, using CEMS data for compliance reduced or
eliminated the need for sampling and analysis of individual containers
to develop waste characterization.  This addressed the concerns Tippe
had regarding the costs and potential exposure required by typical
sampling and analytical programs.  Due to the nature of high-potency
pharmaceutical wastes, reducing or eliminating the typical sampling and
analysis activities provided significant safety, operational, and
economic benefits. 

Second, Tippe was also interested in the benefits from operational
flexibility that could be realized by removing many of the prescribed
HWC MACT operating parameters limits on the incinerator’s air
pollution control system.

Tippe installed continuous monitoring of the combustion (stack) gas
emissions on the solid and liquid waste incinerator designated T149
utilizing three CEMS:

An EcoChem MC3 or equivalent for CO, O2, and HCl

A Sigrist CTNR or equivalent for particulate emissions

A Cooper Environmental Services XACT or equivalent for metals.

Since the required use of CEMs for parameters other than CO and O2 was
removed from the HWC MACT requirements, the use of CEMs for PM, HCl and
multi-metals required the submission and approval of an alternative
monitoring petition.  The requirements for petitioning the EPA for
alternative monitoring for compliance is found in 40 CFR Part 63.8 (f),
the General Provisions of the National Emission Standards for Hazardous
Air Pollutants (NESHAP) for Source Categories. Paragraph (f)(4)(ii)
states:

“The application must contain a description of the proposed
alternative monitoring system which addresses the four elements
contained in the definition of monitoring in §63.2 and a performance
evaluation test plan, if required, as specified in paragraph (e)(3) of
this section. In addition, the application must include information
justifying the owner or operator's request for an alternative monitoring
method, such as the technical or economic infeasibility, or the
impracticality, of the affected source using the required method.”

Tippe initiated the process for the alternative monitoring petition
(AMP) in early 2003. The AMP was approved by the EPA in January 2006. 
Implementation of the AMP then required a major modification to the
facility’s Title V permit.

CPT OBJECTIVES

To demonstrate compliance with the HWC MACT performance standards and
emission limits, Tippe proposed a single, low-temperature, combustion
chamber test condition, treating solid and liquid wastes with enhanced
ash, metals and chloride contents. This single test condition
demonstrated the ability of the combustion system to comply with the
applicable performance standards at the worst-case conditions of minimum
combustion temperatures and maximum combustion air flow, and maximum
waste, ash, and chloride feed rates.  

Since T149 has not been modified in a way which would affect its ability
to meet the DRE standard since DRE testing was conducted in September
2005, Tippe did not conduct a DRE test during the testing described
here. This is in accordance with the provisions of 40 CFR
63.1206(b)(7)(i)(A), the HWC MACT Replacement Standard requires that
compliance with the destruction and removal efficiency (DRE) standard
must only be documented once, provided that the source is not modified
after the DRE test in a way that could affect the ability of the source
to achieve the DRE standard.

Under the provisions of the AMP approved in a letter dated January 27,
2006, the use of a particulate matter CEMS, a multi-metals CEMS, and an
HCl CEMS to demonstrate compliance with the particulate matter, metals,
and HCl/Cl2 HWC MACT emission standards was approved and Tippe used that
provision to demonstrate compliance for this CPT test.

The performance standards established that are applicable for the T149
incinerator are summarized in   REF _Ref289155249 \h  Table 1 .

Table   SEQ Table \* ARABIC  1 . HWC MACT Replacement Rule Performance
Standards

Performance

Standard	

HWC MACT 40 CFR 63 Subpart EEE 

g/dscm1

	40 CFR 63.1219(a)(4)	As, Be & Cr combined: 92 g/dscm1

	40 CFR 63.1219(a)(2)	Hg: 130 g/dscm1

Dioxin/furan emissions	40 CFR 63.1219(a)(1)(ii)	0.40 g/dscm TEQ1

1. Corrected to 7% Oxygen.

CONTINUOUS MONITORING SYSTEMS PERFORMANCE EVALUATION TEST

As a part of the CPT, a performance evaluation of the CEMS was conducted
in accordance with the CMS PET Plan.  The evaluation showed that the
Evonik’s CMS was operating in compliance with the HWC MACT
requirements [40 CFR 63.1209(a)-(b)] as: 

The CEMS monitoring CO and O2 stack gas concentrations met the
appropriate performance specifications promulgated by the EPA. 

The CEMS monitoring metals, HCl, and particulate met the appropriate
performance specifications and requirements in the approved AMP using
the annual auditing process.

Table   SEQ Table \* ARABIC  2 .  CEMS Audit Summary

Monitoring System	Audit Description	Audit Date	Result

CO CEMS	Relative Accuracy Test Audit (RATA)	8/24/2010	Pass

CO CEMS	Calibration Drift	8/18-24/2010	Pass

CO CEMS	Calibration Error	8/24/2010	Pass

CO CEMS	Response Time	8/24/2010	Pass

O2 CEMS	Relative Accuracy Test Audit (RATA)	8/24/2010	Pass

O2 CEMS	Calibration Drift	8/18-24/2010	Pass

O2 CEMS	Calibration Error	8/24/2010	Pass

O2 CEMS	Response Time	8/24/2010	Pass

PM CEMS	Response Correlation Audit (RCA)	8/11-12/2010	New Calibration
Curve

PM CEMS	Absolute Correlation Audit (ACA)	8/12/2010	Pass

HCL CEMS	Accuracy (Dynamic Spiking)	8/24/2010	Pass

HCL CEMS	Seven Day Drift	8/22/2010	Pass

Metals CEMS	Total System Flow (Sample Volume)	8/4/2010	Pass

Metals CEMS	X-Ray Fluorescence Calibration Audit	8/9/2010	Pass

Metals CEMS	Accuracy (Dynamic Spiking)	8/5-6/2010	Pass



HCl CEMs Audit Methodology

The accuracy and precision of the HCl CEMS is determined by dynamically
spiking a known concentration of HCl reference gas into the sample
system and subsequently measuring the reference spike with the HCl CEMS.

Daily, prior to testing, the zero and upscale drift is checked and
recorded.  The HCl CEMS must pass the daily calibration requirements
prior to any testing.

This testing was conducted as part of the CMS performance evaluation
test.

Figure 1 provides an overview of the regression analysis for the HCl
dynamic spiking indicating the resulting performance for the HCl CEMs.

PM CEMs Audit Methodology

The objective of the annual relative calibration audit is to verify that
the current calibration correlation is within statistical requirements
for predicting total particulate mass based upon previous calibration to
a particulate mass Reference Method (Method 5).

The RCA is performed by collecting a minimum of 12 valid (meets quality
assurance requirements) particulate mass Reference Method samples
(Method 5) over the expected operating range of the control device.  The
response of the PM CEMS is recorded during each Method 5 test and the
data averaged to provide a PM CEMS response value (PLA) which correlates
to a Method 5 data point (mg/scm).

Dual train Reference Method 5 sampling trains are used during the
testing ensure the quality of the Reference Method particulate mass
data.

Once all data is collected, both the Reference Method 5 and PM CEMS data
are screened, according the Performance Specification 11 and Procedure 2
 guidance, and only valid data is used for the calibration correlation
check.

Using the current calibration correlation as the baseline, the Method 5
data is plotted against the respective PM CEMS averaged response.  A
minimum of 75% of the RCA data must fall within the current tolerance
intervals of the existing calibration correlation.

If the new data does not meet this requirement, Performance
Specification 11 provides guidance on how to establish a new or updated
calibration correlation.

PM CEMS Operating Range

The operating range of the PM CEMS is 0 – 1 PLA.  This range will be
used for operation of the PM CEMS.

PM CEMS Daily Linearity Check 

Daily, prior to any RCA testing, the linearity of the Sigrist photometer
is checked.  The Sigrist must pass applicable quality assurance
requirements, daily, and prior to the relative calibration audit
testing.

Reference Method 5 Data Quality Assurance

Reference Method 5 data collected during the RCA is analyzed to ensure
it meets applicable quality control requirements.  This is completed by
performing relative standard deviation for each Method 5 test pair (dual
train) along with linear regression analysis.  In addition, a
standardized residual test along with a bias check of the data may be
used to ensure the Reference Method paired data sets are statistically
sound for use in the relative calibration audit test.  Reference Method
data which does not meet the quality assurance guidance may be
discarded, as long as a minimum of 12 Reference Method data points
remain.  

Figure   SEQ Figure \* ARABIC  2 .  Calibration Correlation between the
Sigrist PM CEMs and Method 5 train

Particulate Emission Range of Data during the Relative Calibration Audit

The range of the total particulate emissions during the RCA should
represent expected particulate emission levels during normal operation
of the incinerator. 

Per the T149 Alternative Monitoring Petition, Evonik will strive to
provide a range of particulate emissions in which a minimum of 20% of
the data falls into each of three ranges (1) 0-50% (2) 25-75% (3)
50-100%, with 100% being the highest Sigrist output observed during the
testing.  In addition, the data set of particulate mass emissions should
fall within the data range represented by the current calibration
correlation.  The total particulate emission data collected, based upon
the mg/scm concentrations, fell within three distinct ranges

Figure   SEQ Figure \* ARABIC  3 .  Performance Test Fit of Data for
Sigrist PM CEMs

Xact Multi-Metal CEMs Audit Methodology

The annual audit requirements for the Xact multi-metal CEMS are to
perform a sample volume audit, a thin film standard audit, and an
accuracy test.  The accuracy test, per the T149 Alternative Monitoring
Petition, is performed using dynamic spiking of lead, cadmium, arsenic,
chromium, and mercury. Linear regression is used to assess the accuracy
and precision of the Xact CEMS.  The Xact CEMS met each of the audit
requirements.

The accuracy and precision of the Xact CEMS is demonstrated by
quantitatively spiking each of the regulated metals (arsenic, chromium,
cadmium, lead, mercury) at a minimum of three levels.  The metals are
contained in a stock nitric acid solution and quantitatively spiked into
the sample system, directly in back of the sample probe, using a
quantitative aerosol generator. The concentration for each spiking trial
is controlled by altering the mass loss rate of the aerosol generator
and/or changing the concentration of the stock metals solution. 

The range of metal concentrations tested encompasses the regulatory
limits for each metal and are within the linear range of the Xact CEMS.

For each test level, a minimum of nine sample points are recorded.  The
output of the Xact multi-metals CEMS, for each MACT metal, is plotted
against the reference metal value from the dynamic spiking.  Regression
analysis is performed and the correlation coefficient, slope, and
intercept, for each MACT metal recorded.

Figure   SEQ Figure \* ARABIC  4 .  Plot of Linear Regression for Lead
from Xact CEMs

Figure   SEQ Figure \* ARABIC  5 .  Plot of Linear Regression for
Mercury from Xact CEMs

Figure   SEQ Figure \* ARABIC  6 .  Plot of Linear Regression for
Arsenic from Xact CEMs

Figure   SEQ Figure \* ARABIC  7 .  Plot of Linear Regression for
Cadmium from Xact CEMs

Figure   SEQ Figure \* ARABIC  8 .  Plot of Linear Regression for
Chromium from Xact CEMs

COMPREHENSIVE PERFORMANCE TEST RESULTS

The T149 CPT was conducted September 28-29, 2010, utilizing the of a
particulate matter CEMS, a multi-metals CEMS, and an HCl CEMS.  

Table   SEQ Table \* ARABIC  3 .  T149 HWC MACT Compliance Performance
and Emissions Summary

 	 	HWC MACT 	T149 2010 CPT Results

Parameter	Units	Standard	Run 1	Run 2	Run 3	Average













Stack gas particulate matter (Note a)	gr/dscf	0.013	 	0.013	 	0.012	 
0.012	 	0.012

Stack gas HCl/Cl2 (Notes a, b)	ppmv, dry	32	 	4.2	 	3.0	 	3.9	 	3.7

Stack gas LVM (Note a)	µg/dscm	92	<	2.7	<	2.7	<	3.3	<	2.9

Stack gas SVM (Note a)	µg/dscm	230	 	19	 	17	 	18	 	18

Stack gas mercury (Note a)	µg/dscm	130	 	21	 	23	 	18	 	21















Notes:

(a)  Corrected to 7% oxygen.

(b)  HCl and Cl2 combined, expressed as HCl or Cl- equivalents.







The following figures (Figures 9 -13) show the operating data from
theT149 CEMs systems during the performance of the CPT.

Figure   SEQ Figure \* ARABIC  9 .  

Figure   SEQ Figure \* ARABIC  10 .  

Figure   SEQ Figure \* ARABIC  11 .  

Figure   SEQ Figure \* ARABIC  12 .  

Figure   SEQ Figure \* ARABIC  13 .  

Table   SEQ Table \* ARABIC  6 .  Metals SRE Comparison

Metal	2010 (CEMs)	2005 (Method 29)

Total LVM	99.998%	99.998%

Pumpable LVM	99.93%	99.85%

SVM	99.83%	99.74%

Hg	45.70%	17.6%





SUMMARY

Specific conclusions drawn from the 2010 CPT are as follows:

PM emission standard was met.  A maximum ash feed rate limit can be
appropriately developed from the T149 CPT results.

Metal emission standards were met for Hg, SVM, and LVM.  Maximum metal
feed rates can be reliably and appropriately determined using the T149
CPT results.

Stack gas HCl/Cl2 emission standard was met.  A maximum total chlorine
feed rate limit can be appropriately established from the T149 CPT
results.

Use of CEMs is a viable approach for CPT compliance demonstration.

For certain incineration facility, a number of benefits can be realized
through use of CEMs for demonstration of compliance for emissions:

Improved safety, improved operation efficiency, and economic advantages
for waste characterization

Operational flexibility for the incineration system

Optimization of the establishment of operating parameter limits during
the performance of the CPT.

ACKNOWLEDGEMENTS

REFERENCES

American Society for Testing and Materials, "Annual Book of ASTM
Standards," latest annual edition.

Code of Federal Regulations, 40 CFR 264, 40 CFR 270, 40 CFR 63 Subpart
EEE, and 40 CFR 63 Subpart A.

USEPA, "New Source Performance Standards, Test Methods and
Procedures,” Appendix A, 40 CFR 60.

USEPA, "Test Methods for Evaluating Solid Wastes," Third Edition, 1986,
and Updates.

T149 SOLID-LIQUID WASTE INCINERATOR, COMPREHENSIVE PERFORMANCE TEST
PLAN, Evonik Degussa Corporation., Tippecanoe Laboratories, September
2009

CONTINUOUS MONITORING SYSTEMS PERFORMANCE EVALUATION TEST REPORT, ,
Evonik Degussa Corporation., Tippecanoe Laboratories, December, 2010

T149 SOLID-LIQUID WASTE INCINERATOR, COMPREHENSIVE PERFORMANCE TEST
REPORT, Evonik Degussa Corporation., Tippecanoe Laboratories, December
2010

INTIAL PERFORMANCE TEST AND USE OF ALTERNATIVE MONITORING SYSTEMS FOR
METALS, PM AND HCL FOR COMPLIANCE WITH THE HAZARDOUS WASTE COMBUSTOR
MACT, Eli Lilly and Company, Richard H Lambert, Michael L Foster

KEY WORDS

1. Continuous Emission Monitoring Systems

2. Hazardous Waste Combustor MACT Emission Compliance

3. Incinerator Emission Monitoring

4. Multi-Metals Continuous Emission Monitoring

5. Total Particulate Continuous Emission Monitoring

6. Hydrogen Chloride Continuous Emission Monitoring

7. Comprehensive Performance Test

í

]

x

—

Å

Æ

Ç

Ô

í

Ç

Ô

	”

kdH

欀Ṥ

  PAGE   \* MERGEFORMAT  20 

Table   SEQ Table \* ARABIC  4 .  T149 CPT Particulate, HCl, and Arsenic
Emissions Summary

Table   SEQ Table \* ARABIC  5 .  T149 CPT Cadmium, Chromium, Lead and
Mercury Metals Emissions Summary

