Supplement to EPA Compendium Method T0-15 Reduction of Method Detection Limits to Meet Vapor Intrusion Monitoring Needs


Supplement to EPA Compendium Method TO-15—
Reduction of Method Detection Limits to Meet Vapor
Intrusion Monitoring Needs
£. Hunter Daughtrey, Jr., Karen D. Oliver, and H. Herbert Jacumin, Jr.
ManTech Environmental Technology, Inc., P.O. Box 12313, Research Triangle Park, NC
27709
William A. McCIenny
National Exposure Research Laboratory, U.S. Environmental Protection Agency,
Research Triangle Park, NC 27711
ABSTRACT
The Supplement to EPA Compendium Method TO-15 provides guidance for reducing the
method detection limit (MDL) for the compound 1,1-dichloroethene (1,1-DCE) and for
other volatile organic compounds (VOCs) from 0.5 parts per billion by volume (ppbv), as
cited in Method TO-15, to much lower concentrations. Revisions were made to the
wording of Method TO-15 where the original language proved limiting to the goal of
extending Method TO-15 to low parts per trillion by volume (pptv) levels or where minor
omissions were observed. Also, recommendations in the form of additions were made on
aspects of laboratory procedure deemed critical to low-pptv-level analysis. Specifically,
the MDL for 1,1-DCE was determined to be 6 pptv. During this effort, a capability for
preparing 1,1-DCE sample concentrations of 30 pptv and 60 pptv in ambient air was
developed. Using this capability and the capability to prepare samples of humidified zero
air, samples were prepared in canisters and sent to three contract laboratories as
unknowns. Subsequent comparison of results indicated close agreement among the
laboratories while maintaining the performance standards for replicate precision (25%)
and audit accuracy (30%) originally specified in Method TO-15. The following
compounds were also detected at low pptv levels in canisters filled with spiked ambient
air: chloroethene, dichloromethane, cis-1,2-dichloroethene, trichloromethane, 1,2-
dichloroethane, benzene, 1,1,1-trichloroethane, trichloroethene, and tetrachloroethene.
Since the different laboratories employed different analytical procedures, the use of a
performance-based method appears justified.
INTRODUCTION
TO-15 is a performance-based method prepared by the United States Environmental
Protection Agency (EPA) as a guidance document for monitoring subsets of those
volatile organic compounds (VOCs) that are mentioned in Title III of the Clean Air Act
Amendments of 1990.1 The TO-15 performance criteria are based on data from existing
databases compiled in national monitoring programs (e.g., the Toxics Air Monitoring
System [TAMS] and Urban Air Toxics Monitoring Program [UATMP]) using canister-
based sampling and bench-top quadrupole mass spectrometers. These performance
criteria specify a method detection limit (MDL), a method replicate precision, and a
1

-------
method audit accuracy. The sampling and analytical approaches are not restricted in
any sense as long as the performance criteria are met. Examples of possible approaches
to analysis, generation of calibration mixtures, and use of quality control measures
(technical acceptance criteria) are provided in the text of TO-15. These examples are
intended to be instructive, not prescriptive.
TO-15 has the following performance specifications: a MDL, defined as 3.143 times the
standard deviation for seven replicates, of 0.5 parts per billion by volume (ppbv);
replicate precision, defined as [(A - B) x 200] / (A + B), of ± 25%; and accuracy, defined
as [(Measured - True) x 100] / True, of ± 30%. The goal of the supplement is to extend
the MDL to concentration levels needed to achieve the 10"6 risk levels (see Table 1).
Further details concerning 10"6 risk levels are available at www.epa.gov/iris.
Table 1.10"6 Risk levels for NAT A compounds.

Risk Level

Risk Level
Compound
(pptv)
Compound
(pptv)
Vinyl chloride
90
trans-1,3-Dichloropropene
44
1,1 -Dichloroethene
50
1,1,2-Trichloroethane
11
Dichloromethane
576
1,2-Dibromoethane
1
Trichloromethane
8
Tetrachloroethene
NE
1,2-Dichloroethane
10
1,1,2,2-T etrachloroethane
3
Benzene
41
Hexachlorobutadiene
5
Carbon tetrachloride
11
Acrylonitrile
5
1,2-Dichloropropane
NE
1,3-Butadiene
1
Trichloroethene
NE
Ethylene oxide
NE
cis-1,3-Dichloropropene
44


NE = not established
We prepared caiiisters filled with various levels of 1,1-DCE in a mixture and as a single
compound in ambient air, as well as canisters filled with humidified zero air. These
samples have been analyzed by four laboratories to obtain an idea of the agreement
expected and to verify that low concentration levels corresponding to 10"6 risk levels can
actually be quantified. While these tests demonstrate how well such samples are likely to
be analyzed, it does not mean that a non-canister approach to sampling would not do as
well or better.
In summary, the supplement acknowledges the need for sampling and analytical
protocols that reduce the MDLs for certain types of measurements and provides examples
of achieving this reduction.2 The analytical guidelines developed by the Colorado
Department of Public Heal th and Environment (CDPHE) for use by its contract
laboratories, for example, provide a useful and practical approach for current monitoring
applications. The agreement among the four laboratories establishes that more than one"
2

-------
analytical approach is viable and, furthermore, that the preparation of canisters and
standards for sampling 1,1-DCE is possible at low parts per trillion by volume (pptv)
levels. The extension to other single compounds and to multiple compounds should be
straightforward.
TEST METHODS
To verily that TO-15 can be modified to meet the requirements for monitoring at 10"6 risk
levels, the factors needed to improve detection limits must be identified and the
performance of methods must be tested.
Improving Method Detection Limits
Canister Cleanliness
Canister cleanliness is the first key to improving detection limits. TO-15 specifies 0.2
ppbv as a canister-cleanliness standard, developed when MDL goals were more relaxed.
Blank values must be less than MDLs. Our experience has been that to achieve a realistic
measure of MDLs, humidified scientific-grade (HSA) air must be used for the evaluation.
We also strongly recommend that an oil-free vacuum pump be used for evacuation of the
canisters, as backstreaming of vacuum pump oil vapors can contaminate the canisters.
Use of a humidified airstream as part of the cleaning cycle has been shown to greatly
improve the attainable MDLs. Finally, it is critical that canister cleanliness be verified by
periodic testing.
Standards
Two issues are raised for the calibration and quality control samples to be used. The first
is that the standards should be traceable to those prepared by the National Institute of
Standards and Technology (NIST). It is preferable to have a NIST standard of the
compounds of interest, but this is difficult to obtain for all compounds. Calibration for a
subset of the compounds of interest for which NIST standards are available, coupled with
certification of NIST-traceable gravimetric standards for the balance of the compounds, is
acceptable when NIST standards are not available. No matter which method is chosen,
monitoring of standard concentration over the life of a cylinder is needed. We have our
standards periodically analyzed by another researcher who uses a NIST-traceable propane
standard for gas chromatography (GC) with flame ionization detection (FID) analysis.
The second issue is that the calibration standard must be appropriate to the expected
analytical range. For our "normal" TO-14 or Photochemical Assessment Monitoring
Station (PAMS) analyses, we usually employ a 10-ppbv working standard prepared by
dynamic dilution of a 2 parts per million by volume (ppmv) cylinder standard. This is
clearly too high to be able to accurately calibrate at the tens of parts per trillion range
required for this study. We procured a 10-ppbv 1,1-DCE standard, which allowed for
preparation of standards as low as 2.5 pptv using our dynamic dilution technique.
3

-------
Storage Stability
Demonstrated storage stability of 30 days is needed to allow for transit time and for
unanticipated delays. A goal of no change greater than 20% over the 30-day period is
desired.
GC-MS Conditions
Two methods were considered for evaluation. Our laboratory used sorbent
preconcentration followed by GC-ion trap detection.2,3 CDPHE proposed selected ion
monitoring mass spectrometry (MS) as a candidate method to achieve the required
detection limits.4 This does not preclude other methods that meet the performance
criteria.
Performance Testing
Choice of Laboratories
We requested from CDPHE the names and contact information for contract laboratories
with whom they did business. We contacted three of these to obtain information
regarding services they offered, prices, and turnaround time. We ordered each
laboratory's modified TO-15 analysis with lull data report and one TO-14 analysis.
Experimental Design
The following test samples were generated:
¦	Samples to test canister cleanliness (3 cans/lab)
¦	1,1 -DCE at 20-40 pptv in a humid ambient air matrix (3 cans/lab) to test the
method slightly above the detection limit
¦	1,1-DCE at 50-80 pptv in the presence of a mixture of 14 chlorinated VOCs in a
humid ambient air matrix (2 cans/lab) to test at a slightly higher level in the
presence of other chlorinated VOCs
¦	1,1-DCE at low ppbv levels in the presence of ppbv levels of 60 hydrocarbons in
a synthetic air matrix using Method TO-14 (1 can/lab) to test each contract
laboratory's ability to analyze the TO-14 target list
RESULTS AND DISCUSSION
Canister Cleanliness
Clean, evacuated canisters were received from the contract laboratories. We filled the
canisters with HSA and analyzed them after they were allowed to sit for 24 hours. The
HSA-filled canisters were then returned to the contract laboratories for analysis. All
results for modified TO-15 analytes were below detection limits for our laboratory/The
same held true for the contract laboratories except laboratory 1 found four compounds
slightly above their reporting limits and laboratory 3 found two compounds above their
detection limits but below the practical quantitation limits. Toluene, which was not on the
4

-------
target list, was consistently above the MDL specified by each laboratory, including one
canister at 714 pptv that would have failed the original TO-15 criteria.
Analysis of Synthetic Samples
Nine canisters loaded with a nominal 30 pptv 1,1-DCE standard diluted with ambient air
were analyzed in our laboratory and. of these, three canisters were submitted to each
contract laboratory for analysis. The results are shown graphically in Figure I. As can be
seen, the agreement between the nominal concentration and the results from our
laboratory and the three contract laboratories (CL 1, CL 2. and CL 3) is well within the
criteria for precision and accuracy
Figure 1: Comparison of Results of Analysis for 1,1-DCE of
Ambient Air Spiked with a 30-pptv 1,1-DCE Standard by ManTech
and Three Contract Laboratories
35
2
Q.
CL
C
o
•a
re
30
25
20
t 15
c
-------
This was also reflected in the results for other compounds for laboratory 3, but not to the
same extent.
Figure 2: Comparison of Results of Analysis for 1,1-DCE of Ambient
Air Spiked with a Chlorinated Gas Mixture Containing 60 pptv 1,1 -DCE
by ManTech and Three Contract Laboratories
Can 1
Can 2
Nominal
MDL
MT
Scan
CL 1
SIM
MT
Scan
CL 2
SIM
MT
Scan
CL 3
SIM
Analysis Lab
Figure 3: Comparison of Results of Analysis for 1,1-DCE of
PAMS/Terpene Standard Spiked with a 5-pptv 1,1-DCE Standard
by ManTech and Three Contract Laboratones
>
XJ
a
a
c
o
is
c
a>
o
c
o
o
15
10
iFound
• Nominal
MT
Scan
CL 1
SIM
MT
Scan
CL 2
SIM
MT
Scan
CL 3
SIM
Analysis Lab
6
-------
CONCLUSIONS
• The TO-15 Supplement2 provides guidance for sampling and analysis of 1,1-DCE,
and by implication other VOCs, in air at levels lower than the TO-15 MDL of 0.5
ppbv, with the specific level depending on the data quality objectives (DQOs) for the
project at hand. The performance criteria are an MDL at the customized DQO levels,
replicate precision of at least 25%, and audit accuracy of 30%.
• The supplement includes revisions and additions by section to the original Method
TO-15. Two examples of technical approaches to meet the performance criteria are
provided: One is the guidance developed during this project by EPA on-site
contractor ManTech Environmental Technology, Inc., and the other is a concise
restatement of the guidance developed by CDPHE for the analysis of high-risk
compounds associated with the problem of vapor intrusion into buildings.
• Samples of 30 and 60 pptv of 1,1-DCE in ambient air prepared by ManTech
Environmental Technology, Inc., were analyzed by four laboratories, and the results
showed that the TO-15 Supplement performance criteria could be met at
concentrations as low as 30 pptv. One of the laboratories was the EPA on-site
laboratory operated by ManTech, and at least one of the other contract laboratories
used the CDPHE guidance.
RECOMMENDATIONS
• The technical acceptance criteria provided in the original TO-15 and in the TO-15
Supplement must be recognized as guidance. Other technical acceptance criteria can
be used for meeting the performance criteria of TO-15 and the TO-15 Supplement.
This point is evidenced by the close agreement of results obtained by four
independent laboratories analyzing identical samples, each using their own standard
operating procedures.
• Laboratories that wish to perform analyses of VOCs at low pptv levels must exercise
diligence in all aspects related to cleanliness (canister cleanup and certification,
carryover issues, instrument background levels, etc.). In addition, accurate calibration
standards at the appropriate concentrations must be obtained or generated. Finally, the
MS method will need to be optimized according to the specific analytical system used
and the analyte(s) chosen.
• Agreement on the audit standards to be used in monitoring low-level VOCs is
necessary, whether the audit standard is to be the average of analysis results from
different laboratories, diluted NIST-traceable standards from commercial suppliers, or
fundamentally derived standards.
• Caution should be exercised when working at low pptv levels, due in part to the need
for a more rigorous investigation of storage stability and sample integrity issues as
well as a general need for more laboratory tests in the low-pptv range of sample
7
-------
concentrations. Extreme conditions of humidity (<15% RH for any sample and high
humidity for positive-pressure samples) and of co-collected reactive compounds may
complicate the sampling and analytical conditions. More experience is needed in
monitoring at low pptv levels.
• To confirm consistent sampling technique, a number of replicate samples should be
collected and analyzed.
ACKNOWLEDGMENTS
The authors thank Tom Aalto of EPA Region 8 for his assistance in starting this research
and the scientists at CDPHE, particularly Ken Niswonger and Edgar Ethington, for their
input to the TO-15 Supplement. Also, thanks to Bill Lonneman, an experienced senior
scientist now working with EPA as part of the Senior Environmental Employment
program, for providing advice on analytical procedures and helping with the
substantiation of target compound concentration levels in gas standards. The authors
would like to acknowledge the efforts of Stacy Henkle of KulTech, Inc., in editing and
formatting this document.
REFERENCES
1. Compendium of Methods for the Determination of Toxic Organic Compounds in
Ambient Air, 2nd edition; U.S. Environmental Protection Agency. U.S. Government
Printing Office: Washington, D.C. 1999; EPA/625/R-96/0106.
2. Oliver, K.D.; Jacumin, H.H., Jr.; Daughtrey, E.H., Jr., McClenny, W.A. Method
TO-15 Supplement: Analysis of 1,1-DCE at pptv Concentrations; U.S. Environmental
Protection Agency. U.S. Government Printing Office: Washington, D.C. 2002;
EPA/600/R-03/109.
3. Oliver, K.D.; Adams, J.R.; Daughtrey, E.H., Jr.; McClenny, W.A.; Yoong, M.J.;
Almasi, E.B.; Kirshen, N.A. Environ. Sci. Technol. 1996,30,1939-1945.
4. Guidance for Analysis of Indoor Air Samples—April 2000; Colorado Department of
Public Health and Environment, Hazardous Materials and Waste Management
Division, Denver, CO.
KEYWORDS
1,1-DCE
1,1-dichloroethene
method detection limit
parts per trillion
TO-15
vapor intrusion
VOC
volatile organic compound
8
-------
DISCLAIMER
This work has been partially funded by the United States Environmental Protection
Agency under Contract 68-D-00-206 to ManTech Environmental Technology, Inc. This
paper has been reviewed in accordance with the Agency's peer and administrative review
policies and approved for presentation and publication. Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.
9
-------
TECHNICAL REPORT DATA

1. Report No. 2.

4. Title and Subtitle
Supplement to EPA Compendium Method TO-15 - Reduction of Method
Detection Limits to Meet Vapor Intrusion Monitoring Needs
5. Report Date
31 March 2004
6. Performing Organization Code
7. Author(s)
E. Hunter Daughtrey, Jr., Karen D. Oliver, H.Herbert Jacumin, Jr. and William
A. McClenny
8. Performing Organization Report
No.
9.Performing Organization Name and Address
ManTech Environmental Technology, Inc.
P.O. Bos 12313
Research Triangle Park, NC 27711 J
10. Program Element No.
11. Contract/Grant No.
12.Sponsoring Agency Name and Address
National Exposure Research Laboratory
109 T.W. Alexander Drive
Research Triangle Park, NC 27709
13. Type of Report and Period
Covered
14.Sponsoring Agency Code
15. Supplementary Notes
16. Abstract
The Supplement to EPA Compendium Method TO-15 provides guidance for reducing the method detection limit (MDL)
for the compound 1,1-dichloroethene (1,1-DCE) and for other volatile organic compounds (VOCs) from 0.5 parts per
billion by volume (ppbv), as cited in Method TO-15, to much lower concentrations. The MDL for 1,1-DCE was
determined to be 6 pptv. During the experimental effort, a capability for preparing 1,1-DCE sample concentrations of 30
pptv and 60 pptv in ambient air was developed. Samples were prepared and sent to three contract laboratories as
unknowns. Subsequent comparisons of results indicated close agreement among the laboratories while maintaining the
performance standards for replicate precision (25%) and audit accuracy (30%) originally specified in Method TO-15.
Since the different laboratories employed different analytical procedures, the use of a performance-based method appears
justified.
17. KEY WORDS AND DOCUMENT ANALYSIS
A. Descriptors: dichloroethene, method detection limit,
vapor intrusion, volatile organic compound
B. Identifiers / Open Ended
Terms
C. COSATI

18. Distribution Statement
19. Security Class (This
Report)
21. No. of Pages
20. Security Class (This Page)
22. Price
Form Available: Network Neighborhood\Knight\Groups\HEASDYForms\Technical-Report-Data-2220-l
-------