United States	National Exposure

Environmental Protection	Research Laboratory

Agency	Research Triangle Park, NC 27711

Research and Development	EPA/600/SR-97/033 June 1997

oEPA Project Summary

Field Test of a Generic
Method for Halogenated
Hydrocarbons: SemiVOST
Test at an Agricultural Chemical
Manufacturing Facility

Joan T. Bursey, James F. McGaughey, and Raymond G. Merrill

A field evaluation study was con-
ducted for the halogenated semivolatile
organic compounds (SVOCs) listed in
Title III of the Clean Air Act Amend-
ments (CAAA) of 1990 that may be
sampled and analyzed by the
SemiVOST method. The performance
of the halogenated SVOCs in the test
methodology had been evaluated in the
laboratory and in two previous field
studies. Dynamic spiking techniques
were applied in the field, using a sam-
pling strategy statistically designed to
meet the requirements of EPA Method
301 for method evaluation. The second
field test provided insufficient valid
SemiVOST data to allow thorough sta-
tistical evaluation of the results because
of problems encountered in the prepa-
ration of the samples for analysis. A
protocol to address SemiVOST sample
preparation was written (Draft Method
3542), and this new sample prepara-
tion protocol was applied to this field
evaluation to improve compound re-
covery and precision. Using the crite-
ria for acceptable method performance
in the EPA Handbook for Hazardous
Waste Incineration, 25 of 28 analytes
tested show acceptable performance in
the SemiVOST method. Using the crite-
ria of EPA Method 301 for acceptable
performance, 18 of 28 analytes show
acceptable performance in the
SemiVOST method.

This Project Summary was developed
by EPA's National Exposure Research
Laboratory, Research Triangle Park, NC,
to announce key findings of the re-

search project that is fully documented
in a separate report of the same title
(see Project Report ordering informa-
tion at back).

Introduction

The evaluation of a stationary source test
method for a particular analyte or group of
analytes requires that the precision and bias
of the method be established experimentally.
The SemiVOST (Semivolatile Organic Sam-
pling Train, SW-846 Method 0010 for sam-
pling and Method 8270 for analysis), appli-
cable to all organic compounds with boiling
points above 100°C, was used to evaluate
the halogenated SVOCs listed in the CAAA
The U.S. Environmental Protection Agency
(EPA), under the authority of Title III of the
CAAA of 1990, has evaluated selected halo-
genated SVOCs in the laboratory1 and in field
tests at two different source categories,2'3
The second field test, at a high moisture
source, did not produce sufficient data to
perform a Method 301 statistical test.

Lack of clarity in the existing sample
preparation procedure was identified as
the cause of the loss of SemiVOST data
in the second field test. To clarify the
existing sample preparation procedure,
SW-846 Draft Method 3542 was written to
address sample preparation. As a test of
the new procedure and to provide method
evaluation data for a second source cat-
egory, a third method evaluation field test
was conducted. To challenge the sample
preparation methodology, a test site was
selected at which moisture levels were
high (approximately 55%). At this test site,
a field evaluation of the SemiVOST sam-

-------


pling procedure was performed with dy-
namic spiking of halogenated SVOCs.
Quadruple collocated probes with four simi-
lar sampling trains (two spiked trains, two
unspiked trains) were used. EPA Method
301 provided statistical guidelines for de-
sign of the sampling strategy.

Procedure

The field evaluation was conducted at a
chemical manufacturing facility that oper-
ates a multipurpose incinerator to burn
aqueous waste, with small amounts of
chloroacetic acid, trichloroethylene, and
toluene. During a presurvey visit to the
test site, grab samples for SemiVOST were
taken and analyzed in order to character-
ize the test matrix and evaluate any po-
tential interferents. No significant levels of
any of the compounds of interest were
present in the background emissions ma-
trix.

Sampling was performed by withdraw-
ing stack gas from a single port in the
stack, through a quad probe, into four
similar sampling trains. Two of the trains
for each quad sampling run were dynami-
cally spiked and two were unspiked. Sam-
pling methodology followed SW-846
Method 0010, except that

•	a quad probe was used instead of
the regular single probe;

•	a heated glass elbow equipped with
a dynamic spiking injection port con-
nected the probe to the heated filter;

•	because of the high moisture levels,
a single condenser was not sufficient
to cool the stack gas entering the
XAD-2® module to the temperature of
20°C (68°F) required by Method 0010.
A second condenser with its own re-
circulating pump was added to cool
the stack gas sufficiently.

For dynamic spiking, halogenated
SVOCs were introduced into the sampling
system in methylene chloride solution by
syringe injection through the heated glass
elbow. Liquid feed rates of the spiking
solution were metered by motor-driven sy-
ringe pumps.

Reagent and field blanks were col-
lected, with one field blank for every three
Quad sampling runs.

Train components were prepared for
analysis using the Draft Method 3542 pro-
tocol, "Preparation of Modified Method 5
(SW-846 Method 0010) Train Components
for Analysis by SW-846 Method 8270."
The parts of the SemiVOST train yield
three extracts to be analyzed according to the

procedures of Method 8270: (1) the particu-
late matter/filter extract, combined with the
extract of the front half rinse, (2) the conden-
sate rinse and condenser rinse fractions, and
(3) the combination of the XAD-2® extract with
the rinse of the back half of the filter holder
and the rinse of tubing connecting the filter
holder to the condenser. A critical step in the
preparation of SemiVOST train samples, es-
pecially with high levels of moisture present
during sampling, is the correct transfer of the
wet XAD-2® from the sampling module to the
Soxhlet extractor, a procedure specifically ad-
dressed by Draft Method 3542.

Analyses were performed according to
the protocol of Method 8270, with the fol-
lowing exceptions:

•	Each final extract volume was 5 mL,
rather than 1 mL as specified in
Method 8270 for the extraction of wa-
ter or soil;

•	Filters, XAD-2®, and condensate were
extracted separately to generate three
extracts for analysis;

•	Impinger contents were archived.
Results and Discussion

Overall recoveries for spiked surrogate com-
pounds (spiked in the laboratory immediately
prior to preparation of samples) were 92.3%,
so no correction for surrogate recoveries was
made in the statistical treatment of the data.
As anticipated, the most and highest recover-
ies for the dynamically spiked compounds are
obtained from the XAD-2®. Only the least
volatile compounds are recovered from the
filter and front half rinse, including hexachloro-
benzene, pentachlorophenol, pentachloronitro-
benzene, chlorobenzilate, and 3,3'-
dichlorobenzidine. Possibly because of the
high level of moisture in the source, relatively
high (> 10% of the 500-600 ng spiked) recov-
eries are obtained from the condensate for
epichlorohydrin, dichloroethyl ether, and
hexachlorobenzene. Other compounds fre-
quently observed in the condensate ex-
tracts include pentachloronitrobenzene,
chlorobenzilate, 2,4,5-trichlorophenol, 2,4,6-
trichlorophenol, hexachlorobenzene, and 2-
chloroacetophenone.

The "Protocol for the Field Validation of
Emission Concentrations from Stationary
Sources" (EPA Method 301) was used to
calculate both bias and precision of emission
concentration data. Semrvolatile halogenated
organic compounds that have bias correction
factors in the 0.70 to 1.30 window show ac-
ceptable performance in the sampling and
analytical method. Precision estimates equal
to or less than 50% are considered accept-

able. According to the criteria of EPA Method
301, the following compounds demonstrated
a correction factor and precision within the
acceptable range: frans-1,2-dichloropropene,
1,1,2-trichloroethane, ethylene dibromide, bro-
moform, 1,1,2,2-tetrachloroethane, dichloroet-
hyl ether, benzyl chloride, 1,4-dichlorobenzene,
1,2-dibromo-3-chloropropane, hexachloroet-
hane, 1,2,4-trichlorobenzene, hexachlorobuta-
diene, 2,4,6-trichlorophenol, 2,4,5-trichloro-
phenol, hexachlorobenzene, pentachloronitro-
benzene, and chlorobenzilate. The following
compounds failed to meet the acceptance
criteria of EPA Method 301: epichlorohydrin,
bis(chloromethyl) ether, c/s-1,3-dichloro-
propene, tetrachloroethylene, chlorobenzene,
benzotrichloride, 2-chloroacetophenone, hexa-
chlorocyclopentadiene, pentachlorophenol, and
3,3'-dichlorobenzidine. Chloroacetic acid was
not recovered.

Statistical calculations for the SemiVOST
compound recoveries were also performed
using the methods presented in the EPA
Handbook for QA/QC Procedures for Haz-
ardous Waste Incineration,5 To meet the
acceptance criteria for the EPA Handbook,
the compound recovery must be in the
range of 50% to 150%, with a precision of
50% or less. The following compounds met
the acceptance criteria of the EPA Hand-
book: epichlorohydrin, c/s-1,3-dichloro-
propene, frans-1,3-dichloropropene, 1,1,2-
trichloroethane, ethylene dibromide, tetra-
chloroethylene, chlorobenzene, bromoform,
1,1,2,2-tetrachloroethane, dichloroethyl
ether, benzyl chloride, 1,4-dichlorobenzene,
1,2-dibromo-3-chloropropane, hexachloroet-
hane, 1,2,4-trichlorobenzene, hexa-
chlorobutadiene, benzotrichloride, 2-
chloroacetophenone, 2,4,6-trichlorophenol,
hexachlorocyclopentadiene, 2,4,5-trichloro-
phenol, hexachlorobenzene, pentachlo-
rophenol, pentachloronitrobenzene, and
chlorobenzilate. Bis(chloromethyl) ether and
3,3'-dichlorobenzidine failed to meet accep-
tance criteria, and chloroacetic acid was
not recovered at all. The compound recov-
eries in this third field test are all higher
than the compound recoveries in the first
field evaluation study.2 The difference in
recoveries is consistent and is usually in
the range of 10% to 20% higher in the
second data set from the third field test.
However, the most noticeable difference in
the two field data sets is the precision:
most of the values for the first data set are
in the acceptable range (< 50% relative
standard deviation), whereas most values
for the second field test are 20% relative
standard deviation or less. Careful and con-
scientious application of the procedures

2

-------
of Draft Method 3542 raised the recover-
ies and improved the precision of the final
data set.

Conclusions

Based on the field evaluation of the
SemiVOST method for halogenated SVOCs
listed in Title III of the CAAA, the following
conclusions may be drawn:

•	Most of the'semivolatile halogenated
organic compounds selected for test-
ing show successful performance in
the SemiVOST methodology: 18 of
28 compounds meet EPA Method 301
test criteria, and 25 of 28 compounds
meet EPA Handbook test criteria for
successful performance of the
SemiVOST methodology.

•	The only semivolatile halogenated or-
ganic compounds tested that do not
meet any criteria for acceptable per-
formance of the test methodology are
bis(chloromethyl) ether and 3,3'-
dichlorobenzidine.

•	Modified sample preparation proce-
dures described in Draft Method 3542
were tested under severe field condi-

tions at a stationary source with a
moisture level of 55%. When the pro-
cedures of Draft Method 3542 are
followed, acceptable SemiVOST
method performance can be obtained
for the majority of the analytes.

•	In a comparison of the results for
recovery and precision from a test at
a coal-fired power plant (the first
SemiVOST field evaluation test) and
this test at an agricultural chemical
facility, the application of the proce-
dures described in Draft Method 3542
elevated recoveries for most of the
analytes by 10% to 20% and improved
precision by a factor of four.

•	For compounds that show consistent but
low recovery, the SemiVOST method
could be used as a screening method to
establish the presence or absence of
these compounds at a test site.

References

1. Laboratory Validation of VOST and
SemiVOST for Halogenated Hydrocar-
bons from the Clean Air Act Amend-
ments List. Volumes 1 and 2. U.S.

Environmental Protection Agency.
EPA/600/R-93/123a and b. NTIS
PB93-227163 and PB93-227171. July
1993.

2.	Field Test of a Generic Method for Halo-
genated Hydrocarbons. U.S. Environ-
mental Protection Agency. EPA/600/
R-93/101. NTIS PB93-212181.

3.	Field Test of a Generic Method for Halo-
genated Hydrocarbons: SemiVOST
Test at a Chemical Manufacturing Fa-
cility. U.S. Environmental Protection
Agency. In Press. 1996.

4.	Protocol for the Field Validation of
Emission Concentrations from Station-
ary Sources, U.S. Environmental Pro-
tection Agency, EPA/450/4-90-015.
April 1991. Federal Register, Vol. 57,
No. 250. Appendix A to Part 63. De-
cember 1992.

5.	Handbook. Quality Assurance/Quality
Control (QA/QC) Procedures for Haz-
ardous Waste Incineration. EPA/625/
6-89/023. January 1990.

3

-------
Joan T. Bursey, James F. McGaughey, and Raymond G. Merrill are with Eastern

Research Group, Inc., Morrisville, NC 27560.

Merrill D. Jackson is the EPA Project Officer (see below).

The complete report, entitled "Field Test of a Generic Method for Halogenated
Hydrocarbons: SemiVOST Test at an Agricultural Chemical Manufacturing Facil-
ity, " (Order No. PB97-162499; Cost: $44.00, subject to change) will be available
only from:

National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:

National Exposure Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711

United States

Environmental Protection Agency

Center for Environmental Research Information

Cincinnati, OH 45268

BULK RATE
POSTAGE & FEES PAID
EPA

PERMIT NO. G-35

Official Business
Penalty for Private Use $300

EPA/600/SR-97/033

-------