United States
Environmental Protection
Agency
Atmospheric Research and
Exposure Assessment Laboratory
Research Triangle Park. NC 27711
Research and Development
EPA/600/SR-93/123 October 1993
Project Summary
Laboratory Validation of
VOST and SemiVOST for
Halogenated
Hydrocarbons from the Clean
Air Act Amendments List
Joan T. Bursey, Raymond G. Merrill, Jr., Robert A. McAllister and James F.
McGaughey
The Clean Air Act Amendments of
1990, Title ill, present a need for sta-
tionary source sampling and analytical
methods for the list of 189 compounds.
EPA has used Volatile Organic Sam-
pling Train (VOST) and Semivolatile
Organic Sampling Train (SemiVOST)
sampling and analytical methods for
this type of sampling of organic com-
pounds in the past, but these method-
ologies have been completely validated
for only a few of the organic com-
pounds. In this study, the applicability
of VOST and SemiVOST techniques to
Clean Air Act halogenated compounds
has been evaluated under laboratory
conditions. The methods were evalu-
ated first to determine whether the com-
pounds could be analyzed successfully.
For SemiVOST and VOST compounds,
the analytes were analyzed by gas chro-
matography/mass spectrometry (GC/
MS) techniques. Retention times for the
analytes were determined, and refer-
ence mass spectra were generated so
that primary and secondary quantitation
ions could be selected. Recovery of
the compounds from the sorbents was
evaluated, and analytical detection lim-
its were determined from spiked sor-
bents. Quadruple sampling trains were
used to collect replicate samples for
statistical evaluation of the dynamic
spiking techniques for liquids
(SemiVOST). This report presents the
results of the laboratory experiments.
This Project Summary was developed
by EPA's Atmospheric Research and
Exposure Assessment Laboratory, Re-
search Triangle Park, NC, to announce
key findings of the research project
that is fully documented in a separate
report of the same title (see Project
Report ordering information at back).
Introduction
The Clean Air Act Amendments of 1990,
Title III (CAAA), present a need for sta-
tionary source sampling and analytical
methods for the list of 189 analytes. The
U.S. Environmental Protection Agency
(EPA) has used VOST (SW-846 Meth-
ods 0030 and 5040 or 5041) and
SemiVOST (SW-846 Methods 0010 and
8270) sampling and analytical methods
for sampling and analysis of a wide vari-
ety of organic compounds in the past, but
these methodologies have been com-
pletely validated for only a few of the
compounds to which they have been ap-
plied. Validation of the methodology es-
tablishes how well the methodology will
perform for a given compound under a
defined set of conditions, i.e., the bias
and precision when the method is applied
to a given compound at a particular sta-
tionary source.
In this study, the applicability of VOST
and SemiVOST techniques to the CAAA
halogenated organic compounds has been
evaluated under laboratory conditions.
Analytical methods were evaluated first to
determine whether the compounds could
be analyzed successfully. For SemiVOST
compounds, a methylene chloride solu-
tion of the analytes was analyzed by gas
chromatography/mass spectrometry (GC/
MS) techniques. Retention times for the
analytes were determined, and reference
spectra were generated so that primary
and secondary quantitation ions could be
selected. Compounds were also assigned
T/y5 Printed on Recycled Paper
-------
to the closest-editing Internal Standard for
quantitative calculations. For the VOST
technique, a methanol solution of the
analytes was spiked into water and the
analytes were purged from the water in
order to determine retention times, gener-
ate reference mass spectra, assign
quantrtation standards, and select appro-
priate quantttation ions. Compounds which
did not survive the application of the ana-
lytical methodology were dropped from fur-
ther evaluation in the VOST and
SemiVOST methods. In the SemiVOST
method, chloroacetic acid could not be
chromatographed successfully. In the
VOST method, bis(chloromethyl) ether,
chloromethyl methyl ether, and epichloro-
hydrin could not be analyzed. However,
bis(chloromethyl) ether and epichlorohy-
drin were also tested by the SemiVOST
methodology. Modification of the existing
methodology or development of new meth-
ods will be required for the compounds
which could not be analyzed successfully.
Recovery of the compounds from the sor-
bents was evaluated, and analytical de-
tection limits were determined from spiked
sorbents.
Quadruple sampling trains were used
for simultaneous collection of replicate
samples for statistical evaluation of the
dynamic spiking techniques for gases
(VOST) and liquids (SemiVOST). With suc-
cessful execution and statistical evalua-
tion of the dynamic spiking techniques,
the VOST and SemiVOST methodology,
with dynamic spiking, will be subjected to
complete validation in the field. This re-
port presents the results of the laboratory
experiments.
Experimental Procedures
The habgenated compounds listed under
CAM, Title III, that were evaluated under this
set of experiments are listed in Table 1. Neither
2,3,7,8-tetrachbrodibenzodbxin, dibenzofurans,
nor the PCBs were evaluated in this experiment
since EPA has specialized methods for these
compounds in statbnary source sampling and
analysis. Some compounds are listed for both
VOST and SemiVOST evaluation since there is
potential overlap in the range that each method
collects. (VOST is used for compounds boiling
between 30 and 100°C with some albwanoe to
130°C and some extensbn below 30°C with
appropriate precautions, whereas SemiVOST
is used to collect compounds boiling above
100°C.) The overlap occurs in the boiling range
between 100°C and 130°C, where compounds
might be appropriateV assigned to either method.
VOST Method
The GC/MS retention times, character-
istic ions and reference mass spectra were
developed for each of the compounds of
Table 1. Clean Air Act Amendments Halogen Compounds Investigated
Compound
Allyl chloride
bis(Chloromethyl) ether
Carbon tetrachloride
Chlorobenzene
Chloroform
Chloromethyl methyl ether
Chloroprene
1 ,3-Dichloropropylene
Epichlorohydrin
Ethyl chloride
Ethylene dibromide
Ethylene dichloride
Ethylidene dichloride
Methyl bromide
Methyl chloride
Methyl cloroform
Methylene chloride
Methyl iodide
Propylene dichloride
Tetrachloroethylene
1, 1,2-Trichloroethane
Trichloroethylene
Vinyl bromide
Vinyl chloride
Vinylidene chloride
Benzotrichloride
Benzyl chloride
Bromoform
Chloroacetic acid
2-Chloroacetophenone
Chlorobenzilate
1 ,2-Dibromo-3-chloropropane
1 ,4-Dichlorobenzene
3,3'-Dichlorobenzidine
Dichloroethyl ether
Hexachlorobenzene
Hexachlorobutadiene
Hexachlorocyclopentadiene
Hexachloroethane
Pentachloronitrobenzene
Pentachlorophenol
1, 1,2,2-Tetrachloroethane
1,2,4- Trichlorobenzene
2, 4, 5- Trichlorophenol
2,4, 6- Trichlorophenol
Boiling
Point
°C
44-46
106
77
132
60.5 - 61.5
55-57
59.4
105 - 107/730 mm
115- 117
12"
131 - 132
83
57
4*
-24.2'
74- 76
39.8 - 40
41-43
95-96
121
110- 115
86.9
16/750 mm*
-13.4"
30-32
219 - 223
177- 181
150 - 151
189
244 - 245
147
196
173
mp = 765
65 - 67/T5 mm
323 - 326
210 - 220
239
186
328
309.5
147
214
248/740 mm
246
VOST
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
SemiVOST
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
'Below the recommended lower boiling point limit of3O°C for VOST.
interest for the VOST method. The ana-
lytical method was EPA Method 5041. The
GC column used was DB-624, 0.53 ID, 3 u,
film thickness, on a Finnigan-MAT 4500 GC/
MS system. Reference mass spectra, pri-
mary and secondary quantrtation ions for
the compounds, and relative retention
times are available in the complete EPA
report.
To determine the recovery of the com-
pounds from VOST tubes, clean VOST
tubes were spiked with a methanol solu-
tion containing approximately 50 ng of
each compound using the flash evapora-
tion technique. The quantitatbn standards
were spiked into the water purge flask
and spiked tubes were desorbed as a pair
through the purge trap employing stan-
dard VOST methodology. A pair of
unspiked tubes was analyzed as a blank.
The recoveries were based on the com-
parison of the amount calculated to the
amount spiked. The analytical system was
calibrated by spiking the purge water with
methanolic solutions of the compounds of
interest at appropriate concentrations. Dif-
-------
ferences in observed concentrations be-
tween direct desorption from the purge
water and desorption from spiked VOST
tubes at a given level was attributed to
the efficiency of desorption of the com-
pounds from the VOST tubes. Five repli-
cations were performed to provide data
for statistical analysis (Table 2).
Detection limits were determined by Fed-
eral Register procedure. The detection limit
for a range of compounds on VOST tubes
was estimated to be 10-20 ng, based on
previous determinations of VOST method
detection limits for similar compounds. Ten
pairs of VOST sorbent tubes were spiked
at two times the estimated method detec-
tion limit (20 ng). The analytical system
was calibrated using spiked VOST tubes,
according to the procedure specified in
Method 5041. The standard deviations of
the determinations and the actual detec-
tion limits were calculated using the Fed-
eral Register procedure (Table 3).
The entire sampling (Method 0030) and
analytical (Draft Method 5041) procedure
was evaluated using quadruple sampling
trains set up in the laboratory. The qua-
druple trains as originally configured in
the laboratory consisted of four complete
trains with a gaseous dynamic spiking sys-
tem using a certified cylinder of a gas-
eous mixture of the compounds of interest
to provide the spike immediately prior to
the entry of the stack sample to the VOST
sampling train. The spike was a gas mix-
ture in a pressurized cylinder with certified
concentration. The gas was mete red into
each VOST sampling train through a mass
flow controller to control the flow precisely
and Teflon® lines to minimize interaction
of the halogenated compounds with reac-
tive surfaces such as stainless steel. Labo-
ratory experiments demonstrated that the
trains as configured did not obtain propor-
tional response to changes in the appar-
ent flow rate of the pressurized gas. The
metering system was changed to needle
valves with the flow rate verified by mea-
surement with bubble flowmeters before
and after each sampling run. Teflon® lines
were also heat-traced to 130°C all of the
way from the regulator of the gas cylinder
to the entry point into the sampling train.
The exact point of spiking was changed to
ensure that the standard gaseous mixture
was being spiked directly into the flowing
gas stream. Accurate spiking of an accu-
rately known quantity of the compounds
of interest allows a complete evaluation of
the sampling and analysis methodology of
the VOST. A quad train or a dual train is
required during stack evaluation under
Method 301 (Protocol for the Field Valida-
tion of Emission Concentrations from Sta-
tionary Sources). Preliminary laboratory
experiments demonstrated that the modi-
fications to the gaseous dynamic spiking
system on the quadruple VOST trains
could be used to obtain accurate and re-
producible spiking with target analytes.
A Latin Square experimental design was
used to evaluate dynamic spiking of the
quadruple VOST trains in their original
configuration. The Latin Square is a sta-
tistical experimental design that was used
to test run, train, and concentration as
variables to determine if a variable has a
significant effect. The evaluation of the
results of the Latin Square experiment for
the VOST trains demonstrated that a pro-
portionate response to apparent changes
in gas flow rate was not being obtained
and catalyzed the reconfiguration of the
quadruple VOST trains to provide accu-
rate and reproducible spiking.
SemiVOST Method
The GC/MS retention times, character-
istic ions and reference mass spectra were
developed for each of the compounds of
interest for the SemiVOST method. The
GC column used was DB-5, 0.32 mm ID,
30 m , 1.0 p. film thickness, on a Finnigan-
MAT 4500 GC/MS system. Reference
mass spectra, primary and secondary
quantitation ions, and retention times are
found in the complete EPA report describ-
ing this work.
To determine the recoveries of the com-
pounds from the XAD-2® sampling me-
dium, XAD-2® sampling cartridges were
spiked with a methylene chloride solution
containing approximately 250 u.g of each
halogenated organic compound. Surrogate
standards were also spiked into XAD-2®
cartridges to monitor the performance of
the sample preparation methodology. The
spiked cartridges were extracted and con-
centrated employing standard SemiVOST
methodology. An unspiked cartridge was
analyzed as a method blank. The final
volume for analysis was 5 mL, the normal
final extract volume for the SemiVOST
procedure. All recoveries were based on
the amount added. Five replications were
performed to allow calculation of the mean
and standard deviation, with statistical
evaluation of the outliers. The GC/MS sys-
tem was calibrated with methylene chlo-
ride solutions of the compounds of inter-
est, according to the standard SemiVOST
procedure (Table 4).
Method detection limits were determined
by Federai Register procedure. Ten
cleaned XAD-2® sampling cartridges were
spiked at two times the method detection
limit that was estimated from the results of
the recovery study. The actual method
detection limits and standard deviations
were calculated using the Federal Regis-
ter procedure (Table 5).
The entire SemiVOST sampling (Method
0010) and analysis method (SemiVOST;
analytical procedure the same as Method
8270 with modified sample preparation pro-
cedures) was evaluated using quadruple
trains set up in the laboratory. The quad
train consists of four complete trains with
a dynamic spiking system to provide the
spike into a heated line just after the probe.
The liquid dynamic spiking system con-
sisted of a constant flow syringe pump
with Teflon® lines to a glass-lined stain-
less steel needle introduced into the sam-
pling lines just behind the probe. The sy-
ringe pump flow was set to provide about
10 mL of solution over a 2-hour sampling
period with a gas flow rate of 0.5 cfm
through the probe. The dynamic spiking
system temperatures were regulated to
provide a drop of spiking solution at the
beveled tip of a glass-lined stainless steel
needle. The drop was not allowed to
evaporate nor to drop to the heated glass
surface of the sampling line. Use of a
dynamic spiking system allows a com-
plete check of the SemiVOST sampling
and analysis methodology. Either a dual
or a quad train is required during station-
ary stack emissions evaluation under EPA
Method 301. In order to statistically evalu-
ate the train and allow for run and train
differences and spiking levels, a Latin
Square design was used. In the Latin
Square there were four replications, four
spiking levels and four trains. The spiking
levels were 100, 300, 500 and 700 jig of
each compound. In the laboratory, the
sampling trains were operated at stack
temperature conditions with nitrogen gas
as the diluting gas to make up the total
volume required for SemiVOST sampling
(Table 6).
Results
GC retention times, reference mass
spectra, and primary and secondary ions
used for the quantitative calculations de-
veloped for each compound are available
in the full report and are not presented
here.
VOST. All of the candidate VOST target
compounds except bis(chloromethyl) ether,
chloromethyl methyl ether, and epichloro-
hydrin were chromatographable and could
be identified using the VOST GC/MS ana-
lytical procedure.
The results of the spiking studies of the
VOST sorbents are presented in Table 2.
The recoveries range from 90.4% for vinyl
chloride to 127.2% for allyl chloride. All of
the compounds that were observed
-------
Table 2. Recoveries of Compounds from VOST Sorbents (Tenax GC®- Tenax Gd*/Petroleum-
Based Charcoal)
Compound
Ethyl chloride
Ethylene dichloride
Methyl iodide
Altyl chloride
Methylene chloride
Ethylidene dichloride
Chloroprene
Methyl chloride
Chloroform
Carbon tetrachloride
1 ,2-Dichloroethane
Vinyl chloride
Trichloroethylene
Propylene dichloride
cis- 1, 3-Dichloropropene
trans- 1 , 3-Dichloropropene
1, 1,2-Trichloroethane
Tetrachloroethylene
Ethylene dibromide
Methyl bromide
Chlorobenzene
Vinyl bromide
Methyl chloroform
Mean*
(Percent)
95.8
123.0
127.2
101.6
42-10
31,90
29.80
92,64
36.40
30,30
33.30
31.90
28.20
30.60
31.60
31.10
106.4
111.6
97.0
97.4
94.2
110.8
103.4
Standard
Deviation
(Percent)
10.73
5.61
6.91
288
862
631
748
20.30
548
546
707
768
454
558
556
568
14.58
750
1442
953
953
1030
1270
Percent
Coefficient
of
Variation
11.20
4.56
5.43
2.84
24.32
17.81
21.10
46.30
15.46
15.40
19.96
21.67
12.81
15.75
15.69
16.04
13.71
6.72
14.86
9.78
9.78
9.30
12.28
'Average of 5 values.
Table 3. VOST Method Detection Limits
Compound
Ethyl chloride
Ethylene dichloride
Methyl iodide
Allyl chloride
Methylene chloride
Ethylidene dichloride
Chloroprene
Methyl chloride
Chloroform
Carbon tetrachloride
1 ,2-Dichloroethane
Vinyl chloride
Trichloroethylene
Propylene dichloride
cis- 1 , 3-Dichloropropene
trans- 1 , 3-Dichloropropene
1, 1,2-Trichloroethane
Tetrachloroethylene
Ethylene dibromide
Methyl bromide
Chlorobenzene
Vinyl bromide
Methyl chloroform
Mean*
(ng)
34.20
24.90
30.50
29.80
42.10
31.90
29.80
92.64
36.40
30,30
33.30
31.90
28,20
30.60
31.60
31.10
32.80
29.30
29.80
43.70
29.80
30.60
43.80
Standard
Deviation
(ng)
8.72
6.12
7.11
5.14
8.62
6.31
7.48
2030
5.48
546
7.07
768
4.54
5.58
556
568
5.92
542
575
10.19
4.64
640
786
Method
Detection
Limit
(ng)
24.59
17.26
20.05
14.49
24.32
17.81
21.10
46.30
15.46
15.40
19.96
21.67
12.81
15.75
15.69
16.04
16.71
15.28
16.22
28.74
13.08
18.05
22.16
'Average of 10.
showed recoveries that were acceptable
for further study. Recovery from sorbent is
essential for analytical determination us-
ing Method 5041.
Method Detection Limits for the candi-
date VOST analytes are reported in
Table 3. The highest value is 46 ng/sample
with most between 10 and 20 ng/sample.
A full VOST sample of 20 L of stack emis-
sions would then have a range of 0.5 - 2.3
ng/L of gas sampled. This range of Method
Detection Limits is acceptable.
The laboratory experiment with the qua-
druple train set up for evaluation of VOST
gaseous dynamic spiking was performed.
The Latin Square experimental design re-
quired four spiking levels, provided from a
pressurized cylinder by means of mass
flow controllers, with four runs on four
trains. The results of the Latin Square
experiment showed that only one effec-
tive level of spiking had been achieved in
spite of apparent changes in flow rate.
The mass flow controllers functioned er-
ratically, most of the time in a fully open
position. A reconfiguration of the quadruple
VOST train system to use heated needle
valves to regulate the gas flow, bubble
flow meters to measure actual flow before
and after sampling, and heated Teflon®
lines throughout the dynamic spiking sys-
tem to ensure that compound condensa-
tion does not occur was developed. Pre-
liminary results obtained in the laboratory
indicate that the modified spiking system
provides a constant flow. However, the
reconfigured spiking system has not yet
been evaluated completely. Preliminary
results demonstrate that the delivery of
compounds is consistent at a given level.
SemiVOST. Chbroacetic acid was the only
SemiVOST candidate target compound that
could not be chromatographed successfully
using the standard conditions for SemiVOST.
Erratic results during calibration suggested
possible problems with stability in solution for
bis(chbromethyl) ether, epichbrohydrin, and
3,3'-dichbrobenzidine.
The recoveries from the spiked XAD-2®
resin sampling cartridges are shown in
Table 4. The recoveries ranged from 38%
for pentachloronitrobenzene to 275% for
3,3'-dichlorobenzidine. The method states
that a range of 50 to 150% is acceptable.
Using the criteria from the SemiVOST
method five compounds would not achieve
an acceptable recovery from the XAD-2® sor-
bent. The compounds with bw recoveries were
hexachlorobutadiene, hexachlorobenzene,
pentachbronitrobenzene, chbrobenzilate and
3,3'-dichbrobenzidine. Even though these five
recoveries were out of range the compounds
were retained as candidate targets for the
SemiVOST method for the rest of the study.
-------
Table 4. Recoveries of Compounds from SemiVOST Sorbents (XAD-2® Resin)
Compound
bis(Chloromethyl) ether
Epichlorohydrin
cis- 1, 3-Dichloropropene
frans- 1 , 3-Dichloropropene
1, 1,2-Trichloroethane
Ethylene dibromide
Tetrachloroethylene
Chlorobenzene
Bromoform
1, 1,2,2-Tetrachloroethane
Dichloroethyl ether
1 ,4-Dichlorobenzene
Benzyl chloride
Hexachloroethane
1 ,2-Dibromo-3-chloropropane
1,2,4- Trichlorobenzene
Hexachlorobutadiene
Benzotrichloride
Chloroacetophenone
Hexachlorocyclopentadiene
2,4, 6- Trichlorophenol
2, 4, 5- Trichlorophenol
Hexachlorobenzene
Pentachlorophenol
Pentachloronitrobenzene
Chlorobenzilate
3,3'-Dichlorobenzidine
Mean*
(Percent)
59.3
75.2
71.0
79.4
78.8
89.2
61.1
96.6
80.8
102.0
104.4
95.0
103.2
87.4
92.0
90.6
47.8
76.8
141.6
53.0
93.8
108.2
4.8
69.8
38.0
47.6
275.0
Standard
Deviation
(Percent)
8.10
11.10
10.46
12.01
9.98
12.56
7.66
12.10
11.30
14.05
11.80
12.43
13.08
12.46
13.27
13.35
6.42
11.80
21.43
9.51
15.16
15.24
5.63
10.55
4.58
6.88
5.83
* Average of 5 values.
Table 5. SemiVOST Method Detection Limits
Detection Limit
Total
Compound
\ig/mL
bis(Chloromethyl) ether
Epichlorohydrin
cis- 1, 3-Dichloropropene
trans- 1 , 3-Dichloropropene
1, 1,2-Trichloroethane
Ethylene dibromide
Tetrachloroethylene
Chlorobenzene
Bromoform
1, 1,2,2-Tetrachloroethane
Dichloroethyl ether
1 ,4-Dichlorobenzene
Benzyl chloride
Hexachloroethane
1 ,2-Dibromo-3-chloropropane
1,2,4- Trichlorobenzene
Hexachlorobutadiene
Benzotrichloride
Chloroacetophenone
Hexachlorocyclopentadiene
2,4, 6- Trichlorophenol
2, 4, 5- Trichlorophenol
Hexachlorobenzene
Pentachlorophenol
Pentachloronitrobenzene
Chlorobenzilate
3,3'-Dichlorobenzidine
11.4
9.8
5.8
6.5
9.0
10.7
13.4
9.5
10.6
8.2
11.0
12.9
12.0
10.9
12.6
13.1
15.7
12.7
13.9
14.5
11.6
16.5
13.4
30.7
13.0
15.6
19.3
57.0
49.0
29.0
32.5
45.0
53.5
67.0
47.5
53.0
41.0
55.0
64.5
60.0
54.5
63.0
65.5
78.5
63.5
69.5
72.5
58.0
58.0
67.0
153.5
65.0
78.0
96.5
Percent
Coefficient
of
Variation
13.67
14.76
14.74
15.13
12.67
14.08
12.20
12.52
13.99
13.78
11.30
13.08
12.68
14.26
14.42
14.74
13.43
15.36
15.14
17.95
16.16
14.08
12.29
15.11
12.06
14.45
20.31
Method detection limits (Table 5) for
candidate SemiVost compounds ranged
from 29.0 to 153.5 jig/sample based on a
final concentration volume of 5 ml. As-
suming a sampling rate of 0.5 dm and a
two-hour sampling period, the limits would
range from 0.5 to 2.5 |ig/cf of emission
gas sampled. The majority of the com-
pounds tested would be near 1 (ig/cf.
A full SemiVOST Latin Square qua-
druple train spiking experiment was per-
formed. The average recoveries ranged
from 8.9% for pentachlorophenol to 513%
for hexachlorocyclopentadiene. Eighteen
of the twenty-seven targeted compounds
had an average recovery between 50 and
150 percent. Difficulties with recovery of
pentachlorophenol are illustrative of the
erratic behavior exhibited by this com-
pound under test conditions; pentachloro-
phenol is outstandingly sensitive to chco-
rn at ographic conditions such as cleanli-
ness of the injector port and condition of
the chromatographic column. Problems with
the recovery of hexachlorocyclopentadiene
are attributed to difficulties in calibration,
possibly due to stability problems with this
compound in the calibration solution.
Conclusions
Gas chromatographic retention times, mass
spectra and primary secondary quantrtation
ions were determined for most of the haloge-
nated compounds listed under Title III, CAAA.
Of the targeted 45 compounds, only four
could not be chromatographed successfully:
chloroacetic acid, bis(chloromethyl) ether,
chloromethyl methyl ether, and epichlorohy-
drin. The full set of Latin Square SemiVOST
quad train experiments and tests performed
to evaluate run-to-run reproducibility indicate
that 70% of the candidate target com-
pounds should be evaluated on actual sta-
tionary sources. Although the complete
Latin Square VOST experiment was not
successful, consistent results obtained with
the modified dynamic spiking system indi-
cate that the VOST may be useful on
most of the targeted VOST compounds.
Since standard solutions of both volatile
and semivolatile compounds containing all
of the compounds of interest that could be
chromatographed successfully are avail-
able, no compounds will be removed from
the set for field testing of the trains.
-------
Table 6. Recoveries of Compounds from SemiVost Latin Square Experimental
Runs
Compound
bis(Chloromethyl) ether
Epichlorohydrin
cis- 1, 3-Dichloropropene
trans- 1 , 3-Dichloropropene
1, 1,2-Trichloroethane
Ethylene dibromide
Tetrachloroethylene
Chlorobenzene
Bromoform
1, 1,2,2-Tetrachloroethane
Dichloroethyl ether
1 , 4-Dichlorobenzene
Benzyl chloride
Hexachloroethane
1 , 2-Dibromo-3-chloropropane
1,2,4- Trichlorobenzene
Hexachlorobutadiene
Benzotrichloride
Chloroacetophenone
Hexachlorocyclopentadiene
2, 4, 6- Trichlorophenol
2,4,5-Trichlorophenol
Hexachlorobenzene
Pen?achlorophenol
Pentachloronitrobenzene
Chlorobenzilate
3, 3 '-Dichlorobenzidine
Mean*
(Percent)
18.28
75.20
21.90
20.34
53.13
66.31
49.68
75.98
99.27
81.05
75.73
68. 15
78.72
85.43
66.24
58.20
58.34
67.02
79.64
513.M
4.61
52.69
32.85
8.93
38.24
43.63
86.42
Standard
Deviation
(Percent)
9.22
24.11
6.55
5.80
14.82
14.56
14.48
13.46
22.25
12.77
11.99
10.90
20.43
35.1
6.91
10.94
10.69
16.58
18.03
254.26
16.30
39.78
18.35
10.50
20.66
35.49
165.82
'Four quadruple runs were performed (total of 16 samples); two sets of results
were rejected as outliers, leaving 14 samples.
a-U-S. GOVERNMENT PRINTING OFFICE: t993 - 750471/801»
-------
-------
Joan T. Bursey, Raymond G. Merrill, Jr., Robert A. McAllister, and James F.
McGaughey are with Radian Corp., Research Triangle Park, NC 27709.
Merrill D. Jackson is the EPA Project Officer (see below).
The complete report, entitled "Laboratory Validation of VOSTand SemiVOSTfor
Halogenated Hydrocarbons from the Clean Air Act Amendments List" consists
of two volumes:
Volume 1: (Order No. PB93-227163/AS; Cost: $36.50, subject to change) and
Volume 2 (Order No. PB93-227 171/AS; Cost: 36.50, subject to change).
The above reports 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:
Atmospheric Research and Exposure Assessment 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-93/123
------- |