EPA/80Q/A-93/2Q3
CHEMOSPHERE, Vol. 25, Nos. 7-10, pp 1317-1322, Oct.-Nov., 1992
EVALUATION OF A SAMPLING AND ANALYSIS METHOD FOR DETERMINATION OF
POLYHALOGENATED DIBENZO-/>-DlOXINS AND DIBENZOFURANS IN AMBIENT AIR
Robert L. Harless and Robert G. Lewis
U. S. Environmental Protection Agency
Atmospheric Research and Exposure Assessment Laboratory
Research Triangle Park, North Carolina 27711
Danny D. McDaniel, James F. Gibson and Aubry E. Dupuy, Jr.
U. S. Environmental Protection Agency
Environmental Chemistry Laboratory
Stennis Space Center, Mississippi 39529
Abstract
General Metals Works PS-1 PUF air samplers and an analytical method based on high resolution gas
chromatography - high resolution mass spectrometry (HRGC-HRMS) were evaluated for determination of
polyehlorinated dibenzo-p-dioxins and dibenzofurans (PCDDs/PCDFs), polybrorninated dibenzo-p-dioxins and
dibenzofurans (PBDDs/PBDFs) and bromo/chloro dibenzo-p-dioxins and dibenzofuraas (BCDDs/BCDFs)
in ambient air. Dilute solutions of these compounds and °Cu-l,2,3,4-TCDD were used to spike the filters
of PS-1 air samplers which were then operated 24 hr to sample 350-400 m3 ambient air. After sampling, each
quartz-fiber filter and polyurethane foam (PUF) were spiked with "Cy-Iabeled PCDD, PCDF, PBDD, and
PBDF internal standards before separate Soxhlet extractions with benzene. The extracts were subjected to
an acid/base clean-up procedure followed by clean-up on mierocolumns of silica gel, alumina, and carbon and
then analyzed by HRGC-HRMS. Results derived from this study indicated the PS-1 ambient air samplers
and the analytical procedures were very efficient and that pg/m3 and sub-pg/m' levels of total
PCDDs/PCDFs, PBDDs/PBDFs, BCDDs/BCDFs, and 23,7,8-substituted congeners could be accurately
measured. Background levels of these compounds in the ambient air were also determined. Total PCDDs,
PCDFs, TBDFs, and PeBDFs were detected in a low concentration range of 03 to 3.0 pg/m'.
Introduction
Many types of sampling devices and analytical methods have been developed and evaluated for
determination of semivolau'le organic chemicals, including polychlorinated dibenzo-/?-dioxins (PCDDs) and
dibenzofurans (PCDFs) in ambient air.1AJ Because PCDDs and PCDFs can be formed by thermal reactions,
there has been an increasing interest in ambient monitoring, especially in the vicinities of combustion and
-------
incineration processes such as municipal waste incinerators and resource recovery facilities.4-3-6
, Polybrominated dibenzo-p-dioxins (PBDDs) and dibenzofurans (PBDFs) can also be created thermally'' and
mixed bromo/chloro dibenzo-p-dioxins and dibenzofuraas BCDDs/BCDFs have been detected to ash from
. combustion/incineration processes.*
This research was designed to fully evaluate the air sampler and analytical method for determination
of the polyhalogenated dibeozo-p-diorins and dibenzofurans (PHDDs/PHDFs) in ambient air. The data from
this study and previous studies is being used to prepare well-documented sampling procedures and analytical
procedures for tbe EPA (AREAL) Compendium of Methods for the Determination of Toxic Organic Compounds
in Ambient Air.
Experimental
Sampling. Three General Metals Works PS-1 air samplers were used in this study. The samplers were
equipped with precleaned quartz fiber filters and polyurethane foam (PUF) sorbent traps to collect
paniculate-associated and vapor-phase PHDDs and PHDFs. A calibrated orifice was used to calibrate each
sampler prior to initiation of the study and at intervals during the study. Air flow rates of 0.24 to 0.28 mj/min
were used in 24-hr sampling sessions to provide sample volumes between 346 and 400 m3.
The three samplers were operated for 24-hr periods to sample 350-400 m3 ambient air. Each of the
three PS-1 filters containing paniculate matter loadings was then spiked with 50 pL of a solution containing
16 pg/^L of 1,2,3,4-TCDD in benzene. Two of the PS-1 filters were also spiked with 20 to 80 ML of similar
PHDD/ PHDF spike solution. Spike solutions containing all tetra- through octa-congeners and solutions
containing only the 23,?,§-substitiited congeners were used. The three samplers were then operated another
24 hr. Upon completion of this sampling period, each filter was removed from the sampler, folded, wrapped
in aluminum foil, and placed on top of tbe respective PUF sorbent The glass cartridges containing the filter
and PUF were wrapped and sealed in aluminum foil, packed securely and shipped to the laboratory for
sample preparation. Experiments were also performed in which the spikes were applied to clean filters
immediately prior to initiation of 24-hr sampling period. Also, two PS-1 samplers were spiked with the UCU-
1A3.4-TCDD and operated for seven days and sampled 2660 m3 ambient air in the final experiment.
Sample Preparation. Sample preparation was performed on a set of 12 samples consisting of nine test
samples, method blank, field blank and laboratory method spike. The filter and PUF from the PS-1 sampler
used to collect background ambient air data were combined for extraction. Filter and PUF from the other
two PS-1 samplers were isolated for separate extraction. Aliquots of spike solutions containing 1.0 ng each
of "Cjj-labeled tetra-, penta-, hexa-PCDD/PCDF, 2.0 ng I*CJ2-labeled HpCDD and OCDD and 0.86 ng each
of "C^-labeled PBDD/PBDF internal standards were spiked into each sample immediately before Soxhlet
extraction for 16 hr with benzene. Qean-up of extracts was accomplished using an acid/base procedure
-------
followed by microcolumns of silica gel, alumina and carbon. An aliquot of a standard solution containing
•03 ng S7Ca4-2f3»7,8-TCDD was spiked into each extract prior to final concentration to 60 nL for analysis. The
glass tubes containing the extracts were fire-sealed and shipped to the AREAL laboratory for analysis.
Instrumentation. A Finnigan MAT 90 HRGC/HRMS system operating in the electron impact
ionization and multiple ion detection mode at 7,000 to 10,000 mp-« resolution was used for the analysis of
samples. Chromatographie resolution of components was accomplished using a 30 m x 025 mm ED SE-54
fused silica capillary column which was inserted directly into the HRMS ion source. Operating parameters
were: initial column temp, 170 * C; hold for 7 min; program to 305 * C at 8 * C; hold for 35 min; injection temp,
310' C; transfer line, 295 * C.
Quantification. The peak areas of exact masses of the molecular ion clusters of uC12-labeled and
unlabeled PCDDs, PCDFs, PBDDs, PCDFs, 37Cl4-2378"TCDD and respective response factors were used for
quantification purposes. The s7Q4-23,7,8-TCDD, spiked to the extract prior to final concentration, was used
to determine the method efficiency achieved for the "Cjj-labeled PHDD/PHDF internal standards which
were spiked to sample prior to extraction. The "C^-labeled PHDD and PHDF internal standards and
response factors were used for quantification of unlabeled PHDDs and PHDFs and for determination of the
minimum limits of detection with but two exceptions; "Cu-labeled HpCDD was used for HpCDFs and "C12-
OCDD was used for OCDF. The "C^-l^A^-TCDD spiked to the PS-1 filter prior to ambient air sampling
was used to determine the PS-1 air sampler collection and retention efficiency for the 24-hr sampling period.
Quality Assurance/Quality Control, Analytical criteria used for confirmation of PHDDs and PHDFs
were: chlorine isotope ratios or bromine isotope ratios of molecular ions (within ± 15% of theoretical values);
correct responses for exact masses of uCjj-labeIed and unlabeled PHDDs/PHDFs in the known retention time
window of the specific homologue to be determined; analysis that confirmed the absence of chlorinated
diphenyletbers; responses of unlabeled PHDD/PHDF masses must be greater than 23 x area of noise level,
The data from a set of samples was evaluated using the analytical criteria and following QA/QC
requirements: method efficiency achieved for "Cjj-labeled tetra-, penta-, and hexa-HDD and HDF internal
standards, 50 to 120%; HpCDD and OCDD, 40 to 120%; accuracy and precision achieved for laboratory
method spike(s) at 0.5 to 2.0 pg/ms, ±50%; method blank and field blank free of significant PHDD/PHDF
contamination at the MLDs required for generation of meaningful data for ambient air, 0.02 to 0.15 pg/m3
for tetra-, penta-, and hexa-HDDs/HDFs. Trace levels, 0.05 to 025 pg/ms, of HpCDDs and OCDD are
always detected in the method blanks and field blanks, method background level.
-------
Results and Discussion
The results for distribution of PCDDs and PCDFs in the filter and PUF after sampling 400 m3 ambient
air are shown in Tables 1,2, and 3; those for BCDD/BCDF are shown in Table 4; and those for the PBDDs
and PBDFs are shown in Table 5. Most of the HpCDDs/HpCDFs and OCDD/OCDF are retained by the
filter, and most of the tetra-, penta-, and bexa«CDDs/CDFs and BCDD/BCDF volatilize and are collected
by the PUF. As shown in Table I, there were no significant differences delectable in the volatility of specific
23,7,8-substituted HxCDD, PeCDF, and HpCDF isomers. The volatility of the PBDD and PBDF congeners,
however, is quite different from the PCDDs/PCDFs. As shown in Table 5, most of the penta- and bexa- and
about 60% of the tetra-BDD/BDF are retained by the filter, and only small amounts are volatilized and
collected by the PUF. However, when spiked to clean filters, the PBDDs/PBDFs, especially the PeBDF and
the HxBDF, decompose to form other tetra- and penta-isomers during the 24-hr sampling period.
The concentrations of PCDDs/PCDFs detected in the ambient air during this study are shown in
Table 6. These low and variable concentrations are similar to those that have been detected in other EPA
studies.* As also shown in Table 6, PBDDs/PBDFs with the exception of TBDFs were not detectable in most
cases. A minimum of eight TBDF isomers were detectable. Average values are shown for the duplicate
samples collected on August 8,1991, which were derived from the two PS-1 samplers that bad been operated
seven days and sampled 2660 mj ambient air to provide large duplicate samples for confirmation purposes.
Two PeBDF isomers and one HxBDF isomer were also detected in both samples. No problems were
encountered in the PS-1 sampler and analytical procedures due to the long sampling period.
The PS-1 air sampler proved to be very dependable and reliable in all of the studies. The average
efficiencies that have been achieved, based on retention of 0.8 ng uCn-li3,4-TCDD spiked to the filter prior
to sampling 325-400 rn3 ambient air are: this study, 91%; Green Bay Study, 92%; Vermont Study, 85%.
The average efficiencies achieved with the analytical method for the "Cylabeled standards
were: 23,7,8-TCDD, 89%; 2^,7,8-TCDF, 90%; 1,23,7,8-PeCDD, 88%; 1,23,7,8-PeCDF, 88%; 1,2,3,6,7,8-
HxCDD, 87%; 1 A3,4,7,8-HxCDF, 88%; 1,23,4,6,7,8-HpCDD, 83%; OCDD, 73%; 23,7,8-TBDD, 96%; 2,3,7,8-
TBDF, 103%; 1,23,7,8-PeBDF, 103%. Efficiencies for the PCDDs/PCDFs are based on analyses of hundreds
of air samples, including those from this study. Efficiencies for the PBDD/PBDFs are from this study only.
The current data base for these compounds Is very small. Accuracy achieved for spike levels of the
PHDDs/PHDFs is adequate. Minimum limits of detection achieved for PHDDs/PHDFs were in the range
of 0.02 to 02 pg/ms. The data derived from this study demonstrate and confirm the capabilities of the PS-1
sampler and the analytical method for determination of pg/ms and sub pg/m3 levels of PHDDs/PHDFs in
ambient air.
-------
Table 1. Distribution of Spiked
2,3,7,8-Substituted PCDD/PCDF Congeners
after Sampling 400 m' Ambient Air
2,3.7.8-TCDD
1,2,3,7,8-PeCDD
1.2.3.4. 7.8-HxCDD
1.2,3,6,7.8-HxCDD
1.2,3.4,6,7,6-HpCDO
OCDD
2.3.7,8-TCDF
1_2,3,7,8-PeCDF
2,3,4,7.8-PeCDF
1.2,3,4.7,8-HxCDF
1,2.3,6.7,8-HxCDF
12,3,7,8.9-HxCDF
2.3,4,6,7.8-4-lxCDF
12.3,4,6.7.8-HpCOF
1,2.3.4,7,6.9-HpCDF
OCDF
«-n— ,„% Pereent Recovered
Filter
200 0
200 7
200 16
200 15
200 52
800 78
200 0
200 e
200 6
200 8
200 8
200 7
0 0
200 60
200 55
800 67
PUF Total
100 100
94 101
64 100
84 99
22 74
9 87
98 98
7S 61
92 98
67 75
67 75
67 74
0 0
34 94
34 89
7 74
Table
2. Distribution of Low
-Level and
High-Leve! Spikes of PCDDs/PCDFs after
Sampling 400 m* Ambient Air
Low
TCDDs
PeCDOs
HxCDDs
HpCDDs
OCDD
TCDFs
PeCDFs
HxCDFs
HpCDFs
OCDF
High
TCDDs
PeCDDs
HxCDDs
HpCDDs
OCDD
TCDFs
PeCDFs
HxCDFs
HpCDFs
OCDF
Tables
Percent Recovered
Splice (ng)
0.30
0.66
0.74
0.68
0.29
0.64
0.79
0.76
0.34
0.07
1.6
4.4
4.4
5.1
2.8
3.9
5.5
5.3
2.7
0.6
Filter
20
16
30
64
82
19
16
20
50
99
7
8
14
61
86
13
15
13
36
90
PUF
74
6?
82
23
14
73
64
61
33
10
102
98
90
33
5
97
90
84
49
8
Total
94
78
112
87
96
92
80
81
83
109
109
106
104
84
91
110
105
97
85
98
, Distribution of Duplicate Low-Level
Spikes of PCDDs/PCDFs after Sampling
400 m9 Ambient Air (Spiked to Clean Filters)
TCDDs
PeCDDs
HxCDDs
HpCDDs
OCDD
TCOFi
PeCDFs
MxCDFs
HpCDFs
OCOF
-..
115
263
322
260
148
224
340
266
144
36
Percent Recovered
Finer*
F-1 F-2
0 0
6 9
6 0
23 21
56 53
6 6
S 3
7 5
15 13
51 44
PUF»
PUF-1
77
86
88
69
41
69
75
79
58
42
PUF-2
97
68
97
71
32
81
88
83
58
33
Total
1 2
77 97
92 97
96 97
92 92
97 85
75 87
60 91
86 88
73 71
93 77
Table 4. Distribution of Duplicate Low-Level
Spikes of BCDD/BCDF after Sampling 400 m9
Ambient Air
2-Br-3,7,8-Trichlero-
(*S«nzop-Dtoxin
3-Br-7,S.9-Trtchlort>-
Abenzoluran
Bpik*
)
TCDDs
PeCDDs
HxCDDs
HpCDDs
OCDD
TCDFs
PeCDFt
HxCDFi
HpCDFs
OCDF
Sample
Collected
3-28-90
NO (0.02)
0.03
0.06
0.09
0.15
0.10
0.07
0.07
0.09
0.18
Sample
Collected
10-1MO
0.14
0.19
0.31
O.S1
0.59
0.51
O.SO
0.69
0.70
0.49
Sample
Collected
6-19-91
O.OB
0.06
0.10
0.08
0.10
023
0.18
0.10
0.06
0.04
Sample
Collected
8-8-91
0.03
0.05
0.05
0.09
0.12
0.14
0.11
0,06
0.04
0.03
Analytical Data for PBDDs/PBDFs Detected
in Ambient Air at RTP, NC (pg/m3)
TBDDs
PeBDDs
HsBDDs
TBDFs
PeSDFs
HxBDFs
Sample
Collected
3-28-90
ND (0.04)
ND (0.02)
ND(0.11)
0.13
ND (0.03)
ND(0.11)
Sample
Collected
10-15-90
ND (0.02)
ND (0.04)
ND (0.12)
OJO
ND (0.08)
ND(OJ1!
Sample
Collected
t-1741
ND (0.02)
ND (0.10)
ND (0.08)
0.16
ND (0.10)
ND (0.08)
Sample
Collected
8-16-91
ND (0.02)
ND (0.03)
ND (0.05)
OJO
ND (0.03)
ND (0.19)
Sample
Collected
•4-91
ND(0.01)
NDI0.01)
ND (0.04)
0.19
0.22
0.30
-------
Acknowledgement
The efforts of Malcolm C, Willtins in calibrating the PS-1 air samplers (hiring this study are sincerely
appreciated.
Disclaimer
This paper has been reviewed in accordance with the U. S. Environmental Protection 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.
References
1 R, G. Lewis and M. D. Jackson, "Modification and evaluation of a high-volume air sampler for
pesticide and semivolatile industrial organic chemicals,* AnaL Chen. 54: 592,1982.
2. B. D. Eltzer and R. A. Kites, Tolychlorinated dibenzo-p-dioxins and dibenzofurans in the ambient
atmosphere of Bloomington, Indiana," Environ. Set Technol 23 (11), 1989.
3. A. Szakolcai and W. Chan, "Intercomparison study of ambient air dioxin/furan sampling and analytical
methods," pp. 170-175 in Proceedings of the 1988 EPA/APCA International Symposium on Measurement
of Toxic and Related Air Pollutants, VIP-10, Air Pollution Control Association, Pittsburgh, 1988.
4. T. Tieman, D. Wagel, G. Vanness, J, Garrett, J. Soleh, and L. Harden, "PCDD/PCDF in the ambient
air of a metropolitan area in the U.S.,* Chemosphere 19: 541-546, 1989,
5. G. Hunt, "Measurement of PCDDs/PCDFs in ambient air,"/. AirPollut. Control Assoc. 39 (3), 1989.
6. R. L. Harless, R. G. Lewis, D. D. McDaniel, and A. E. Dupuy, Jr., "Determination of polychlorinated
dibenzo-j?-dioxins and dibenzofurans in stack gas emissions and ambient air," in Proceedings of the 1988
EPA/APCA International Symposium on Measurement of Toxic and Related Air Pollutants, VIP-10, Air
Pollution Control Association, Pittsburgh, 1988.
7. H. R. Buser, "Polybrominated dibenzofurans and dibenzo-j>-dioxins: Thermal reaction products of
polybrominated diphenyl ether flame retardants," Environ. ScL TechnoL 20 (4), 1988.
8. R. L Harless, R. G. Lewis, D. D. McDaniel, and A. E. Dupuy, Jr., "Identification of bromo/cWoro
dibenzo-p-dioxins and dibenzofurans in ash samples,* Chemosphere 18 (1-6): 201-208, 1989.
9. R. L Harless, R. G. Lewis, D. D. McDaniel, and A. E. Dupuy, Jr., "Sampling and analysis for
polychlorinated dibenzo-p-dioxins and dibenzofurans in ambient air," Chemosphere, in press, 1991.
-------
TECHNICAL REPORT DATA
ff lease read Instructions on the reverse before complet
1. REPORT NO.
EPA/6QO/A-93/203
2.
3,
4. TITLE AND SUBTITLE
Evaluation of a Sampling and Analysis Method for
Determination of Polychlorinated Dibenzo-p-Dioxins and
Dibenzofurans in Ambient Air
5. REPORT DATE
6 PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
R. Harless, R. Lewis, D. McDaniel, J. Gibson, and
A, Dupuy
8. PERFORMING ORGANIZATION REPORT NO,
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Atmospheric Research and Exposure Assessment Laboratory
Methods Research and Development Division
Ambient Methods Research Division (MD-44)
Research Triangle Park, NC 27711
10, PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
Atmospheric Research and Exposure Assessment Laboratory
Methods Research and Development Division
Ambient Methods Resarch Division (MD-44)
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
General Metals Works PS-1 PUF air samplers and an analytical method based on high
resolution gas chromatography - high resolution mass spectrometry (HRGC-HRMS) were
evaluated for determination of polychlorinated dibenzo-p-dioxins and dibenzofurans
(PCDDs/PCDFs), polybrominated dibenzo-p-dioxins and dibenzofurans (PBDDs/PBDFs) and
bromo/chloro dibenzo-p-dioxins and dibenzofurans (BCDDs/BCDFs) in ambient air. Dilute
solutions of these compounds and C12-l»2,3,4-TCDD were used to spike the filters
of PS-1 air samplers which were then operated 24 hrs to sample 350-400 m ambient air.
After sampling, each quartz-fiber filter and polyurethane foam (PUF) were spiked with
r ••-'--•'-^ pcDD, PCDF, PBDD, and PBDF internal standards before separate Soxhlet
extractions with benzene. The extracts were subjected to an acid/base clean-up
procedure followed by clean-up on microcolumns of silica gel, alumina, and carbon
and then analyzed by HRGC-HRMS. Results derived from this study indicated the PS-1
ambient air samplers and the analytical procedures were very efficient and that
pg/m3 and sub-pg/m3 levels of total PCDDs/PCDFs, PBDDs/PBDFs, BCDDs/BCDFs, and
2,3,7,8-substituted congeners could be accurately measured. Background levels of
these compounds in the ambient air were also determined. Total PCDDs, PCDFs, TBDFs,
and PeBDFs were detected in a low concentration range of 0.3 to 3.0 pg/m .
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
8. DISTRIBUTION STATEMENT
19. SECURITY CLASS (This Report!
21. NO. OF PAGES
6
20. SECURITY CLASS (This page)
22. PRICE
EPA Form 2220-1 (R«». 4-77) PREVIOUS EDITION is OBSOLETE
------- |