vvEPA
United States
Environmental Protection
Agency
Environmental Research
Laboratory
Duluth, MN 55804
EPA/600/3-90/022
March 1990
Research and Development
Analytical
Procedures and Quality
Assurance Plan for the
Determination of
PCDD/PCDF in Fish
-------
EPA/600/3-90/022
March 1990
U.S. Environmental Protection Agency
National Oioxin Study - Phase II
Analytical Procedures and Quality Assurance Plan
for the Determination of PCOD/PCOF in Fish
Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Duluth, MN 55804
-------
NOTICE
The information in this document has been funded wholly or in part by the U.S.
Environmental Protection Agency. It has been reviewed technically and
administratively. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
12/89 QA/OC PCOD/PCDF it
-------
ACKNOWLEDGEMENTS
Technical contributions to this research were made by:
U.S. Envi ronmentaI Protection Agency
Brian C . Butterworth
Douglas W. Kuehl
ASc I Corporation
Phillip J. Harqui s
Marie L . Larsen
Larry G. Holland
Christine E. Soderberg
Jennifer A. Johnson
Kevin L . Hogfeldt
ih t State Un i ve r s i t v
Dr. Thomas Tiernan
Dr. Michael Taylor
University of Wisconsin-Superior
Elizabeth A. Lundmark
Daniel M. Fremgen
Sandra Naumann
Murray Hackett
Kent Johnson
Harvey D. Corbin, Jr.
Dr. Ray L. Hanson
12/89 QA/QC PCOD/PCDF
i i i
-------
FOREWORD
Directed by Congressional mandate, the U.S. Environmental Protection Agency
during 1983 initiated the National Dioxin Study, a survey of environmental
contamination by 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in the United
States. Results of this study are published in the National Dioxin Study:
Tiers 3,5,6, and 7, EPA 400/4-82-003. This laboratory, the Environmental
Research Laboratory- Duluth, was responsible for one part of the Study, the
analysis of fish samples. The most significant findings of these analyses was
the observation that fish contamination was more widespread than previously
thought, and that a primary source of TCDD was discharge from pulp and paper
production using chlorine.
A second more detailed characterization of anthropogenic organic chemical
contaminants in fish was conducted in subsequent analyses during what is now
called Phase II of the National Dioxin Study. This document describes the
analytical methods used for the determination of the level of contamination of
fifteen biosignificant polychlorinated dibenzo-p-dioxins and dibenzofurans in
fish. A companion document (EPA /600/3-90/023) describes the analytical methods
used for the determination of levels of contamination of polychlorinated
biphenyls, pesticides, and industrial compounds in those same fish.
12/89 QA/QC PCDD/PCDF iv
-------
TABLE OF CONTENTS
DISCLAIMER ii
ACKNOWLEDGEMENTS Hi
FOREWORD iv
I. Introduction 1
II. Sample Preparation
A . Grinding 3
B . Extraction 3
C. Percent Lipid Determination 3
D. Anthropogenic Chemical Isolation 5
E. Florisil Chromatography 5
F. PCDD/PCDF Isolation 5
III. Reagents and Standards 6
A. Reagents 6
B. Standards 7
IV. Instrumental Parameters 12
V. Quality Assurance/Quality Control 13
A. General Procedures of Operation 13
B. Instrumental Quality Control 20
C. Evaluation of Data 21
1. Accuracy 21
2. Precision 23
3 . Signal Quality Assurance Requirements 23
4. Polar Gas Chromatographic Confirmation Analysis 23
D. Quality Assurance Problems and Corrective Actions 24
12/89 QA/QC PCDD/PCDF
-------
VI. Quantification Procedures 25
A. Initial and Daily Calibration of the HRMS 25
B. Signal Quality 27
C. Quantification of PCOD/PCDF 29
0. Method Efficiency 30
E . Integration of Automated Data Processing and Quality
Assurance 31
TABLES
Table 1 -- Biosignificant PCDDs/PCDFs 1
Table 2 -- Minimum Level of Detection Limit 2
Table 3 -- Internal Standard Solutions 4
Table 4 -- Calibration Standards 9
Table 5 -- Relative Retention Times 4-8 PCDD Isomers 10
Table 6 -- Relative Retention Times 4-8 PCDF Isomers 11
Table 7 -- HRGC/HRMS Operating Parameters 12
Table 8 -- Native PCDD/PCDF Spiking Solution 14
Table 9 •- Codes for the SCC Number and Matrix Type 19
Table 10-- GC Column Performance Quality Control 20
Table 11-- GC Etution Window Defining Solutions for
DB-5 Column 21
Table 12-- Quality Assurance Parameters 22
Figures
Figure 1 - - Database Format for Sample Information 17
Figure 2 -- 2,3,7,8-TCDD Weighted Calibration Curve 26
Figure 3 -- Data Reduction for PCDD/PCDF National Dioxin Study.32
12/89 QA/QC PCDD/PCDF vi
-------
I . Introduct ion
This document, "Analytical Procedures and Quality Assurance Plan for the
Determination of PCDD/PCDF in Fish" has been drafted in response to the need
for the Environmental Research Laboratory of Duluth (ERL-D) to perform analysis
for tetrachloro- to octachloro- congeners/isoners of po I ychIorinated dibenzo-
p-dioxins and dibenzofurans (PCDD/PCDF), Table 1.
2378-TCD F
2367- TCD F
3467- TCD F
2378-TCDD
12378-PeCDF
23478- PeCD F
23467-PeCDF
12378-PeCDD
123467-HxCDF
1 23478-HxCD F
123678-HxCDF
234678-HxCD F
1 23789- HxCD F
1 23478-HxCDD
1 23678-HxCDD
1 23789-HxCDD
1 234678- HpCDF
1234789-HpCDF
1234678-HpCDD
CASRN
51207-31-9
1746-01-6
57117-41-6
57117-31-6
70648-29-9
40321-76-4
70648-26-9
57117-44-9
60851-34-5
7291 8-21-9
32598- 13-3
57753-85-7
19408-74-3
67562-39-4
55673-89-7
37871-00-4
12/89 QA/QC PCDD/PCDF
-------
These analyses are limited by lack of analytical standards; however isomer
specificity may be determined using specially developed standards. Analytical
results will, therefore, be reported as concentration (pg/g) for each gas
chromatography (GC) peak in a congener class by making the assumption that
the response for the molecular ion of all isomers in that class is equal to
the response observed for the isomer for which ERL-0 does have a standard.
The target minimum level of detection (MID) for specific PCDD/PCDF isomers is
given in Table 2 below. This document is meant to be only a guideline for
analyses and may be modified as needed to satisfactorily analyze any sample.
Target Minimum
k«vei_of .Detect ion.
TCDD, TCDF 1 pg/g
PeCDD, PeCDF 2 pg/g
HxCOO, HxCDF 4 pg/g
HpCDD, HpCDF 10 pg/g
12/89 QA/QC PCDD/PCDF
-------
I I. Sample Preparat ion
A. Grinding; Frozen fish wrapped in aluminum foil are sent to
the ERL-Duluth laboratory. How the fish is ground, (whole body
or fillet), is dependent on the species. Bottom feeders are ground
whole and predators are filleted with the skin off. Fish tissue is
ground frozen in a stainless steel power meat grinder. Each
sample is processed through the grinder three times which
homogenizes it thoroughly. The ground tissue is stored at
-20° C in solvent rinsed glass jars with aluminum lined
plastic lids.
B. Extraction; Tissue (20 g) is blended with enough anhydrous
sodium sulfate to dry the tissue (100 g). Two-thirds of the sample
is placed in a glass Soxhlet thimble, spiked with 100 ul of each Standard
Solution A and B (Table 3) and then the remainder of the sample
is added to the thimble. The sample is extracted at least twelve
hours with a 1:1 mixture of hexane and methylene chloride in a
Soxhlet extractor. The sample is quantitatively transferred to
a 500 ml Kuderna-Oanish apparatus and prewashed boiling chips
are added.
C. Percent Ii P i d Determination: The sample extracted in
section I.B. of sample preparation is used to determine percent
lipid. After sample concentration, the KO lower tube is placed in a
60° C water bath under a gentle stream of dry carbon filtered
air. After any remaining solvent has been evaporated, the lower
12/89 QA/QC PCDO/PCOF
-------
tube and contents are weighed. The lipid is then quantitatively
transferred to the macro column as described in Section I.D. of
sample preparation. After transfer, the empty lower tube and
boiling chips are weighed. The percent lipid is calculated from
the weight differences.
Table 3. Internal Standard Solutions.
Concentration Concentration
_Comgound_
37c
13c
i 3
15c
i •»
13C
13c
i 3
Il5c
i 3
13C
i 3
13C
i 3
13C
i 3
nic
3 7
irc
1,2
1,2
1,2
1,2
I,
4
12
12
12
12
12
12
12
12
1j
C
14
,3,
,4,
,3,
,3,
2,
2,
2,
1,
1,
1
1,
1,
1,
3,
3,
3,
2,
2,
?
2,
2,
2,
l
7.8-
7,8-
7,8-
3,7,
3,7,
3,4,
3,4,
3,4,
3,4,
nternal Standard
TCDD
TCDO
TCD
8
8
7
7
6
6
- P
- P
8
,8
,7
,7
F
eCDD
eCOF
- HxCOO
- HxCD F
,8-HpCOD
, 8- HpCD F
OCDO
2,
4-
7.
4-
6,
3,
7,8-
TCD
F
Internal
TCOD
8-
PeCDD
TCDF
7-
PeCO
F
2
5
5
5
5
12
12
12
12
25
2
Solution A. (100 uL
.0
.0
.0
.0
. 0
.5
.5
.5
.5
.0
.0
JL
1 0
25
25
25
25
62
62
62
62
125
10
.0
.0
. 0
. 0
. 0
.5
.5
.5
.5
.0
. 0
Standard Solution 8.
1
1
1
1
.0
.0
.0
. 0
5
5
5
5
.0
.0
. 0
. 0
Internal Standard Solution C.
13C12 1,2,3,4-TCDD 50.0 50.0
* Assumes a 20 g sample.
12/89 QA/QC PCDD/PCDF
-------
0. Anthropogeni c^ chemical Isolation: The sample extract is
quantitatively transferred to a 30 cm x 2.5 cm glass chromatography
column (MACRO-columns) fitted with a 300 ml reservoir on top.
The column has been packed with a plug of glass wool (bottom to
top), 2 g silica gel, 2 g potassium silicate, 2 g sodium sulfate
10 g celite/sulfuric acid and 2 g sodium sulfate, and previously
washed with 100 ml hexane. The column is eluted with 100 ml
benzene/hexane ( 5 X) and the eluent is collected in a Kuderna-Danish
(KD) apparatus (Caution: benzene is a known carcinogen). Isooctane
(1.0 ml) is added, the volume is reduced and then transferred to the
florisil column.
E. FIo r i s iI Chromatography: A 1.0 cm x 20.0 cm glass chromatography
column fitted with a 100 ml reservoir is packed with a plug of glass
wool (bottom to top), 5.0 cm (1.5 g) activated florisil and 1.0 cm
sodium sulfate. The florisil is activated at 120° C for 24 hours.
The column is washed with 20 ml methylene chloride followed by 10 ml
hexane. Sample and two 1 ml hexane rinses are quantitatively
applied in small "plugs". The column is eluted with 20 ml 2 %
methylene chloride/hexane and the eluate discarded. This wash is
followed by 50 ml methylene chloride which flows directly onto the
micro carbon/si lea gel column for PCDO/PCDF isolation.
f. PCDO/PCDF Isolation: Effluent from the florisil column is
passed onto a 4 mm x 200 mm column (micro-column) containing
300 mg silica gel/carbon (see sec. III.A.6) which was previously
rinsed with 10 ml toluene followed by 10 ml methylene chloride.
The column is fitted with a solvent reservoir. After the sample
has almost completely eluted from the micro-column, the reservoir
is washed twice with 2 ml 25X benzene/methylene chloride and the
12/89 QA/OC PCDD/PCDF 5
-------
column is finally eluted with an additional 11 m L 25X benzene/
methylene chloride. The column is inverted on the reservoir and
the PCDD/PCDF are eluted with toluene (25 ml). The toluene
fraction is collected in a pear shaped flask (25 ml) and reduced
in volume to 0.1 ml in a 60° C water bath under a gentle
stream of dry carbon filtered air. The sample is transferred to
a microvial using toluene to rinse the flask. Prior to GC/MS
analysis, the sample is allowed to evaporate to dryness and is
spiked with 20 ul of Standard Solution C (Table 3).
III. Reagents and Standards :
A. Reagents:
1. Solvents: Only pesticide grade distilled in glass solvents
are used. They are: hexane, isooctane, methylene chloride, benzene,
toluene, acetone, and methanol (Burdick and Jackson, Fischer
Scientific).
2. Sodium SuIfa t e; Sodium sulfate (Baker Chemical Company reagent
grade anhydrous) is baked at 650 C in a furnace for 24 hours,
cooled, and stored in an empty hexane solvent bottle.
3. Silica Gel: Silica-Gel-60 (Merck-Darmstadt), is Soxhlet
extracted eight hours with methanol, placed on solvent rinsed foil,
air dried for 12 hours, and vacuum oven dried (125 C) for 24
hours. It is stored in an empty hexane solvent bottle. Prior to
use it is activated at 105° C for 24 hours.
4. S u I f u r i c A c i d/C e I i t e : Sulfuric acid (Baker Chemical Company,
Ultrex) (5 ml) is blended in a 250 ml beaker with Celite 545
(Baker) (10 g) .
12/89 QA/QC PCDD/PCDF
-------
5. Potassium Silicate: High purity potassium hydroxide (Aldridge
Chemical Company) (56 9) is dissolved in methanol (300 ml).
Silica-gel (100 g) is added to the mixture and stirred (1 hour,
60° C). The mixture is cooled and the solvent is removed using
a Buchner funnel. The potassium silicate is rinsed twice with
100 ml of methanol and once with 100 ml of methylene chloride.
The solids are placed on aluminum foil in a fume hood and allowed
to dry for approximately 2 hours. The solids are placed in a vacuum
oven and dried overnight at 105 C. The reagent is placed in a
rinsed beaker and stored (activated) at 1 2 0 ° C until use.
6. Silica GeI/Carbon: Silica Gel-60 (100 g) (Merck-Darmstadt) is
Soxhlet extracted with methanol (200 ml) for 24 hours, air dried
in a hood, and further dried in vacuum oven for 24 hours. AMOCO
PX-21 Carbon (5 g) is added and then blended until uniform in
color. The Silica Gel/Carbon is stored in a closed jar at room
temperature until use.
7. FIo r i s iI; Fieri si I 60-100 mesh (Baker Analyzed) is soxhlet
extracted with methanot for 24 hours, placed on solvent rinsed
foil, air dried and stored in an empty hexane bottle. Prior to
use it is activated at 120°C for 24 hours.
B. Standards;
1. A n a I Y ti e a I Standard Spiking Solution
Table 3 provides details of the spiking solutions. The surrogate
analytes are used by the data reviewer to insure that calculated
MLD values are reasonable.
2. Quantification Standards: Quantification standards were prepared
by Wright State University. The concentration of 2,3,7,8-TCDO was
12/89 QA/QC PCDD/PCDF 7
-------
checked against a primary standard obtained from the U.S. National
Bureau of Standards. A table of the concentrations of each isomer
in each standard is given in Table 4.
3. Qualitative Standards; ERL-0 has developed two qualitative
analytical standards, one containing all 75 PCOO's and all 138
PCDF's was developed from an extraction of municipal incinerator
fly ash (Tables 5 and 6) and the other containing only the biosig-
nificant isomers was developed by exposure of fish to an extract
i
of municipal incinerator fly ash and processing the exposed fish
for PCDD/PCDF. These standards will be used to assign
structures for isomer specific ^analyses.
Standard solutions are sonicated for 5 to 10 minutes before use.
4. Mass Spec t rometer Mass Calibration Compounds: Perfluoro-
kerosene (PFK) is used for the initial mass calibration of the
mass spectrometer. Perfluorodecalin (P F 0) is used daily for
determining mass resolution on m/z 392.9761.
12/89 QA/QC PCDD/PCDF
-------
Table 4; Calibration Standards.
Concentrations in Calibration Solutions in pg/ul Tridecane
Calibration Standard
VM
U2
W3
W5
W6
W8
2,3,7,8-TCDO
2,3,7.8-TCDF
1,2,3,7,8-PeCDD
1,2,3,7,8-PeCDF
2,3,4,7,8-PeCDF
1,2,3,4,7,8-HxCDD
1,2,3,6,7,8-HxCDO
1,2,3,7,8,9-HxCDD
1,2,3,4,7,8-HxCDF
1.2,3.6,7,8-HxCDF
1.2,3,7,8,9-HxCDF
2,3,4,6,7,8-HxCDF
1,2,3,4,6,7,8-HpCDD
1,2,3,4,6,7,8-HpCDF
1,2,3,4,7,8,9-HpCDF
OCDD
OCDF
J3C12 2,3,7,8-TCDO
3C12 2,3.7,8-TCDF
13C12 1,2,3.7,8-PeCDD
J3C12 1,2,3,7,8-PeCDF
3C12 1,2,3,6,7,8-HxCDD
13C12 1,2,3,4,7,8-HxCDF
13C,, 1,2,3,4,6,7,8-HpCOO
13C12 1,2,3,4,6,7,8-HpCDF
13C,, OCDD
57 12
*'C14 2,3,7,8-TCDD
3^C14 2,3,7,8-TCDF
13C12 1,2,3,4-TCDD
200
200
200
200
200
500
500
500
500
500
500
500
500
500
500
1000
1000
50
50
50
50
125
125
125
125
250
20
20
50
100
100
100
100
100
250
250
250
250
250
250
250
250
250
250
500
500
50
50
50
50
125
125
125
125
250
20
20
50
50
50
50
50
50
125
125
125
125
125
125
125
125
125
125
250
250
50
50
50
50
125
125
125
125
250
20
20
50
25
25
25
25
25
62
62
62
62
62
62
62
62
62
62
125
125
50
50
50
50
125
125
125
125
250
20
20
50
.5
.5
.5
.5
.5
.5
.5
.5
.5
.5
10
10
10
10
10
25
25
25
25
25
25
25
25
25
25
50
50
50
50
50
50
125
125
125
125
250
20
20
50
5
5
5
5
5
12.5
12.5
12.5
12.5
12.5
12.5
12.5
12.5
12.5
12.5
25
25
50
50
50
50
125
125
125
125
250
20
20
50
2
2
2
2
2
6
6
6
6
6
6
6
6
6
6
12
12
50
50
50
50
125
125
125
125
250
20
20
50
.5
.5
.5
.5
.5
.25
.25
.25
.25
.25
.25
.25
.25
.25
.25
.5
.5
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
5
5
50
50
50
50
125
125
125
125
250
20
20
50
.5
.5
.5
.5
.5
.5
.5
.5
.5
.5
12/89 QA/OC PCOD/PCDF
-------
Compound
1368
1379
1369
1378
1469
1247
1248
1246
1249
1268
1478
1279
1234
1236
1269
1237
1238
2378
1239
1278
1267
1289
12468
12479
12469
12368
12478
RRT
DBS
0.814
0.838
0.861
0.912
0.912
0.912
0.912
0.921
0.921
0.934
0.940
0.960
0.985
0.98S
0.985
0.993
0.993
1 .000
1 .009
1 .028
1 .048
1 .079
1 .224
1 .224
1 .265
1 .293
1 .308
S
0
0
0
0
1
0
0
1
1
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
RRT
P2330
.826
.871
.948
.916
.072
.948
.948
.014
.014
.972
.990
.027
.014
.027
. 105
.014
.014
.000
. 088
.072
. 130
.216
1 1 1
1 1 1
268
148
188
Compound
12379
12369
12467
12489
12347
12346
12378
12367
12389
124679
124689
123468
123679
123689
123469
123478
123678
123467
123789
1234679
1234678
12346789
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
RRT
DBS
.320
.348
.348
.348
.368
.368
.400
,415
..443
. 620
.620
.673
. 700
.700
.700
.764
.775
.802
.802
.976
.023
.234
RRT
SP2330
1.209
1.307
1.321
1.321
1 . 268
1.352
1 .288
1 .363
1 .463
1 .473
1 .473
1 .473
1.546
1.546
1.681
1 .604
1.618
1 . 789
1 . 721
2.135
2.297
3 . 225
12/89 QA/QC PCDD/PCDF
10
-------
Compound
1368
1468
2468
1247
1347
1378
1346
2368
1367
1348
1379
1268
1248
1467
1478
1369
1237
2467
1234
2349
1236
1469
1238
1278
1349
1267
2378
2348
2347
2346
1246
1249
1279
2367
1239
1269
3467
1289
13468
12468
23479
12368
12478
13467
12467
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
1 .
1 .
1 .
1 .
1 .
1 .
1 .
1 .
RRT
DBS
730
752
763
782
782
782
782
782
801
801
801
835
835
853
853
863
863
863
880
880
880
880
880
902
920
920
939
939
939
939
939
939
939
973
988
988
988
071
120
120
190
202
202
202
202
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
1
0
0
0
1
0
1
1
1
1
1
1
1
0
1
1
1
RRT
SP2330
.777
.875
.989
.885
.865
.853
.919
.071
.881
.900
.853
.943
.919
.989
.943
.943
.943
. 109
.977
.977
.989
. 061
.989
.017
.013
.049
.169
.175
.140
. 193
.940
.071
.049
.206
. 140
.162
.264
.341
.008
.028
.065
. 103
.121
.142
. 160
Compound
1 3478
13479
23469
12479
13469
23468
12469
12347
12346
12348
12378
12367
23489
12379
23478
12489
13489
12369
23467
12349
12389
123468
134678
124678
134679
124679
124689
123467
123478
123678
123479
123469
123679
123689
234678
123789
123489
1234678
1234679
1234689
1234789
12346789
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
RRT
DBS
.202
.217
.217
.233
.253
.253
.253
.253
.253
.280
.280
.295
.309
.309
.359
.359
.359
.359
.371
.392
.446
.556
.570
.570
.570
.602
.621
.663
.663
.676
.676
.712
.730
.744
.744
.827
.827
.954
.979
.024
.043
.240
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
1
1
1
2
2
2
3
RRT
SP2330
. 083
. 103
. 173
. 142
.204
.278
.278
. 173
.231
.216
.216
.252
.388
.237
.557
.446
.350
.373
.612
.420
.590
.336
.370
.348
.348
.428
.521
.533
.489
.502
.489
.668
.562
.668
.012
.871
.940
.936
.001
. 161
.463
. 165
12/89 QA/OC PCDD/PCOF
11
-------
I V . 1 ns trumen t a I Parameters:
All gas chromatography/mass spectrometry analyses (GC/MS) will be done
on a Mnnigan-MAT 8230 high resolution GC/high resolution MS (HRGC/
HRMS) system. Instrumental parameters are given in Table 7.
Data Acquisition: Multiple Ion Selection Electric Sector Scan.
Compound Mass Window m/z value
Quant. Confir.
TCDF
37C t
13c 4
C12
TCDD
3^C I
13c 4
C12
PeCDF
13C
C12
PeCDO
13C
C12
HxCDF
13C
C12
HxCDD
13C
C12
HpCDF
13C
C12
HpCDD
13c
C12
OCDF
13C
C12
OCDD
13C12
-TCDF
-TCDF
-TCDD
-TCDD
-PeCDF
-PeCDD
-HxCDF
- HxCDD
- HpCDF
- HpCDD
-OCDF
-OCDD
1
1
1
1
1
1
2
2
2
2
3
3
3
3
4
4
4
4
5
5
5
5
305
311
317
321
327
333
339
351
355
367
373
385
389
401
407
419
423
435
443
455
459
471
.8986
.8898
.9389
.8936
.8847
.9338
.8597
.9000
.8546
.8949
.8207
.8610
.8156
.8559
. 781 7
.8220
.7766
.8169
.7498
.7801
.7348
.7750
303.
315.
319.
331 .
341 .
349.
353.
369.
375.
387.
391 .
403.
409 .
421 .
425.
437.
445 .
453.
457.
473.
9016
9419
8965
9368
8567
9029
8576
8919
8178
8580
81;>7
8530
7788
8191
7737
8140
7369
7831
7377
7721
Sample Introduction:
I o n i z a t i o n :
Source Pressure:
Ionizer Temperature:
Mass Resolution:
Scan Rate:
GC Column:
Linear Velocity:
Temperature Program:
Capillary Column, Splitless Injection.
Electron Impact, 70eV, 1mA Emission Current
1 X 10"5 torr .
250° C.
5000, 10% valley.
1 MIS cycle per second.
30 m DB-5, 60 m SP2330
35 cm/sec Helium.
180° C (hold 1 min); 13°/min to 200°;
3°/min to 270°;
270 hold 4 min.
Mass windows are monitored sequentially during the temperature
Quant. = Quantifeat ion ion; Confir. = Confirmation ion.
12/89 QA/QC PCDD/PCDF
12
-------
V. Quality Assurance/Quality Control (QA/QC)
A. General Procedures of Operati on
1. Analysis of Samples: Samples are analyzed in sets of
twelve consisting of:
a. Blank: Method Blank (extraction apparatus) is prepared in
the laboratory and subjected to the same sample preparation
procedures as environmental samples. The Method Blank is
used in every sample set.
b. Fortified Matrix: Native analytes (100 uL) (Table 8)
are added to a blank sample matrix. The levels of fortifi-
cation of native analytes in the matrix spike will be above
the target detection limit to provide an estimate of the
method's sensitivity, and for determination of percent
accuracy of quantification. This sample may be substituted
with a reference sample that has been analyzed at least
three times and a mean value of contamination has been
established.
c. Detection Limit Verification Sample; An environmental
sample with nondetectable amounts of native analyte (determined
from a previous analysis) will be spiked with native analytes
(Table 8) and analyzed with the next sample set. The addition
of the QA/QC sample will be done for only the first three
sample sets of any matrix type to establish that the
calculated MLD is achievable. If analytical results show
difficulty in obtaining the MLD, then this QA/QC sample must
be in each set. If no problem is experienced, then this
QA/QC sample may be dropped.
12/89 QA/OC PCDD/PCDF 13
-------
Table 8: Native PCDO/PCDF spiking solution (100 uL)
Compound
Concentration
(pg/uL Tridecane)
Solution A Solution B Solution C
2,
2,
1,
1,
2,
1,
1,
1
1
1.
2
1,
1
1
3,
3.
2,
2,
3.
2,
2,
?
?
2,
T
2,
?
?
7,
7,
3,
3,
4,
3,
3,
T
1
3,
4
3,
1
1
8
8
7
7
7
4
6
7
4
6
6
7
4
4
-T
- T
,8
,8
,8
,7
,7
8
7
, 7
7
,8
6
6
OCDD
COD
CDF
-PeCDD
-PeCDF
-PeCDF
, 8- HxCDD
,8-HxCDD
9- HxCDD
8-HxCDF
, 8- HxCD F
8- HxCDF
, 9- HxCD F
78- HpCDD
7 8- HpCDF
OCDF
0
0
0
0
0
1
1
1
1
1
1
1
1
1
2
2
.50
.50
.50
.50
.50
.25
.25
.25
.25
.25
.25
.25
.25
.25
.50
.50
1
1
1
1
1
2
2
2
2
2
2
2
2
2
5
5
. 00
.00
.00
.00
.00
.50
.50
.50
.50
.50
. 50
.50
. 50
. 50
.00
. 00
1
1
1
1
1
3
3
3
3
3
3
3
3
3
7
7
.50
.50
.50
.50
.50
.75
. 75
.75
. 75
.75
. 75
.75
. 75
. 75
.50
.50
d. 0 u pIi c a t e Sample: Two separate portions of the same
environmental sample are processed and analyzed.
e. Environmental Samples: The total number of environmental
samples analyzed is eight if the Detection Limit Verification
sample is used; otherwise nine samples are analyzed.
2. Sample Tracking and label ing of Samp Ies :
a. Logging Incoming Samples: ERL-D completes the chain of
custody forms and informs the Sample Control Center (SCC)
that samples arrived safely or informs SCC of any problems
with the samples. Each sample received by ERL-D had
previously been assigned two numbers by the Sample Control
Center, the Sample Control Center number ( S C C #) and an Episode
number. The SCC* number is unique for each sample and provides
12/89 OA/QC PCDD/PCDF
14
-------
a means for tracking a given sample throughout its analysis
and its permanent storage at the locker plant. The samples
are placed into freezer A upon arrival at ERL-Duluth,
homogenized, (see II.A.), and an aliquot (100-500 g) is placed
into freezer B. After the samples are extracted they are put
into freezer C. If all the data meets QA requirements after
mass spectral analysis and quantification, the samples are
transferred to a locker plant for permanent storage (-20° C).
b. Logging and labeling Samples During Preparation: A laboratory
identification code (lab ID) is randomly assigned to each
sample in a set of twelve at the start of sample preparation.
The code consists of a letter, A through L, date of
extraction, and two initials of the sample preparation
chemist, (e.g. A091587HL). This code is used to identify the
sample throughout the analysis period. The S C C #, lab ID,
sample description, weight of sample, and amount of analytical
standards added to each sample are recorded in the sample
preparation log book at the start of extraction. The lab
ID is written on labeling tape which is transferred from
beaker to flask during sample preparation. The lab ID is
written into the MS log book along with the mass spectra
analysis number.
3. Data System Sample Tracking: ERL-D has developed the National
Dioxin Study (NDS) Phase II, Bioaccumulative Pollutants in Fish:
Sample Tracking Database to facilitate record keeping and
summary report generation for each sample on the DEC-VAX 11/785
(Digital Equipment Corporation). For each sample, including QA
samples, information pertinent to each sample is entered into the
12/89 QA/QC PCDO/PCDF 15
-------
database. Quantification data (final concentration, ion ratios
percent recovery, M ID s , and signal to noise) are automatically
uploaded to the database once all OA criteria have been met.
Figure 1 is an example of the NDS database.
The first two letters of the SCC number indicate whether
the sample is an Environmental, Method or Matrix Blank,
Duplicate Sample or a mass spectral confirmation analysis of
an environmental sample. All environmental samples begin
with the letter D, or S if it is a mass spectral confirmation
analysis of a previously analyzed environmental sample.
The Blank and Duplicate samples begin with the letter Q
followed by a D or an R for duplicate or reference fish
sample, respectively. Table 9 lists the possible codes
for the SCC number, and matrix type. Episode numbers for
Blanks and Fortified Matrix samples are entered as 0000.
12/89 QA/QC PCDD/PCDF 16
-------
NDS Phase II: Bioaccumulative Pollutants in Fish:
Sample Tracking System ERL-D
loc:25
EP I SODE #: 0000
Sampling Information:
Sampling Office:
State & City:
Sampling Contact:
Date Sampl ed: 0/0/0
Site Location:
Latitude: N 0 0' 0'
Analysis Lab: D
Matrix Type: R
SCC #: QR071486
Analytical :
Extraction Date:
GC/MS ID:
LAB ID:
Weight:
% L i p i d :
PCDD/PCDF
7/14/86
MAT86824
K07H86LH
20.00
5 .2
Longitude: W 0 0' 0"
Date Received: O/ O/ 0
Rerun: 0
Pesticide & Industrial Chemicals
O/ O/ 0
0.00
0.0
Mass Lipid on GPC: 0.00
Comments: Reference fish 86
12/89 QA/OC PCDD/PCDF
17
-------
NDS Phase II: Bioaccumulative Pollutants in Fish
EPISODE *: 0000 SCC #: QR071486
DATA FOR BIOSIGNIFICANT POLYCHLOR I NAT ED D I BENZOD I OX INS AND FURANS:
Analytc CAS NO. I/R S/N XREC DL
2,3,7,8-TCDF
2 , 3,6,7-TCDF
3,4,6,7-TCDF
2,3,7,8-TCDD
1 ,2,3,7.8-PeCDF
2,3,4, 7,8-PeCDF
2,3,4,6,7-PeCDF
51207-31-9
1746-01-6
57117-41-6
57117-31-6
70648-29-9
0.74
1 .00
1 .71
0. 78
1 .33
1.10
0.00
0.0000
0.9726
0.4863
0 . 0000
1 .0892
1 .6357
2. 1784
ERL-D Loc: 25
Amount < pg/g)
5.26
ND
ND
15.63
ND
ND
ND
1,2,3
1,2,3
1,2,3
1.2,3
214
1,2,3
1,2,3
1,2,3
1,2,3
1,2,3
1,2,3
1,2,3
,7,8-
,4,6,
.4,7,
,6,7,
,6,7,
,7,8,
,4,7,
,6,7,
,7,8.
,4,6,
,4,7.
,4,6,
PeCDO
7
8
8
8
9
8
8
9
7
8
7
- HXCDF *
-HxCDF
- HXCDF
- HxCDF
-HxCDF
- HxCDO
- HxCDD
-HxCDO
,8-HpCDF
,9-HpCDF
, 8- HpCDD
40321 -
70648-
571 17-
60851 -
72918-
32598-
57753-
19408-
67562-
55673-
37871-
76-
26-
44-
34-
21 -
13-
85-
74-
39-
89-
00-
4
9
9
5
9
3
7
3
4
7
4
0.
0.
0.
1 .
0.
0.
1 .
0.
0.
0.
1 .
25
00
67
25
00
00
31
00
62
00
13
4.
57.
28.
57.
57.
29.
4.
29.
18.
37.
10.
24
03
52
03
03
08
67
08
97
94
50
57
47
47
47
47
49
49
49
39
39
39
4
0
1
0
0
1
0
1
0
0
0
.0729
.7327
.4654
.7327
.7327
.3863
.0000
.3863
.0000
.0000
. 0000
ND
ND
ND
ND
ND
ND
3.23
ND
ND
ND
5 . <3T
* Coelutes with 1,2,3,4,6,7-HxCDF on a DBS.
I/R - Ion Ratio; S/N = Signal to Noise; DL = Detection Limit
12/89 OA/QC PCDD/PCDF
18
-------
_Tabie_9.1 £odes_.£o r_ t h e_S C C_N umber_and_M a t r j_ x
SCC number first letter options:
D -- Environmental samples
Q - - QA samples
S •• MS confirmation analysis
Second letter options for Environmental Samples
A - Region 1
B - Region 2
C - Region 3
D • Region 4
E - Region 5
F - Region 6
G - Region 7
H - Region 8
Y - Region 9
J - Region 10
T - All regional data
Second letter options for QA samples:
8 - Method or matrix blank
0 - Labrotory duplicate
R - Reference fish or fortified matrix
Matrix Type:
PF •
WB -
UP -
BF -
R
Y
L
Predator Fillet
Whole Bottom
Whole Predator
Bottom Fillet
Reference
Blank
Laboratory Duplicate
12/89 OA/QC PCOO/PCDF
19
-------
B. 1ns t rumenta 1 Quality Control
1. Gas Chromatonraph
a. Qperat ion and Maintenance: Operation and maintenance of
the gas chromatograph will be done according to manufacturer's
recommendations.
b. Co I umn Performance: GC column performance will be
evaluated by:
i. Resolution of 1,2,3,4-TCDD from 2,3,7,8-TCOO
I
(Table 10).
i i . The R value of the regression of the sample
relative retention time of all biosignificant PCOD/PCDF,
to the library relative retention should not be less
than 0 .995.
iii. Elution of all PCDD/PCOF during analysis from a GC window
defining solution of select PCDO/PCDF (Table 11).
i __ ££_£2iu.m.n._P.S£l2£!D§n.2£_2u.2iil¥_c.2Dl£2i
Resolution of 1,2,3,4-TCDD from 2,3,7,8-TCDD will
be used to evaluate general column performance.
Resolution (R) must be 0.75 or greater.
2d
12/89 QA/OC PCDO/PCDF
20
-------
Congene£ Groug
TCDD 1,3,6,8 1,2,8,9
TCOF 1,3,6,8 1,2,8,9
PeCDD 1,2,4,7,9 /1,2,4,6,8 1,2,3,8,9
PeCDF 1,3,4,6,8 1,2,7,8,9
HxCDD 1,2,4,6,7,9 / 1,2,4,6,8,9 1,2,3,4,6,7
HxCDF 1,2,3,4,6,8 1,2,3,4,8,9
HpCDD 1,2,3,4,6,7,9 1,2,3,4,6,7,8
HpCDF 1,2,3,4,6,7,8 1,2,3,4,7,8,9
2. Mass Spectral Performance: The performance of the mass
spectrometer is evaluated for resolution, sensitivity and
linearity. The mass resolution used for these analyses is set at
a minimum of 5000 (10% valley definition). The mass spectrometer
is tuned each day to the required resolution according to the
procedures established by the instrument manufacturer. Sensitivity
and linearity is evaluated by the use of calibration standards
varying in concentration (Table 4). A calibration curve is
established for each standard. The curve must be linear over the
range of concentrations used in the calibration standards. The
percent relative standard deviations for the mean response factors
must be less than 20 percent.
C. Evaluation of Data:
1. Accuracy: Accuracy, the degree to which the analytical
measurement reflects the true level present, will be evaluated in
two ways for each sample set. These are: the difference of
measurement of a PCDD/PCDF isomer added to a blank matrix, or
difference of measurement of a PCDD/PCDF from the level in an
established reference material; and the efficiency for recovery
12/89 QA/QC PCDD/PCDF 21
-------
of the internal standard added for each congener group. The QA
requirements for accuracy and method efficiency are provided in
Table 12. Percent Accuracy and Percent Method Efficiency
are defined as follows:
% accuracy
% Method efficiency
measured value
amount native isomer
added to blank matrix
X 1 00
measured value
amount internal standard
added to each sample
X 100
T_able
Ion R a t i
TCDD
PCDD
HxCDD
HpCDD
OCDD
TCDF
PCOF
HxCDF
HpCDF
OCDF
*
* *
0
0
1
1
0
0
1
1
1
1
Var
Var
.76+ 1
.61+ 1
.23+ 1
.02+ 1
.88+ 1
.76+ 1
.53+ 1
.23+ 1
.02+ 1
.53+ 1
i a n c e
i a n c e
0
5X
5%
5%
5X
5%
5%
5%
5%
5%
5%
of
of
* * * *
Method Accuracy Precision
Efficiency at 10 pg/g at 10 pg/g
>40%, <120X
>40X, <120X
>40%, <120%
>40%, <1 20%
>40%, < 1 20%
>40%, < 1 20%
>40%, <120X
>40X, <1 20X
>40X, <120X
>40X, <120X
measured value from
difference of duplic
+50% +50X
+50% +50X
+100% +100%
+100% +100%
+.200% +100%
+50% +50%
+50X +50X
+100X +100X
+200X + 200X
+200X +200X
actual .
a t es f r om mean .
S/N
Minimum
3
3
3
3
3
3
3
3
3
3
.0
.0
.0
. 0
. 0
.0
.0
. 0
.0
.0
12/89 QA/QC PCDO/PCOF
22
-------
2. Precision; Precision, a measure of mutual agreement among
individual measurements of the same pollutant in replicate
samples, is evaluated for each sample set by the ratio of
the difference of duplicate values to their mean value.
Table 12 provides QA requirements for precision. Precision is
determined only when both values are above the detection limit.
Precision is defined as follows:
difference between duplicate samples
Precision * X 100
mean value for the duplicates
3. Signal Quality: The quality of the mass spectral signals used
for qualitative and quantitative analysis is evaluated
using two parameters: the ion intensity ratio for the two ions
monitored in each congener group, and the signal to noise (S/N)
ratio. Table 12 provides Q A requirements for signal quality.
In addition, qualitative identification will be based on
coelution with the stable isotope labeled compound, or relative
retention time correlation (Tables 5 and 6).
4. Polar Gas Ch romatographic Confirmation Analysis: Ten
percent of the sample extracts analyzed are seleceted for
GC/MS confirmation analysis on the more polar SP2330 column,
(Supelco, Belafonte, PA). Samples which were positive for
2,3,7,8-TCOO were selected for analysis.
12/89 QA/OC PCOO/PCDF 23
-------
D. Quality Assurance Problems and Corrective Actions:
Problem
MS performance outside QA
Co£r ec£j.ye_
Adjust MS parameters for resolution,
rerun initial curve and reanalyze
samp Ie(s).
Reanalyze standards and samples on
modified or alternate column.
If 2378-TCDD method efficiency < 4 0 % ,
reanalyze sample set. If method
efficiency < 4 0 X for analytes other
than 2378-TCDD, flag and report data.
If more than 20% of the analytes a,re
outside of QA for accuracy and pre-
cision, reanalyze the sample set.
Reextract and reanalyze all samples
for which the level of contamination,
or MID, is < 2.5 x blank level.
Record blank concentration in comment
field of samples.
Measure method efficiency. Dilute
sample 100:1 respike with each
standard solution (A and B ) , adjust
volume and reanalyze.
Reextract and reanalyze all positives
in set.
GC column performance
outside QA.
Method efficiency outside
of QA .
Accuracy outside of QA for
spiked matrix.
Precision of duplicates
outside QA.
Detection of analyte in
blank for 2,3,7,8- TCDD ,
2,3,7,8-TCDF and
1,2,3,7,8-PCDD
For other analytes in
bl ank
Analyte exceeds calibration
standard range.
Method efficiency for blank
outside of QA or blank lost
Because of the complexity of these analyses types, it is not expected that
all analytes will meet all QA criteria. Therefore, a complete review of
the data by a chemist is essential. Responsibility for the evaluation of
data is that of the sample preparation chemist and the mass spectrometer
operator. Review of the data, including QA, and resolution of data quality
problems is the responsibility of the Principal Investigator/Program Manager
Resolution of data questions may require reanalysis of samples to include
the addition of confirmatory ions or analysis on different types of
GC columns .
12/89 QA/QC PCOD/PCDF
24
-------
VI . Quantification Procedures
Quantification of analytes is accomplished by assigning isomer
identification, integrating the area of mass specific GC peaks, and
calculating an analyte concentration based upon an ion relative
response factor between the analyte and standard.
A. Initial and Dai I y C a I i b r a t i o n o f the HUMS: An initial calibration
of the instrument will be performed as needed. This will include
making three replicate injections of each calibration standard
(Table 4). Weighted least-squares linear regression is used to
generate a calibration curve for each analyte. The weighting factor
is inversely proportional to the variance among the replicate
injections of each calibration standard. The slope of the regression
line is the response factor used to quantify the analyte. At least
two calibration standards are injected daily to insure that any
response factors used for quantification and recovery calculations
do not deviate from the initial calibration by more than 20 percent.
If the daily calibration generates values outside this margin, and
less drastic corrective action does not solve the problem, a new set
of initial calibration curves is generated and the old response
factor libraries discarded. An example of a typical calibration
curve, using 2,3,7,8-TCDD as an example, is shown in Figure 2.
12/89 OA/QC PCOD/PCDF 25
-------
Figure 2
2,3,7,8-TCDD
WEIGHTED CALIBRATION CURVE
2 3
CONCENTRATION
SLOPE * RESPONSE FACTOR
CONCENTRATION/
12/89- QA/QC PCDD/PCDF
26
-------
B . Signal Quality
1. Minimum Level o f Detect ion {MLD): Minimum Level of Detection
is defined as the concentration predicted from the ratio of
baseline noise area to labeled standard area, plus three times
the standard error of the estimate derived from the initial
calibration curve for the analyte of interest.
Initial Cal ibration Based Method o f M L 0: MLD is estimated
from the ratio of the noise area to the isotopically labeled
internal standard area, plus three times the standard error of the
estimate (SE) for the area ratio, or Y-axis, of the initial
calibration curve. The Y-intercept (INT) is subtracted from this
quantity, in keeping with the normal formalism for "inverse
prediction" of a point on the X, or concentration ratio axis, from
a point on the Y, or signal ratio axis. The SE term is derived
from an analysis of variance ( A N 0 V A ) performed during the weighted
least squares fit of the initial calibration curve. This term
represents the random error in the replicate injections used to
generate the calibration curve, the error not accounted for by the
linear model. The weighting is necessary because of the relation
often observed in instrumental analysis, of increasing variance
with increasing concentration. MLD, according to this scheme,
is defined below:
[(NA/I334) + ( 3 x SE) - INT] x C334
MLD =
RF ( N/ I 334 ) x K
12/89 QA/QC PCDD/PCDF 27
-------
where: N A = noise area in the window for the major ion
of the native analyte,
1334 = labeled internal standard peak area in the
sample,
INT = the Y-axis intercept on the initial calibration
curve,
C334 = labeled internal standard concentration,
K = constant to adjust for sample size and final
volume,
RF(N/I334) = response factor for major native ion to
13C12 1,2,3,4-TCDD ion, the slope of the
initial calibration curve,
SE = standard error of the estimate of the initial
calibration curve.
In addition, fish tissue is spiked with surrogate analytes
(see Internal Standard Solution B, Table 3) prior to extraction.
The surrogate analytes serve as an added check to insure that
MID values calculated from the initial calibration curve,
as discussed above, are reasonable.
2. Signal t o Noise (S/N): The method of determining the signal
to noise ratio is shown below.
Analyte signal
Noise Signal
Analyte S'
Noise Signal Peak Area
12/89 QA/QC PCDD/PCDF 28
-------
S/N =
Analyte Signal Peak Area
Noise Signal Peak Area
The noise area is calculated by integrating over a peak width
equivalent to the analyte signal, typically about 10 seconds.
C. Quanti fication of PCDD/PCDF : The concentration of a natural
PCDD/PCDF is determined by calculating a response factor between
PCDD/PCDF and the stable isotope labeled PCDD/PCDF for the congener
group. Calculations are performed as follows:
Standard:
RF(N/L )
Sample:
AL x RF(N/L)
where: RF(N/L) = response factor native to labeled,
A ^ = peak area native,
AL = peak area labeled,
C N = concentration of native standard,
CL = concentration of labeled standard,
SL = labeled spiking level in sample,
Vw = level of native analyte in sample.
12/89 QA/OC PCDD/PCDF
29
-------
D. Method Efficiency: The method efficiency for the recovery of stable
isotope labeled compounds is determined by calculating the amount of
stable isotope labeled compound in the final extract and dividing by
the amount spiked into the sample at the start of the cleanup
procedure. This is done by determining the relative response factor
between the Internal Standard Solution C, C12 1,2,3,4-TCDD
and the stable isotope labeled internal standard (Solution A).
Determine Response Factor:
RF
AL x CIS
AIS * CL
where: RF = response factor,
A = area of stable isotope labeled
internal standard, (solution A),
A = area of 13C12 1 , 2 , 3 , 4 - TCDO ,
C, = concentration of stable isotope labeled
internal standard, (solution A),
CIS = concentration of 13C12 1,2,3,4-TCDD.
The response factor is then used in calculating the concentration
of the internal standard in the final solution,
CL
AL X CIS
AIS x RF
where: C. = concentration of stable isotope labeled
internal standard, (solution A).
12/89 QA/QC PCDD/PCDF 30
-------
The concentration in the final solution times the f •' n a I volume
equals the total amount present. The method efficiency is then
calculated by:
CL found
X Recovery = X 100
C L spiked
E. Integration o_f_ Automated Data Processing and Quality Assurance:
QA parameters for method efficiency, ion ratios,, retention time
correlations, signal/noise ratio, accuracy and precision are
monitored with the aid of software either developed in-house, or
modified from existing programs included with the HRMS data system.
Raw data is sorted and edited using the mass spectrometer's dedicated
data system, transferred to the DEC-VAX system and processed using
software programs RFACTOR and DFQUANT (Figure 3.). Data is reviewed
by the Project Director before entering into the NDS data base.
12/89 QA/QC PCDD/PCDF 31
-------
Figure 3
DATA REDUCTION FOR PCDD/PCDF
NATIONAL DIOXIN STUDY
YES
DAILY
CALIBRATION
STANDARDS
<7P)T ^
on i ^*
1
RFACTOR
SOFTWARE
1 /DATAX K
\. PASSES/
\QAr/
INITIAL
to. OAI IRRATIHN
^^ wMLIDnMMwIN
LIBRARIES
nrp \/AY
UCV^'VMA
OD
UR
IBM-PC
!
MASS
SPECTROMETER
DATA
SYSTEM
(BEGIN)
JO '
^"CORRECTIVE"^
ACTION
DATABASE -
lx^ I ^^LJ^\J b ^
SAMPLES
to. 9PT
^ on i
r .
DFQUANT
SOFTWARE
N0 /DAT\
^PASSES/
\QA?/
REVIEW \/
YES
GENERATE
1 FINAL REPORT
12/89 QA/QC PCDD/PCDF 32
GOVERNMENT PRINTING OFFICE 1990/748-159/00433
------- |