United States
Environmental Protection
Agency
Office ot
Toxic Substances
Washington DC 20460
EPA-560/5-85-011
April, 1985
Toxic Substances
Analytical Method:
The Analysis of
By-Product Chlorinated
Biphenyls in Air,
Revision 2
-------
ANALYTICAL METHOD: THE ANALYSIS OF BY-PRODUCT
CHLORINATED BIPHENYLS IN AIR, REVISION 2
by
Mitchell D. Erickson
WORK ASSIGNMENT NO. 6
SPECIAL REPORT NO. 2
EPA Contract No. 68-02-3938
MRI Project No. 8201-A(6)
May 20, 1985
For
U.S. Environmental Protection Agency
Office of Toxic Substances
Field Studies Branch, TS-798
401 M Street, SW
Washington, DC 20460
Attn: Joseph J. Breen, Project Officer
Daniel T. Heggem, Work Assignment Manager
-------
DISCLAIMER
This document has been reviewed and approved for publication by the
Office of Toxic Substances, Office of Pesticides and Toxic Substances, U.S.
Environmental Protection Agency. The use of trade names or commercial prod-
ucts does not constitute Agency endorsement or recommendation for use.
-------
PREFACE
This report contains an analytical method for the analysis of by-
product polychlorinated biphenyls in air. The work was done on Work Assign-
ment No. 6 on US Environmental Protection Agency Contract No. 68-02-3938.
This is the second revision of the method. Previous revisions are cited as
references 3 and 4 in the method. This report was prepared by Mitchell
Erickson. The work on the previous revisions was conducted by Dr. Erickson,
John Stanley, Kay Turman, Gil Radolovich, Karin Bauer, Jon Onstot, Donna Rose,
Margaret Wickham, and Ruth Blair. The work for the previous revisions was
performed on Task 51 of EPA Contract No. 68-01-5915.
Two companion methods have been published which address commercial
products and product wastes (Special Report No. 1, EPA Report No. EPA-560/5-
85-010) and water (Special Report No. 3, EPA Report No. EPA-560/5-85-012).
The EPA Work Assignment Manager, Daniel T. Heggem, of Field Studies
Branch provided helpful guidance.
MIDWEST RESEARCH INSTITUTE
Clarence L. Haile
P/rogram Ma
ger
in E. Going
^rogram Manager
Approved:
James L. Spigarelli, Director
Chemical and Biological Sciences
Department
-------
TABLE OF CONTENTS
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
11.0
12.0
13.0
14.0
15.0
16.0
17.0
References
Scope and Application
Summary
Interferences
Safety.
Apparatus and Materials
Reagents
Calibration
Sample Collection, Handling, and Preservation
Sample Preparation
Gas Chromatographic/Electron Impact Mass Spectrometric
Determination '
Qualitative Identification
Quantitative Data Reduction . . .
Confirmation
Quality Assurance
Quality Control '
Method Performance
Documentation and Records
Page
1
3
5
5
6
15
19
30
38
41
43
44
51
51
52
55
55
56
-------
LIST OF TABLES
Number Page
1 Numbering of PCB Congeners 2
2 DFTPP Key Ions and Ion Abundance Criteria 14
3 Concentrations of Congeners in PCB Calibration Standards
for Full Scan Analysis (ng/uL) 16
4 Concentrations of Congeners in PCB Calibration Standards
for Selected Ion Monitoring and Limited Mass Scan
Analysis (pg/uL) 17
5 Composition of Internal Standard Spiking Solution (SS100)
Containing 13C-Labeled PCBs 20
6 Operating Parameters for Capillary Column Gas Chromato-
graphic System 22
7 Operating Parameters for Packed Column Gas Chromatography
System 23
8 Operating Parameters for Quadrupole Mass Spectrometer
System 24
9 Operating Parameters for Magnetic Sector Mass Spectrometer
System. 25
10 Limited Mass Scanning (LMS) Ranges for PCBs 26
11 Characteristic SIM Ions for PCBs 27
12 Pairings of Analyte and Calibration Compounds 29
13 Relative Retention Time (RRT) Ranges of PCB Homologs Versus
d6-3,3' ,4,4'-Tetrachlorobiphenyl 31
14 Characteristic Ions for Internal Standards and 13C-Labeled
PCB Surrogates. 46
15 Analysis Report 50
VI
-------
LIST OF FIGURES
Number Page
1 PCB sampling train for stack gases 7
2 Florisil adsorbent tube 10
3 PCB sampling train for static air 11
4 Field data sheet. . 35
5 Reconstructed ion chromatogram of calibration solution
FS100 obtained in the full scan mode 42
vn
-------
THE ANALYSIS OF BY-PRODUCT CHLORINATED BIPHENYLS IN AIR
1-0 Scope and Application
1.1 This is a gas chromatographic/electron impact mass spectrometric
(GC/EIMS) method applicable to the determination of chlorinated
biphenyls (PCBs) in air emitted from commercial production through
stacks, as fugitive emissions, or static (room, other containers,
or outside) air. The PCBs present may originate either as syn-
thetic by-products or as contaminants derived from commercial PCB
products (e.g., Aroclors). The PCBs may be present as single
isomers or complex mixtures and may include all 209 congeners from
monochlorobiphenyl through decachlorobiphenyl listed in Table 1.
This method was prepared for use in demonstrating compliance with
the EPA rules regarding the generation of PCBs as byproducts in
commercial chemical production1'2 and is based on earlier ver-
sions.3'4 This revision includes elimination of a calculation
which corrects the native PCB concentration based on the recovery
of the 13C-labeled PCB recovery surrogates. In addition, full
scan is now emphasized over the selected ion monitoring and lim-
ited mass scan mass spectrometric data collection modes. The
latter two techniques provide less qualitative information and
should be used only if needed to achieve the required sensitivity.
Additional background information on selection of the techniques
has also been published.5
1.2 The detection and quantitation limits are dependent upon the vol-
ume of sample collected, the complexity of the sample matrix and
the ability of the analyst to remove interferents and properly
maintain the analytical system. The method accuracy and preci-
sion will be determined in future studies.
1.3 This method is restricted to use by or under the supervision of
analysts experienced in the use of gas chromatography/mass spec-
trometry (GC/MS) and in the interpretation of gas chromatograms
and mass spectra. Prior to sample analysis, each analyst must
demonstrate the ability to generate acceptable results with this
method by following the procedures described in Section 15.2.
1.4 During the development and testing of this method, certain analyt-
ical parameters and equipment designs were found to affect the
validity of the analytical results. Proper use of the method re-
quires that such parameters or designs must be used as specified.
These items are identified in the text by the word "must."
-------
TABLE 1. NTJMBERING OF PCB CONGENERS3
1
2
3
4
5
6
7
8
9
in
n
i?
n
14
15
16
17
18
19
20
22
24
?6
27
28
29
30
31
32
33
14
IS
Ifi
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
2
3
4
Olctilorobipheiiyls
2,2'
2.3
2.3'
2.4
2.4'
2.5
2,6
3.3'
3,4
3,4'
3,5
4,4*
THcMornblpheriyli
2.2'. 3
2, 2', 4
2,2',5
2,2', 6
2,3.3'
2.3,4
2,3.4'
2.3,5
2.3.6
2. 3', 4
2,3'. 5
2, 3'. 6
2,4,4'
2,4.5
2,4.6
2, 4', 5
2.4'.S
2', 3, 4
2'. 3,5
3. 3', 4
3.3',5
3.4,4'
3,4.5
3. 4'. 5
Tgtrartiloroblphtnyls
2.2*. 3,3*
2,2*,3,4
2.2', 3,4'
2.2'.3.5
2.2'. 3,5'
2.2',3,6
2.2'. 3.6'
2.2*.4.4*
2.2'. 4.5
2,2'. 4.5*
2,2'.4.6
2.2'.4.6'
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
RO
81
82
83
84
as
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
TftraehloroblphefiyU
2.2'.5.5*
2,2*.5.6'
2.2'.6.6'
2.3.3',4
2.3, 3*. 4'
2.3,3',5
2.3.3'. 5'
2.3.3'. 6
2.3,4,4'
2.3,4.5
2.3.4.6
2.3. 4'. 5
2.3.4', 6
2,3.5,6
2,3*. 4. 4'
2.3*.4,5
2.3*. 4.5'
2,3*. 4.6
2,3*. 4*. 5
2,3', 4', 6
2,3* .5,5'
2.3*.S'.6
2,4,4'. 5
2.4,4', 6
2'. 3,4, 5
3.3' .4,4'
3,3*. 4,5
3,3'. 4,5'
3,3', 5,5*
3,4,4'.S
Pentaeti 1 oroM pHeny 1 s
2.2'.3.3'.4
2.2'. 3.3*. 5
2,2*.3,3',6
2.2*. 3.4,4*
2.2*. 3,4.5
2,2'. 3. 4,5'
2.2*,3,4,6
2, 2', 3, 4, 6'
2,21.3,4',5
2, 2'. 3. 4'. 6
2.2*.3,S,5'
2,2'. 3.5,6
2,2*.3.5,6*
2.2'. 3,5'. 6
2,2', 3. 6,6'
2,2',3'.4,5
2.2*,3'.4.6
2.2',4.4'.5
2,2*,4.4*,6
2.2', 4,5.5'
2.2' ,4,5.6'
2,2'. 4,5*. 6
2,2*,4.6.6*
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
Pentichl orob 1 pheny 1 s
2,3.3'. 4.4'
2.3.3' ,4.5
2,3,3'.4',5
2,3,3'.4.5'
2.3, 3', 4, 6
2.3.3' .4'. 6
2.3.3'. 5.5'
2.3,3*.5.6
2.3.3*. 5'. 6
2.3.4.4'.5
2.3,4,4'. 5
2.3,4,5,6
2,3,4', 5, 6
2.3'. 4, 4' .5
2.3*. 4.4'. 6
2.3*. 4.5. 5'
2,3*, 4, 5'. 6
2*.3,3'.4,5
2*. 3,4,4'. 5
2* ,3, 4,5, 5'
2'. 3. 4.5, 6'
3.3' .4.4' .5
3,3', 4, 5,5'
H»xach1ereb
-------
Anyone wishing to deviate from the method in areas so identified
must demonstrate that the deviation does not affect the validity
of the data. Alternative test procedure approval must be obtained
from the Agency. An experienced analyst may make modifications
to parameters or equipment identified by the term "recommended."
Each time such modifications are made to the method, the analyst
must repeat the procedure in Section 15.2. In this case, formal
approval is not required, but the documented data from Section
15.2 must be on file as part of the overall quality assurance pro-
gram.
1.5 This method contains many options because of the diversity of ma-
trices and interferences which may be encountered. Once the ap-
propriate options for each sample type have been selected, each
laboratory should prepare a written step-by-step protocol for use
by the analysts. The protocol may contain verbatim sections from
this method, more detailed steps for certain techniques, or totally
different extraction or cleanup techniques.
2.0 Summary
2.1 The air must be sampled such that the specimen collected for
analysis is representative of the whole. Statistically designed
selection of the sampling position (stack, flue, port, etc.) or
time should be employed. Gaseous and particulate PCBs should be
withdrawn isokinetically from stacks, room air exhausts, process
point exhausts, and other flowing gaseous streams using a sam-
pling train. While several sampling methods are available for
collection of PCBs, the modified EPA Method 56 described herein
has been well-validated for PCB collection and recovery. In the
modified EPA Method 5, PCBs are collected in a Florisil adsorbent
tube and in a series of impingers in front of the adsorbent.
Other sample collection systems may be used, provided PCBs are
adequately collected by and recovered from the train.
PCBs are sampled from ambient air and other static gaseous sources
onto a Florisil adsorbent tube. The sample must be preserved to
prevent PCB loss prior to analysis. Storage at 4°C is recommended.
2.2 The Florisil adsorbent is extracted with hexane in a Soxhlet ex-
tractor, the aqueous condensate is extracted with hexane and the
acetone/hexane impinger rinse is back-extracted with water. All
three organic extracts are then combined. Optional cleanup tech-
niques may include sulfuric acid cleanup and Florisil adsorption
chromatography. The sample is concentrated to a final known vol-
ume for instrumental determination.
2.3 The PCB content of the sample extract must be determined by high
resolution (preferred) or packed column gas chromatography/electron
impact mass spectrometry (HRGC/EIMS or PGC/EIMS) operated in the
full scan, selected ion monitoring (SIM), or limited mass scan
(IMS) mode.
-------
2.4 PCBs are identified by comparison of their retention time and
mass spectral intensity ratios to those in calibration standards.
2.5 PCBs are quantitated against the response factors for a mixture
of 10 PCB congeners using the internal standard technique.
2.6 The PCBs identified by the SIM technique may be confirmed by full
scan HRGC/EIMS, retention on alternate GC columns, other mass spec-
trometric techniques, infrared spectrometry, or other techniques,
provided that the sensitivity and selectivity of the technique are
demonstrated to be comparable or superior to GC/EIMS.
2.7 The analysis time is dependent on the extent of workup employed.
The time required for instrumental analysis of a single sample
excluding instrumental calibration, data reduction, and report-
ing, is typically 30 to 45 min.
2.8 A quality assurance (QA) plan must be developed for each labora-
tory.
2.9 Quality control (QC) measures include laboratory certification
and performance check sample analysis, procedural QC (instru-
mental performance, calculation checks), and sample QC (blanks,
replicates, and standard addition).
2.10 While several options are available throughout this method, the
recommended procedure for stack gases to be followed is:
2.10.1 The sample is collected using a modified Method 5 train6
according to a scheme which permits extrapolation of the
sample data to the source being assessed.
2.10.2 The sample is preserved at 4°C to prevent any loss of
PCBs or changes in matrix which may adversely affect re-
covery.
2.10.3 The three sample fractions are extracted and combined.
2.10.4 The extract is cleaned up and concentrated to an appro-
priate volume. Internal standards are added.
2.10.5 An aliquot of the extract is analyzed by HRGC/EIMS oper-
ated in the SIM mode. On-column injections onto a 15-m
DB-5 capillary column, programmed (for toluene solutions)
from 110° to 325°C at 10°/min after a 2 min hold is used.
Helium at 45-cm/sec linear velocity is used as the car-
rier gas.
2.10.6 PCBs are identified by retention time and mass spectral
intensities.
-------
2.10.7 PCBs are, quantitated against the response factors for a
mixture of 10 PCB congeners.
2.10.8 The total PCBs are obtained by summing the amounts for
each homolog found, and the concentration is reported as
micrograms per cubic meter.
3.0 Interferences
3.1 Method interferences may be caused by contaminants, in sample
collection media, solvents, reagents, glassware, and other sample
processing hardware, leading to discrete artifacts and/or elevated
baselines in the total ion current profiles. All of these ma-
terials must be routinely demonstrated to be free from interfer-
ences by the analysis of laboratory reagent blanks as described
in Section 15.
3.1.1 Glassware must be scrupulously cleaned. All glassware
should be cleaned as soon as possible after use by rins-
ing with the last solvent used. This should be followed
by detergent washing with hot water and rinses with tap
water and reagent water. The glassware should then be
drained dry and heated in a muffle furnace at 400°C for
15 to 30 min. Some thermally stable materials, such as
PCBs, may not be eliminated by this treatment. Solvent
rinses with acetone and pesticide quality hexane may be
substituted for the muffle furnace heating. Volumetric
ware should not be heated in a muffle furnace. After it
is dry and cool, glassware should be sealed and stored
in a clean environment to prevent any accumulation of
dust or other contaminants. It is stored inverted or
capped with aluminum foil.
3.1.2 The use of high purity reagents and solvents helps to
minimize interference problems. Purification of solvents
by distillation in all-glass systems may be required.
All solvent lots must be checked for purity prior to use.
3.2 Matrix interferences may be caus'ed by contaminants that are coex-
tracted from the sorbent material or impingers. The extent of
matrix interferences will vary considerably from source to source,
depending upon the nature and diversity of the sources of samples.
4.0 Safety
4.1 The toxicity or carcinogenicity of each reagent used in this
method has not been precisely defined; however, each chemical
compound should be treated as a potential health hazard. From
this viewpoint, exposure to these chemicals must be reduced to
-------
the lowest possible level by whatever means available. The lab-
oratory is responsible for maintaining a current awareness file
of OSHA regulations regarding the safe handling of the chemical
specified in this method. A reference file of material data han-
dling sheets should also be made available to all personnel in-
volved in the chemical analysis.
4.2 Polychlorinated biphenyls have been tentatively classified as
known or suspected human or mammalian carcinogens. Primary
standards of these toxic compounds should be prepared in a hood.
Personnel must wear protective equipment, including gloves and
safety glasses.
Congeners highly substituted at the meta and para positions and
unsubstituted at the ortho positions are reported to be the most
toxic. Extreme caution should be taken when handling these com-
pounds neat or in concentrated solution. The class includes
3,3',4'4'-tetrachlorobiphenyl (both natural abundance and isotop-
ically labeled).
4.3 Waste disposal must be in accordance with RCRA and applicable
state rules.
5.0 Apparatus and Materials
All specifications are suggestions only. Catalog numbers and suppliers
are included for illustration only.
5.1 Stack sampling train6 - See Figure 1; a series of four impingers
with a solid adsorbent trap between the third and fourth impingers.
The train may be constructed by adaptation from a Method 5 train.3
Descriptions of the train components are contained in the follow-
ing subsections.
5.1.1 Probe nozzle - Stainless steel (316) with sharp, tapered
leading edge. The angle of taper shall be ^ 30° and the
taper shall be on the outside to preserve a constant in-
ternal diameter. The probe nozzle shall be of the button-
hook or elbow design, unless otherwise specified by the
Agency. The wall thickness of the nozzle shall be less
than or equal to that of 20 gauge tubing, i.e., 0.165 cm
(0.065 in.) and the distance from the tip of the nozzle
to the first bend or point of disturbance shall be at
least two times the outside nozzle tubing. Other con-
figurations and construction material may be used with
approval from the Agency.
5.1.2 Probe liner - Borosilicate or quartz glass equipped with
a connecting fitting that is capable of forming a leak-
free, vacuum tight connection without sealing greases;
-------
Stack
Wall
Thermometer
Florisil Tube
Check
Valve
Probe tf
Reverse-Type"^
Pi tot Tube
t >
Flow
Control Box
Figure 1. PCB sampling train for stack gases.
-------
such as Kontes Glass Company "0" ring spherical ground
ball joints (model K-671300) or University Research
Glassware SVL teflon screw fittings.
A stainless steel (316) or water-cooled probe may be used
for sampling high temperature gases with approval from
the Agency. A probe heating system may be used to prevent
moisture condensation in the probe.
5.1.3 Pitot tube - Type S, or equivalent, attached to probe to
allow constant monitoring of the stack gas velocity.
The face openings of the pitot tube and the probe nozzle
shall be adjacent and parallel to each other but not
necessarily on the same plane, during sampling. The free
space between the nozzle and pitot tube shall be at least
1.9 cm (0.75 in.). The free space shall be set based on
a 1.3 cm (0.5 in.) ID nozzle, which is the largest size
nozzle used.
The pitot tube must also meet the criteria specified in
Method 27 and be calibrated according to the procedure
in the calibration section of that method.
5.1.4 Differential pressure gauge - Inclined manometer capable
of measuring velocity head to within 10% of the minimum
measured value. Below a differential pressure of 1.3 mm
(0.05 in.) water gauge, micromanometers with sensitivities
of 0.013 mm (0.0005 in.) should be used. However, micro-
manometers are not easily adaptable to field conditions
and are not easy to use with pulsating flow. Thus, other
methods or devices acceptable to the Agency may be used
when conditions warrant.
5.1.5 Impingers - Four impingers with connecting fittings able
to form leak-free, vacuum tight seals without sealant
greases when connected together as shown in Figure 1.
The first and second impingers are of the Greenburg-
Smith design. The final two impingers are of the
Greenburg-Smith design modified by replacing the tip
with a 1.3 cm (1/2 in.) ID glass tube extending to 1.3
cm (1/2 in.) from the bottom of the flask.
One or two additional modified Greenburg-Smith impingers
may be added to the train between the third impinger and
the Florisil tube to accommodate additional water col-
lection when sampling high moisture gases. Throughout
the preparation, operation, and sample recovery from the
train, these additional impingers should be treated
exactly like the third impinger.
-------
5.1.6 Solid adsorbent tube -.Glass with connecting fittings
able to form leak-free, vacuum tight seals without seal-
ant greases (Figure 2). Exclusive of connectors, the
tube has a 2.2 cm inner diameter, is at least 10 cm long,
and has four deep indentations on the inlet end to aid
in retaining the adsorbent. Ground glass caps (or
equivalent) must be provided to seal the adsorbent-filled
tube both prior to and following sampling.
5.1.7 Metering system - Vacuum gauge, leak-free pump, thermom-
eters capable of measuring temperature to within ±3°C
(^ 5°F), dry gas meter with 2% accuracy at the required
sampling rate, and related equipment, or equivalent, as
required to maintain an isokinetic sampling rate and to
determine sample volume. When the metering system is
used in conjunction with a pitot tube, the system shall
enable checks of isokinetic rates.
5.1.8 Barometer - Mercury, aneroid, or other barometers cap-
able of measuring atmospheric pressure to within 2.5 mm
Hg (0.1 in. Hg). In many cases, the barometric reading
may be obtained from a nearby weather bureau station, in
which case the station value shall be requested and an
adjustment for elevation differences shall be applied at
a rate of -2.5 mm Hg (0.1 in. Hg) per 30 m (100 ft) ele-
vation increase.
5.2 Static air sampling train6 - The sampling train, see Figure 3,
consists of a glass-lined probe, an adsorbent tube containing
Florisil, and the appropriate valving and flow meter controls for
isokinetic sampling as described in Section 5.1. The sampling
apparatus in Figure 3 is the same as that in Figure 1 and Section
5.1, except that the Smith-Greenburg impingers and heated probe
are not used. If condensation of significant quantities of mois-
ture prior to the solid adsorbent is expected, Section 5.1 of the
method should be used. Since probes and adsorbent tubes are not
cleaned up in the field, a sufficient number must be provided for
sampling and allowance for breakage.
5.3 Sample recovery
5.3.1 Ground glass caps - To cap off adsorbent tube and the
other sample exposed portions of the train.
5.3.2 Teflon FEP® wash bottle - Two, 500 mL, Nalgene No.
0023A59 or equivalent.
5.3.3 Sample storage containers - Glass bottles, 1 liter, with
TFE®-lined screw caps.
5.3.4 Balance - Triple beam, Ohaus Model 7505 or equivalent.
-------
J28/12
10cm
j 28/12
Figure 2. Florisil adsorbent tube.
10
-------
Probe (to sample from duct) ^M
Glass- lined Probe
Orifice
Manometer :-
Type S
Pitot Tube
/ Integrated j
1 Flow Meter I
Florisil
Glass Wool
Check Valve
Air
Tight
Pump
Vacuum
Line
Figure 3. PCB sampling train for static air.
11
-------
5.3.5 Aluminum foil - Heavy duty.
5.3.6 Metal can - To recover used silica gel.
5.4 Analysis
5.4.1 Glass Soxhlet extractors - 40 mm ID complete with 45/50
J condenser, 24/40 I 250 mL round-bottom flask, heating
mantle for 250 ml flask, and power transformer.
5.4.2 Teflon FEP wash bottle - Two, 500 ml, Nalgene No. 0023A59
or equivalent.
5.4.3 Separatory funnel - 1,000 ml with TFE® stopcock.
5.4.4 Kuderna-Danish concentrators - 500 mL.
5.4.5 Steam bath.
5.4.6 Separatory funnel - 50 mL with TFE® stopcock.
5.4.7 Volumetric flask - 25.0 mL, glass.
5.4.8 Volumetric flask - 5.0 mL, glass.
5.4.9 Culture tubes - 13 x 100 mm, glass with TFE®-lined screw
caps.
5.4.10 Pipette - 5.0 mL glass.
5.4.11 Teflon®-glass syringe - 1 mL, Hamilton 1001 TLL or
equivalent with Teflon® needle.
5.4.12 Syringe - 10 |jL, Hamilton 701N or equivalent.
5.4.13 Disposable glass pipettes with bulbs - To aid transfer
of the extracts.
5.4.14 Gas chromatography/mass spectrometer system.
5.4.14.1 Gas chromatograph - An analytical system com-
plete with a temperature programmable gas chro-
matograph and all required accessories includ-
ing syringes, analytical columns, and gases.
The injection port must be designed for on-
column injection when using capillary columns
or packed columns. Other capillary injection
techniques (split, splitless, "Grob," etc.)
may be used provided the performance specifi-
cations stated in Section 7.1 are met.
12
-------
5.4.14.2 Capillary GC column - A 10-30 m long x 0.25 mm
ID fused silica column with a 0.25 urn thick
DB-5 bonded silicone liquid phase (J&W Scien-
tific) is recommended. Alternate liquid phases
may include OV-101, SP-2100, Apiezon L, Dexsil
300, or other liquid phases or columns which
meet the performance specifications stated in
Section 7.1.
5.4.14.3 Packed GC column - A 180 cm x 0.2 cm ID glass
column packed with 3% SP-2250 on 100/120 mesh
Supelcoport or equivalent is recommended.
Other liquid phases or columns which meet the
performance specifications stated in Section
7.1 may be substituted.
5.4.14.4 Mass spectrometer - Must be capable of scanning
from m/z 150 to m/z 550 every 1.5 sec or less,
collecting at least five spectra per chromato-
graphic peak, utilizing a 70-eV (nominal) elec-
tron energy in the electron impact ionizaton
mode and producing a mass spectrum which meets
all the criteria in Table 2 when 50 ng of deca-
fluorotriphenyl phosphine [DFTPP, bis(perfluoro-
phenyl)phenyl phosphine] is injected through
the GC inlet. Any GC-to-MS interface that
gives acceptable calibration points at 10 ng
per injection for each PCB isomer in the cali-
bration standard and achieves all acceptable
performance criteria (Section 10) may be used.
Direct coupling of the fused silica column to
the MS is recommended. Alternatively, GC to
MS interfaces constructed of all glass or glass-
lined materials are recommended. Glass can be
deactivated by silanizing with dichlorodimethyl-
silane.
5.4.14.5 A computer system that allows the continuous
acquisition and storage on machine-readable
media of all mass spectra obtained throughout
the duration of the chromatographic program
must be interfaced to the mass spectrometer.
The data system must have the capability of
integrating the abundances of the selected
ions between specified limits and relating
integrated abundances to concentrations using
the calibration procedures described in this
method. The computer must have software that
allows searching any GC/MS data file for ions
of a specific.mass and plotting such ion abun-
dances versus time or scan number to yield an
13
-------
Table 2. DFTPP Key Ions and Ion Abundance Criteria
m/zIon abundance criteria
197 Less than 1% of mass 198
198 100% relative abundance
199 5-9% of mass 198
275 10-30% of mass 198
365 Greater than 1% of mass 198
441 Present, but less than mass 443
442 Greater than 40% of mass 198
443 17-23% of mass 442
14
-------
extracted ion current profile (EICP). Software
must also be available that allows integrating
the abundance in any EICP between specified
time or scan number limits.
5.5 Gas chromatograph for GC/FID screening
5.5.1 A temperature-programmable GC equipped with a flame
ionization detector Varian 3740 or equivalent.
5.5.2 A 2 m x 2 mm ID glass column packed with 3% SP-2250
on 100/120 mesh Supelcoport or equivalent. A high
resolution GC column may also be used.
6.0 Reagents
6.1 Sampling
6.1.1 Florisil - Floridin Company, 30/60 mesh, Grade A. The
Florisil is cleaned by 8 hr Soxhlet extraction with hex-
ane and then by drying for 8 hr in an oven at 110°C and
is activated by heating to 650°C for 2 hr (not to exceed
3 hr) in a muffle furnace. After allowing to cool to
near 110°C transfer the clean, active Florisil to a clean,
hexane-washed glass jar and seal with a TFE®-lined lid.
The Florisil should be stored at 110°C until taken to
the field for use. Florisil that has been stored more
than 1 month must be reactivated before use.
6.1.2 Glass wool - Cleaned by thorough rinsing with hexane,
dried in a 110°C oven, and stored in a hexane-washed
glass jar with TFE®-lined screw cap.
6.1.3 Water - Deionized, then glass-distilled, and stored in
hexane-rinsed glass containers with TFE®-lined screw caps.
6.1.4 Silica gel - Indicating type, 6-16 mesh. If previously
used, dry at 175°C for 2 hr. New silica gel may be used
as received.
6.1.5 Crushed ice.
6.2 Solvents - All solvents must be pesticide residue analysis grade.
New lots should be checked for purity by concentrating an aliquot
by at least as much as is used in the procedure.
6.3 Calibration standard congeners - Standards of the PCB congeners
listed in Tables 3 and 4 are available from Ultra Scientific,
Hope, Rhode Island; or Analabs, North Haven, Connecticut.
15
-------
Table 3. Concentrations of Congeners in PCB Calibration Standards
for Full Scan Analysis (ng/(jL)
Homolog
1
2
3
4
5
6
7
8
9
10
4
-
-
13C-Cli
13C-C14
13c-ci8
13c-ci10
Congener
no.
1
7
30
50
97
143
183
202
207
209
210 (IS)
C10H7I (IS)b
Ci8Di2 (IS)C
211 (RS)
212 (RS)
213 (RS)
214 (RS)
FS100
ng PCB
100
100
150
200
200
200
300
300
450
200
250
250
250
100
250
400
500
FS050
ng PCB
50
50
75
100
100
100
150
150
225
100
250
250
250
50
125
200
250
FS010
ng PCB
10
10
15
20
20
20
30
30
45
20
250
250
250
10
25
40
50
FS005
ng PCB
5
5
7.5
10
10
10
15
15
22.5
10
250
250
250
5
12.5
20
25
FS001
ng PCB
1
1
1.5
2
2
2
3
3
4.5
2
250
250
250
1
2.5
4
5
Concentrations given as examples only.
1-lodonaphthalene.
d12~Chrysene.
16
-------
Table 4. Concentrations of Congeners in PCB Calibration Standards
for Selected Ion Monitoring and Limited Mass Scan
Analysis (pg/|A)
Homolog
1
2
3
4
5
6
7
8
9
10
4
-
-
13C-Cli
13C-C14
13c-ci8
13c-ci10
Congener
no.
1 .
7
30
50
97
143
183
202
207
209
210 (IS)
C10H7I (IS)b
C18D12 (IS)C
211 (RS)
212 (RS)
213 (RS)
214 (RS)
SIM1000
pg PCB
1,000
1,000
1,500
2,000
2,000
2,000
3,000
3,000
4,500
2,000
250
250
250
1,000
2,500
4,000
5,000
SIM100
pg PCB
100
100
150
200
200
200
300
300
450
200
250
250
250
100
250
400
500
SIM050
pg PCB
50
50
75
100
100
100
150
150
225
100
250
250
250
50
125
200
250
SIM010
pg PCB
10
10
15
20
20
20
30
30
45
20
250
250
250
10
25
40
50
.Concentrations given as examples only.
1-Iodonaphthalene.
d12~Chrysene.
17
-------
6.4 Calibration standard stock solutions - Primary dilutions of each
of the individual PCBs listed in Tables 3 and 4 are prepared by
weighing approximately 1-10 mg of material within 1% precision.
The PCB is then dissolved and diluted to 1.0 ml with hexane. The
concentration is calculated in mg/mL. The primary dilutions are
stored at 4°C in screw-cap vials with Teflon cap liners. The
meniscus is marked on the vial wall to monitor solvent evaporation.
Primary dilutions are stable indefinitely if the seals are main-
tained. The stock solutions and dilutions should be clearly la-
beled with pertinent information such as sample code, solvent,
date prepared, initials of person preparing the solution, and
notebook reference.
6.5 Working calibration standards - Working calibration standards are
prepared that are similar in PCB composition and concentration to
the samples by mixing and diluting the individual standard stock
solutions. Example calibration solutions are shown in Tables 3
and 4. The mixture is diluted to volume with pesticide residue
analysis quality hexane. The concentration is calculated in
ng/mL as the individual PCBs. Dilutions are stored at 4°C in
narrow-mouth, screw-cap vials with Teflon cap liners. The menis-
cus is marked on the vial wall to monitor solvent evaporation.
These secondary dilutions can be stored indefinitely if the seals
are maintained.
These solutions are designated FSxxx ng PCB and SIMxxx pg PCB
where the xxx is used to encode the nominal concentration of the
lower congeners in ng/uL and pg/i-iL, respectively. The FS prefix
helps aid the analyst in identifying solutions which are appro-
priate for full scan analysis; the SIM prefix is for solutions to
calibrate in the selected ion monitoring and limited mass scan
acquisition modes.
6.6 Alternatively, certified stock solutions similar to those listed
in Tables 3 and 4 may be available from a supplier, in lieu of
the procedures described in Section 6.4.
6.7 DFTPP standard - A 50 ng/(jL solution of decafluorotriphenylphos-
phine (DFTPP), PCR Research Chemicals, Gainesville, Florida, is
prepared in acetone or another appropriate solvent.
6.8 Internal standard stock solution - Solutions of d6-3,3',4,4'-
tetrachlorobiphenyl (KOR Isotopes, Cambridge, MA) and 1-iodonaph-
thalene (Aldrich Chemical Company, Milwaukee, WI) or d12-chrysene
(KOR Isotopes, Cambridge, MA) are prepared at nominal concentra-
tions of 1-10 mg/mL in hexane. The solutions are further diluted
to give working standards.
NOTE - Any internal standard may be used, provided it meets the
following criteria: (a) it is not already present in the sample,
(b) it gives a strong, recognizable mass spectrum, (c) it does
not give mass spectral ions which interfere with PCB quantisation,
(d) it is chemically stable, and (e) it elutes in the PCB reten-
tion window. Ideally, several internal standards are used which
18
-------
have retention times spanning the PCB retention windows to improve
the response factor recision.
Alternatively, the four 13C-labeled PCBs listed in Table 5 may be
used as internal standards. The are available as a certified so-
lution from Toxic and Hazardous Materials Repository, U.S. Environ-
mental Protection Agency, Environmental Monitoring and Support
Laboratory, 26 West St. Clair Street, Cincinnati, Ohio 45268,
(513) 684-7327. This solution may be used as received or diluted
further. These solutions are designated "SSxxx," where the xxx
is used to encode the nominal concentration in ug/mL.
6.9 Solution stability - The calibration standard, internal standard
and DFTPP solutions should be checked frequently for stability.
These solutions should be replaced after 6 months, or sooner if
comparison with quality control check samples indicates compound
degradation or concentration change.
7.0 Calibration
Maintain a laboratory log of all calibrations.
7.1 Sampling train
7.1.1 Probe nozzle - Using a micrometer, the inside diameter
of the nozzle is measured to the nearest 0.025 mm (0.001
in.). Three separate measurements are made using differ-
ent diameters each time and obtain the average of the
measurements. The difference between the high and low
numbers must not exceed 0.1 mm (0.004 in.).
When nozzles become nicked, dented, or corroded, they
must be reshaped, sharpened, and recalibrated before use.
Each nozzle must be permanently and uniquely identified.
7.1.2 Pitot tube - The pitot tube must be calibrated according
to the procedure outlined in Method 2.7
7.1.3 Dry gas meter and orifice meter - Both meters must be
calibrated according to the procedure outlined in APTD-
0581.8 When diaphragm pumps with bypass valves are used,
proper metering system design is checked by calibrating
the dry gas meter at an additional flow rate of 0.0057
mVmin (0.2 cfm) with the bypass valve fully opened and
then with it fully closed. If there is more than ±2%
difference in flow rates when compared to the fully
closed position of the bypass valve, the system is not
designed properly and must be corrected.
19
-------
Table 5. Composition of Internal Standard Spiking Solution (SS100)
Containing 13C-Labeled PCBs
Congener
no.
211
212
213
214
Compound
4-Chloro-d' ,2' ,3' ,4' ,5' ,6'-13C6)-biphenyl
3,3' ,4,4'-Tetrachloro-(13C12)-biphenyl
2,2' ,3,3' ,5,5' ,6,6'-Octachloro-(13C12)-biphenyl
Decachloro-(13C12)-biphenyl
Abbreviations
13C-Cl!
13C-C14
13c-ci8
13c-ci10
Concentration
(|jg/mL)
100
250
400
500
ro
o
-------
7.1.4 Probe heater calibration - The probe heating system must
be calibrated according to the procedure contained in
APTD-0581.8
7.1.5 Temperature gauges - Dial and liquid filled bulb thermom-
eters are calibrated against mercury-in-glass thermometers.
Thermocouples should be calibrated in constant tempera-
ture baths.
7.2 The gas chromatograph must meet the minimum operating parameters
shown in Tables 6 and 7, daily. If all of the criteria are not
met, the analyst must adjust conditions and repeat the test until
all criteria are met.
7.3 The mass spectrometer must meet the minimum operating parameters
shown in Tables 2, 8, and 9, daily. If all criteria are not met,
the analyst must retune the spectrometer and repeat the test un-
til all conditions are met.
7.3.1 Full scan data acquisition - Quadrupole mass spectrom-
eters must meet the tuning criteria in Table 2. The
spectrometer must scan between m/z 150-550, although
wider scan ranges are permissible.
7.3.2 Limited mass scan data acquisition - Table 10 presents
a suggested set of LMS ranges. The mass spectrometer
should be set to at least unit resolution. The com-
puter acquisition parameters should utilize the minimum
threshold filtering necessary so as not to lose pertinent
data. Optimum acquisition parameters will vary depending
on the condition of the mass spectrometer and should be
checked daily.
The dwell times for the mass ranges given in Table 10
will vary with instrument and should be optimized to
allow at least five data points across a chromatographic
peak. Maximum sensitivity will be achieved when utiliz-
ing maximum dwell time.
Instruments having the capability to switch mass ranges
during an analysis required particular attention to the
switching points to assure minimal data loss. Switch-
ing points can be initially determined by analyzing a
highly concentrated Aroclor mixture while in the full
scan mode.
7.3.3 Selected ion monitoring data acquisition - Table 11 pre-
sents a suggested set of characteristic ions for SIM.
The SIM program must include at least two ions for each
analyte, generally the primary and secondary ions in
Table 11. The spectrometer should be set to at least
21
-------
Table 6. Operating Parameters for Capillary Column Gas Chromatographic System
Parameter
Recommended
Tolerance
Gas chromatograph
Column
Liquid phase
Liquid phase thickness
Carrier gas
Carrier gas velocity
Injector
Injector temperature
Injection volume
Initial column temperature
Column temperature program
Separator
Transfer line temperature
Tailing factor
Peak width1
Finnigan 9610
15 -30 m x 0.255 mm ID
Fused silica
DB-5 (J&W)
0.25 urn
Helium
30-45 cm/sb
"Grob" (split/splitless
mode)
250-270°C
1.0-2.0 uL
60-80°C (2 min)d
70°-300°C at 10°C/mine
None
280°C
0.7-1.5
7-10 s
Other*
Other
Other nonpolar
or semipolar
< 1 urn
Hydrogen
Optimum performance
Other0
Optimum performance
Other
Otherd
Other
Glass jet or other
Optimum^
0.4-3
< 15 s
Substitutions permitted with any common apparatus or technique provided
.performance criteria are met.
Measured by injection of air or methane at 270°C oven temperature.
Manufacturer's instructions should be followed regarding injection tech-
.nique.
With on-column injection, initial temperature equals boiling point of the
solvent; in this instance, hexane.
C12C110 elutes at 270°C. Programming above this temperature ensures a
fclean column and lower background on subsequent runs.
Fused silica columns may be routed directly into the ion source to prevent
separator discrimination and losses.
^High enough to elute all PCBs, but not high enough to degrade the column
.if routed through the transfer line.
Tailing factor is width of front half of peak at 10% height divided by
width of back half of peak at 10% height for single PCB congeners in solu-
.tion FSxxx ng PCB or SIMxxx pg PCB.
Peak width at 10% height for a single PCB congener in FSxxx ng PCB or
SIMxxx pg PCB.
22
-------
Table 7. Operating Parameters for Packed Column Gas Chromatography System
Parameter
Gas chromatograph
Column
Recommended
Finnigan 9610
180 cm x 0.2 cm ID
Tolerance
Other3
Other
Column packing
Carrier gas
Carrier gas flow rate
Injector
Injector temperature
Injection volume
Initial column temperature
Column temperature program
Separator
Transfer line temperature
Tailing factor0
Peak widthd
glass
3% SP-2250 on 100/
120 mesh Supelcoport
Helium
30 mL/min
On-column
250°C
1.0 uL
150°C, 4 min
150°-260°C at 8°/min
Glass jet
280°C
0.7-1.5
10-20 sec
Other nonpolar
or semipolar
Hydrogen
Optimum performance
Other
Optimum
^ 5 Mi-
Other
Other
Other
Optimum
0.4-3
< 30 sec
.Substitutions permitted if performance criteria are met.
°High enough to elute all PCBs.
Tailing factor is width of front half of peak at 10% height divided by
width of back half of peak at 10% height for single PCB congeners in solu-
.tion FSxxx ng PCB or SIMxxx pg PCB.
Peak width at 10% hei'ght for a single PCB congener in FSxxx ng PCB or
SIMxxx pg PCB.
23
-------
Table 8. Operating Parameters for Quadrupole Mass Spectrometer System
Parameter Recommended Tolerance
Mass spectrometer Finnigan 4023 Other3
Data system Incos 2400 Other
Scan range 95-550 Other
Scan time 1 sec Other
Resolution Unit Optimum performance
Ion source temperature 280°C 200°-300°C
Electron energy 70 eV 70 eV
^Substitutions permitted if performance criteria are met.
Greater than five data points over a GC peak is a minimum.
cFilaments should be shut off during solvent elution to improve instrument
stability and prolong filament life, especially if no solvent venting is
used.
24
-------
Table 9. Operating Parameters for Magnetic Sector Mass Spectrometer System
Parameter Recommended Tolerance
Mass spectrometer Finnigan MAT 311A Other9
Data system Incos 2400 Other
Scan range 98-550 Other
Scan mode Exponential Other
Cycle time 1.2 sec ' Otherb
Resolution 1,000 > 500
Ion source temperature 280°C 250-300°
Electron energy0 70 eV 70 eV
^Substitutions permitted if performance criteria are met.
Greater than five data points over a GC peak is a minimum.
cFilaments should be shut off during solvent elution to improve instrument
stability and prolong filament life, especially if no solvent venting is
used.
25
-------
Table 10. Limited Mass Scanning (LMS) Ranges For PCBs
Compound Mass range (m/z)
C^HeCli + i3C612C6H9Cl 186-198
C12H8C12 220-226
C12H7C13 254-260
C12H6C14 + C12D6 C14 + 13C12H6C14 288-310
C12H5C15 322-328
C12H4C16 356-362
r u n ^Qn-'iQR
Uj^r^L I 7 OjU Oj\j
C12H2C18 426-434
C12HC19 460-468
C12C110 496-502
C10H7I 254
p n o/i n
^18^12 t*rU
13C12H2C18 440-446
l3C19Clin 508-514
26
-------
Table 11. Characteristic SIM Ions for PCBs
Homolog
Ci2HgCl
C12H8C12
(-12^7(^3
C12H6C14
£12^5^5
C12H4C16
C^HgCly
£12^2^8
"12nd g
C12Cl1o
CioHyl
^12^6^4
^18^12
13C612C6H9C1
13C12H6C14
13C12H2C18
13Ci2Cl10
Primary
188 (100)
222 (100)
256 (100)
292 (100)
326 (100)
360 (100)
394 (100)
430 (100)
464 (100)
498 (100)
254 (100)
298 (100)
240 (100)
194 (100)
304 (100)
442 (100)
510 (100)
Ion (relative intensity)
Secondary
190 (33)
224 (66)
258 (99)
290 (76)
328 (66)
362 (82)
396 (98)
432 (66)
466 (76)
500 (87)
-
300 (49)
.
196 (33)
306 (49)
444 (65)
512 (87)
Tertiary
-
226 (11)
260 (33)
294 (49)
324 (61)
364 (36)
398 (54)
428 (87)
462 (76)
496 (68)
-
296 (76)
-
-
302 (76)
440 (87)
508 (68)
27
-------
unit resolution. The computer acquisition parameters
should utilize the minimum threshold filtering necessary
so as not to lose pertinent data. Optimum acquisition
parameters will vary depending on the condition of the
mass spectrometer and should be checked daily.
The dwell times for the mass given in Table 11 will vary
with instrument and should be optimized to allow at least
five data points across a chromatographic peak. Maximum
sensitivity will be achieved when utilizing maximum dwell
time.
Instruments having the capability to switch mass ranges
during an analysis require particular attention to the
switching points to assure minimal data loss. Switching
points can be initially determined by analyzing a highly
concentrated congener or Aroclor mixture while in the
full scan mode.
7.4 The PCB response factors (RF ) must be determined in triplicate
or other replication, as discussed below, using Equation 7-1 for
the analyte homologs.
RF _ _2 ll Fn 7-1
P • A1s * Mp q'
where RF = response factor of a given PCB isomer
A = area of the characteristic ion. for the PCB congener
" peak
M = mass of PCB congener in sample (micrograms)
A- = area of the characteristic ion for the internal
15 standard peak (d6-3,3',4,4'-tetrachlorobiphenyl,
iodonaphthalene, d12~chrysene or the 13C-labeled
PCBs)
M. = mass of internal standard in sample (micrograms)
I o
If specific congeners are known to be present and if standards
are available, selected RF values may be employed. For general
samples, solutions in Tables 3 and 4 or a mixture may be used as
the response factor solution. The PCB-surrogate pairs to be used
in the RF calculation are listed in Table 12.
Generally, only the primary ions of both the analyte and surrogate
are used to determine the RF values. If alternate ions are to be
used in the quantitation, the RF must be determined using that
characteristic ion.
28
-------
Table 12. Pairings of Analyte and Calibration Compounds
Analyte
a
Congener
no. Compound
1~3 C H Cl
4-15 C12H8C12
16-39 C12H7C13
40-81 C12H6C14
82-127 C12H5C15
128-169 C12H4C16
170~193 C^2rl3Cl7
194-205 C12H2C18
206-208 C12HC19
209 C12C110
Calibration standard
Congener
no.
1
7
30
50
97
143
183
202
207
209
Compound
2
2,4
2,4,6
2, 2', 4, 6
2,2',3',4,5
2, 2', 3, 4, 5, 6'
2, 2', 3', 4, 4', 5', 6
2, 2', 3, 3', 5, 5', 6, 6'
2,2' ,3,3' ,4,4' ,5,6,6'
Cl2dlO
aBallschmiter numbering system, see Table 1.
29
-------
The RF value must be determined in a manner to assure ±20% pre-
cision^ For instruments with good day-to-day precision, a running
mean (RF) based on seven values may be appropriate. A new value
is added each day and the oldest dropped from the mean. Other
options include, but are not limited to, triplicate determinations
of a single concentration spaced throughout a day or determination
of the RF at three different levels to establish a working curve.
If replicate RF values differ by greater than ±10% RSD, the sys-
tem performance should be monitored closely. If the RSD is
greater than ±20%, the data set must be considered invalid and
the RF redetermined before further analyses are done.
7.5 If the GC/EIMS system has not been demonstrated to yield a linear
response or if the analyte concentrations are more than one order
of magnitude different from those in the RF solution, a calibra-
tion curve must be prepared. If the analyte and RF solution con-
centrations differ by more than one order of magnitude, a cali-
bration curve should be prepared. A calibration curve should be
established with triplicate determinations at three or more con-
centrations bracketing the analyte levels.
7.6 The relative retention time (RRT) windows for the 10 homologs and
surrogates must be determined. If all congeners are not available,
a mixture of available congeners or an Aroclor mixture (e.g., 1016/
1254/1260) may be used to estimate the windows. The windows must
be set wider than observed if all isomers are not determined.
Typical RRT windows for one column are listed in Table 13. The
windows may differ substantially if other GC parameters are used.
8.0 Sample Collection, Handling, and Preservation
The sampling shall be conducted by competent personnel experienced with
this test procedure and cognizant of the constraints of the analytical
techniques for PCBs, particularly contamination problems.
8.1 Stack sampling
While several sampling protocols are available for collection of
PCBs, the modified EPA Method 5 described herein has been well
validated for PCB collection and recovery.6 In this protocol,
PCBs are collected in a Florisil adsorbent tube and in a series
of impingers in front of the adsorbent. Other sample collection
protocols may be used, provided that PCBs are quantitatively col-
lected and recovered from the train. A recent protocol for the
collection of semi volatile organics using a modified Method 5
train with XAD-2 as the adsorbent represents an attempt to stan-
dardize stack gas sampling.9 This protocol may be considered an
alternative to that described herein.
30
-------
Table 13. Relative Retention Time (RRT) Ranges of PCB Homologs
Versus d6-3,3',4,4'-Tetrachlorobiphenyl
PCB
homolog
Monochloro
Dichloro
Trichloro
Tetrachloro
Pentachloro
Hexachloro
Heptachloro
Octachloro
Nonachloro
Decachloro
No. of
isomers
measured
3
10
9
16
12
13
4
6
3
1
Observed range
of RRTsa
0.40-0.50
0.52-0.69
0.62-0.79
0.72-1.01
0.82-1.08
0.93-1.20
1.09-1.30
1.19-1.36
1.31-1.42
1.44-1.45
Congener
no.
1
7
30
50
97
143
183
202
207
209
Observed
RRTa
0.43
0.58
0.65
0.75
0.98
1.05
1.15
1.19
1.33
1.44
Projected
range of
RRTsD
0.35-0.55
0.45-0.80
0.55-1.00
0.55-1.05
0.80-1.10
0.90-1.25
1.05-1.35
1.10-1.50
1.25-1.50
1.35-1.50
The RRTs of the 77 congeners and a mixture of Aroclor 1016/1254/1260 were
measured versus 3,3',4,4'-tetrachlorobiphenyl-d6 (internal standard) using
a 15-m J&W DB-5 fused silica column with a temperature program of 110°C
for 2 min, then 10°C/min to 325°C, helium carrier at 45 cm/sec, and an on-
column injector. A Finnigan 4023 Incos quadrupole mass spectrometer oper-
ating with a scan range of 95-550 daltons was used to detect each PCB
congener.
The projected relative retention windows account for overlap of eluting
homologs and take into consideration differences in operating systems and
lack of all possible 209 PCB congeners.
31
-------
8.1.1 Pretest preparation - All train components shall be main-
tained and calibrated according to the procedure de-
scribed in APTD-0581,8 unless otherwise specified herein.
This should be done in the laboratory prior to sampling.
8.1.1.1 Cleaning glassware - All glass parts of the
train upstream of and including the adsorbent
tube and impingers, should be cleaned as de-
scribed in Section 3.1.1. Special care should
be devoted to the removal of residual silicone
grease sealants on ground glass connections of
used glassware. These grease residues should
be removed by soaking several hours in a chromic
acid cleaning solution prior to routine cleaning
as described above.
8.1.1.2 Solid adsorbent tube - 7.5 g of Florisil acti-
vated within the last 30 days and still warm
from storage in a 110°C oven, is weighed into
the adsorbent tube (prerinsed with hexane) with
a glass wool plug in the downstream end. A
second glass wool plug is placed in the tube to
hold the sorbent in the tube. Both ends of the
tube are capped with ground glass caps. These
caps should not be removed until the tube is
fitted to the train immediately prior to sampling.
8.1.2 Preliminary determinations - The sampling site and the
minimum number of sampling points are selected according
to Method I7 or as specified by the Agency. The stack
pressure, temperature, and the range of velocity heads
are determined using Method 27 and moisture content using
Approximation Method 47 or its alternatives for the pur-
pose of making isokinetic sampling rate calculations.
Estimates may be used. However, final results must be .
based on actual measurements made during the test.
The molecular weight of the stack gases is determined
using Method 3.7
A nozzle size is selected based on the maximum velocity
head so that isokinetic sampling can be maintained at a
rate less than 0.75 cfm. It is not necessary to change
the nozzle size in order to maintain isokinetic sampling
rates. During the run, the nozzle size must not be
changed.
A suitable probe length is selected such that all traverse
points can be sampled. Sampling from opposite sides for
large stacks may be considered to reduce the length of
probes.
32
-------
A sampling time is selected appropriate for total method
sensitivity and the PCB concentration anticipated. Sam-
pling times should generally fall within a range of 2 to
4 hr.
A buzzer-timer should be incorporated in the control box
(see Figure 1) to alarm the operator to move the probe to
the next sampling point.
8.1.3 Preparation of collection train - During preparation and
assembly of the sampling train, all train openings must
be covered until just prior to assembly or until sampling
is about to begin. Immediately prior to assembly, all
parts of the train upstream of the adsorbent tube are
rinsed with hexane. The probe is marked with heat resis-
tant tape or by some other method at points indicating
the proper distance into the stack or duct for each sam-
pling point.
Two hundred milliliters of water is placed in each of the
first two impingers, and the third impinger left empty.
CAUTION: Sealant greases must not be used in assembling
the train. If the preliminary moisture determination
shows that the stack gases are saturated or supersaturated,
one or two additional empty impingers should be added to
the train between the third impinger and the Florisil tube.
See Section 5.1.5. Approximately 200 to 300 g or more, if
necessary, of silica gel is placed in the last impinger.
Each impinger (stem included) is weighed and the weights
recorded to the nearest 0.1 g on the impingers and on
the data sheet.
Unless otherwise specified by the Agency, a temperature
probe is attached to the metal sheath of the sampling
probe so that the sensor is at least 2.5 cm behind the
nozzle and pitot tube and does not touch any metal.
The train is assembled as shown in Figure 1. Through all
parts of this method use of sealant greases such as stop-
cock grease to seal ground glass joints must be avoided.
Crushed ice is placed around the impingers.
8.1.4 Leak check procedure - After the sampling train has been
assembled, the probe heating system(s) is turned on and
set (if applicable) to reach a temperature sufficient to
avoid condensation in the probe. Time is allowed for the
temperature to stabilize. The train is leak checked at
the sampling site by plugging the nozzle and pulling a
380 mm Hg (15 in. Hg) vacuum. A leakage rate in excess
of 4% of the average sampling rate or 0.0057 mVmin
(0.02 cfm) whichever is less, is unacceptable.
33
-------
The following leak check instruction for the sampling
train described in APTD-05818 may be helpful. The pump
is started with bypass valve fully open and coarse adjust
valve completely closed. The coarse adjust valve is
partially opened and the bypass valve slowly closed until
380 mm Hg (15 in. Hg) vacuum is reached. The direction
of bypass valve must not be reversed. This will cause
water to back up into the probe. If 380 mm Hg (15 in. Hg)
is exceeded, either the leak check is conducted at this
higher vacuum or the leak check is ended as described
below and start over.
When the leak check is completed, the plug is first slowly
removed from the inlet to the probe and the vacuum pump
is immediately turned off. This prevents the water in
the impingers from being forced backward into the probe.
Leak checks shall be conducted as described above prior
to each test run and at the completion of each test run.
If leaks are found to be in excess of the acceptable rate,
the test will be considered invalid. To reduce lost time
due to leakage occurrences, it is recommended that leak
checks be conducted between port changes.
8.1.5 Train operation - During the sampling run, an isokinetic
sampling rate within 10%, or as specified by the Agency,
of true isokinetic shall be maintained. During the run,
the nozzle or any other part of the train in front of
and including the Florisil tube must not be changed.
For each run, the data required on the data sheets must
be recorded. An example is shown in Figure 4. The dry
gas meter readings are recorded at the beginning and end
of each sampling time increment, when changes in flow
rates are made, and when sampling is halted. Other data
point readings are taken at least once at each sample
point during each time increment and whenever significant
changes (20% variation in velocity head readings) neces-
sitate additional adjustments in flow rate.
The portholes are cleaned prior to the test run to mini-
mize change of sampling deposited material. To begin
sampling, the nozzle cap is removed, the probe heater
operational and temperature up, and the pi tot tube and
probe positions are verified (if applicable). The nozzle
is positioned at the first traverse point with the tip
pointing directly into the gas stream. The pump is
started and the flow adjusted to isokinetic conditions.
Nomographs are available for sampling trains using type
S pitot tubes with 0.85 ± 0.02 coefficients (C ), and
when sampling in air or a stack gas with equivalent
34
-------
FIELD DATA
PLANT.
DATE
PROBE LENGTH AND TYPE.
NOZZLE I.D
SAMPLING LOCATION .
SAMPLE TYPE
RUN NUMBER
OPERATOR
AMBIENT TEMPERATURE
BAROMETRIC PRESSURE .
STATIC PRESSURE. (P$)_
FILTER NUMBER (s)
ASSUMED MOISTURE.'. _
SAMPLE BOX NUMBER
METER BOX NUMBER
METER AHp
C FACTOR
PROBE HEATER SETTING.
HEATER BOX SET! ING
REFERENCE ap
SCHEMATIC OF TRAVERSE POINT LAYOUT
READ AND RECORD ALL DATA EVERY.
MINUTES
TRAVERSE
POINT
NUMBER
\. CLOCK TIME
SrTUNG \^LOCK>
TIMt.mm ^\
~~ ___
GAS METER READING
-------
density (molecular weight, M., equal to 29 ± 4), which
aid in the rapid adjustment of the isokinetic sampling
rate without excessive computations. If C and M , are
outside the above stated ranges, the nomograph cannot be
used unless appropriate steps are taken to compensate for
the deviations.
When the stack is under significant negative pressure
(height of impinger stem), the coarse adjust valve must
be closed before inserting the probe into the stack to
avoid water backing into the probe. If necessary, the
pump may be turned on with the coarse valve closed.
When the probe is in position, the openings around the
probe and porthole must be blocked off to prevent un-
representative dilution of the gas stream.
The stack cross section is traversed, as required by
Method I7 or as specified by the Agency. To minimize
chance of extracting deposited material, the probe nozzle
should not bump into the stack walls when sampling near
the walls or when removing or inserting the probe through
the portholes.
During the test run, periodic adjustments are made to
keep the probe temperature at the proper value. More
ice and, if necessary, salt is added to the ice bath to
maintain a temperature of less than 20°C (68°F) at the
impinger/silica gel outlet, to avoid excessive moisture
losses. Also, the level and zero of the manometer should
be periodically checked.
If the pressure drop across the train becomes high enough
to make isokinetic sampling difficult to maintain, the
test run should be terminated. Under no circumstances
should the train be disassembled during the test run to
determine and correct causes of excessive pressure drops.
At the end of the sample run, the pump is turned off, the
probe and nozzle removed from the stack, and the final
dry gas meter reading recorded. A leak check is performed,
with acceptability of the test run based on the same cri-
teria as in Section 8.1.4. The percent isokinetic is
calculated (see calculation section) to determine whether
another test run should be made. If there is difficulty
in maintaining isokinetic rates due to source conditions,
the Agency should be consulted for possible variance on
the isokinetic rates.
8.1.6 Blank train - For each series of test runs, a blank train
is set up in a manner identical to that described above,
but with the nozzle capped with aluminum foil and the
36
-------
exit end of the last impinger capped with a ground glass
cap. The train is allowed to remain assembled for a
period equivalent to one test run. The blank sample is
recovered as described in Section 8.3.
8.2 Static air sampling6 - The sampling procedure for static air is
identical to that described in Section 8.1 with the following ex-
ceptions: (a) impingers and a heatable probe are not required
prior to the adsorbent tube; and (b) the PCB concentrations may
dictate a longer or shorter sampling time.
The selection of sampling time and rate should be based on the
approximate levels of PCB residues expected in the sample. The
sampling rate should not exceed 14 L/min and may typically fall
in the range of 5 to 10 L/min. Sampling times should be more
than 20 min but should not exceed 4 hr.
8.3 Sample recovery - Proper cleanup procedure begins as soon as the
probe is removed from the stack at the end of the sampling period.
When the probe can be safely handled, all external particulate
matter near the tip of the probe nozzle is wiped off. The probe
is removed from the train and both ends closed off with aluminum
foil. The inlet to the train is capped off with a ground glass
cap.
The probe and impinger assembly are transfered to the cleanup area.
This area should be clean and protected from the wind so that the
chances of contaminating or losing the sample will be minimized.
The train is inspected prior to and during disassembly and any
abnormal conditions noted. The samples are treated as follows:
8.3.1 Adsorbent tube - The Florisil tube is removed from the
train and capped with ground glass caps.
8.3.2 Sample Container No. 1 - The first three impingers are
removed. The outside of each impinger is wiped off to
remove excessive water and other debris. The impingers
are weighed (stem included), and the weight recorded on
a data sheet. The contents are poured directly into
Container No. 1.
8.3.3 Sample Container No. 2 - Each of the first three impingers
are rinsed sequentially with 30-mL acetone and then with
30-mL hexane, and the rinses put into Container No. 2.
Material deposited in the probe is quantitatively recov-
ered using 100-mL acetone and then 100-mL hexane and
these rinses added to Container No. 2.
37
-------
8.3.4 Silica gel container - The last impinger is removed, and
the outside wiped to remove excessive water and other
debris. It is weighed (stem included), and the weight
recorded on the data sheet. The contents are transferred
to the used silica gel can.
8.4 Sample preservation - Samples should be storedxin the dark at 4°C.
Storage times in excess of 4 weeks are not recommended.
9.0 Sample Preparation6
9.1 Extraction
9.1.1 Adsorbent tube - The entire contents of the adsorbent
tube are expelled directly onto a glass wool plug in the
sample holder of a Soxhlet extractor. Although no extrac-
tion thimble is required, a glass thimble with a coarse-
fritted bottom may be used.
The tube is rinsed with 5-mL acetone and then with 15-mL
hexane and these rinses put into the extractor. The ex-
traction apparatus is assembled and the adsorbent ex-
tracted with 170-mL hexane for at least 4 hr. The ex-
tractor should cycle 10 to 14 times per hour. After
allowing the extraction apparatus to cool to ambient
temperature, the extract is transferred into a Kuderna-
Danish evaporator.
The extract is evaporated to about 5 ml on a steam bath
and the evaporator allowed to cool to ambient temperature
before disassembly. The extract is transferred to a 50-mL
separatory funnel and the funnel set aside.
9.1.2 Sample Container No. 1 - The aqueous sample is transferred
to a 1,000-mL separatory funnel. The container is rinsed
with 20-mL acetone and then with two 20-mL portions of
hexane, adding the rinses to the separatory funnel.
The sample is extracted with three 100 mL portions of
hexane and the sequential extracts transferred to a
Kuderna-Danish evaporator.
The extract is concentrated to about 5 ml and allowed to
cool to ambient temperature before disassembly. The ex-
tract is filtered through a micro column of anhydrous
sodium sulfate into a 50-mL separatory funnel containing
the corresponding Florisil extract from Section 9.1.1.
The micro column is prepared by placing a small plug of
glass wool in the bottom of the large portion of a dis-
posable pipette and then adding anhydrous sodium sulfate
until the tube is about half full.
38
-------
9.1.3 Sample Container No. 2 - The organic solution is trans-
ferred into a 1,000-mL separatory funnel. The container
is rinsed with two 20-mL portions of hexane and the rinses
added to the separatory funnel. The sample is washed with
three 100-mL portions of water. The aqueous layer is
discarded and the organic layer transferred to a Kuderna-
Danish evaporator.
The extract is concentrated to about 5 mL and allowed to
cool to ambient temperature before disassembly. The ex-
tract is filtered through a micro column of anhydrous
sodium sulfate into the 50-mL separatory funnel contain-
ing the corresponding Florisil and impinger extracts
(Section 9.1.2).
9.2 Cleanup - Two tested cleanup techniques are described below.10 De-
pending upon the complexity of the sample, one or both of the tech-
niques may be required to fractionate the PCBs from interferences.
If the sample extract is colored, the Florisil column cleanup may
be indicated.
9.2.1 Acid cleanup
9.2.1.1 Add 5 mL of concentrated sulfuric acid to the
separatory funnel containing the sample extract
and shake for 1 min.
9.2.1.2 Allow the phases to separate, transfer the
sample (upper phase) with three 1 to 2 mL
solvent rinses to a clean continer.
9.2.1.3 Back-extract sample extract with 5-10 drops
of distilled water. Pass through a short column
of anhydrous sodium sulfate and concentrate to an
appropriate volume.
9.2.1.4 Analyze as described in Section 10.0.
9.2.1.5 If the sample is highly contaminated, a second
or third acid cleanup may be employed.
9.2.2 Florisil column cleanup
9.2.2.1 Variations among batches of Florisil may affect
the elution volume of the various PCBs. For
this reason, the volume of solvent required to
completely elute all of the PCBs must be veri-
fied by the analyst. The weight of Florisil
can then be adjusted accordingly.
39
-------
9.2.2.2 Place a 20-g charge of Florisil, activated over-
night at 130°C, into a Chromaflex column. Settle
the Florisil by tapping the column. Add about
1 cm of anhydrous sodium sulfate to the top of
the Florisil. Pre-elute the column with 70-80
ml of hexane. Just before the exposure of the
sodium sulfate layer to air, stop the flow.
Discard the eluate.
9.2.2.3 Add the sample extract to the column.
9.2.2.4 Carefully wash down the inner wall of the column
with 5 ml of of the hexane.
9.2.2.5 Add 200 ml of 6% ethyl ether/hexane and set
the flow to about 5 mL/min.
9.2.2.6 Collect 200 mL of eluate in a Kuderna-Danish
flask. All the PCBs should be in this fraction.
Concentrate to an appropriate volume.
9.2.2.7 Analyze the sample as described in Section 10.0.
9.3 Optional Screening for Interferences Using GC/FID
Note: Since many sample matrices are one of a kind or in-
frequently encountered by the analyst, the effectiveness of the
extraction and cleanup for a matrix may be unknown. A simple
screen to assess whether the interferences have been reduced to a
tolerable level can both save GC/MS time and prevent contamination
of the GC/MS instrument with very dirty samples. This screen
should not be used to determine PCB levels under this analytical
method.
9.3.1 Using a GC system as described in Section 5.5.3, analyze
for background interferences.
9.3.2 A 2 m x 2 mm glass column packed with 3% SP-2250 on 1007
120 Supelcoport or equivalent is suggested. A flow rate
of 40 mL/min 95% air/5% methane or nitrogen is recom-
mended. The air and hydrogen flow rates should be suf-
ficient to keep the flame lit and to burn efficiently,
e.g., 300 mL/min air and 30 mL/min H2.
9.3.3 The recommended temperature program is from 50 to 250°C
at 20°C/min with an initial hold of 3 min and a final
hold of 10 min. The injector temperature should be
200°C and the detector 300°C.
40
-------
9.3.4 Set instrumental sensitivity comparable to the antici-
pated mass spectral sensitivity. It is advisable to
establish criteria for rejection of sample at a given
attenuation such as (a) any off-scale peaks in PCB
elution window, (b) a baseline rise of 40% full scale,
(c) other criteria which are indicative of "problem"
samples.
9.3.5 If the FID screen suggests that the sample is not amen-
able to analysis by GC/EIMS, the analyst may either (a)
cycle the sample through the same cleanup again if it
appears that the cleanup technique was overloaded by the
matrix the first time, (b) submit the extract to another
and cleanup techniques which may remove more interfer-
ences, or (c) analyze a new aliquot of sample by another
extraction or cleanup technique.
10.0 Gas Chromatographic/Electron Impact Mass Spectrometric Determination
10.1 Internal standard addition - Pipet an appropriate volume of in-
ternal standard solution into the sample. The final concentra-
tion of the internal standards must be in the working range of
the calibration and well above the matrix background. The in-
ternal standards are thoroughly mixed by mechanical agitation.
Note: The volume measurement of the spiking solution is critical
to the overall method precision. The analyst must exercise cau-
tion that the volume is known ±1% or better. Where necessary,
calibration of the pipet is recommended.
Note: If the 13Olabeled PCB mixture is used for internal stan-
dards, this same solution is used as a surrogate standard solu-
tion in the method for products/product waste11 and for water.12
In this method, the 13C-labeled PCBs are spiked after extraction,
so are used as internal standards.
Alternately, another internal standard solution such as the
d6-3,3',4,4'-tetrachlorobiphenyl, 1-iodonaphthalene and d12~
chrysene used in the product/product waste and water protocols
may be used, if acceptable RF precision and accuracy are shown
across the nomolog range.
10.2 Tables 2, and 6 through 11 summarize the recommended operating
conditions for analysis. Figure 5 presents an example of a
chromatogram.
The analyst may choose to operate the mass spectrometer at any
appropriate sensitivity, using either full scan, limited mass
scanning or selected ion monitoring acquisition. The sensitivity
selected will depend on anticipated PCB levels and the instrumental
41
-------
100.0n
RIC
ro
C12D6C14
13C12H6C14
13C612C6H9C1
C10H?I
50
—\
800
13:20
30
97
183
143
88960.
202
+
C18D12
+
13
C12H2C18
209
13 +
207
1000
16:40
1200
20:00
1400
23:20
1
1500
26:40
2C110
—\
1800
30:00
SCAN
TIME
Figure 5. Reconstructed ion chromatogram of calibration solution FS100 ng PCB obtained in the full scan mode. The
concentration of the 10 PCB calibration congeners, the four 13C-labeled PCB recovery surrogates, and the three inter-
nal standards are in Table 3. See Table 1 for PCB numbering system, Table 6 for capillary GC parameters, and Table 8
for mass spectrometer operating parameters.
-------
LOQ needed meet the required method LOQ. In general, the more
concentrated the PCBs, the greater the precision, accuracy, and
qualitative data confidence. Thus, if possible, the amount of
sample and concentration factor should be scaled so that full
scan acquisition may be utilized.
10.3 While the highest available chromatographic resolution is not a
necessary objective of this method, good chromatographic per-
formance is recommended. With the high resolution of HRGC, the
probability that the chromatographic peaks consist of single
compounds is higher than with PGC. Thus, qualitative and quanti-
tative data reduction should be more reliable.
10.4 After performance of the system has been certified for the day
and all instrument conditions set according to Tables 2, and 6
through 11, inject an aliquot of the sample onto the GC column.
If the response for any ion, including surrogates and internal
standard, exceeds the working range of the system, dilute the
sample and reanalyze. If the responses of surrogates, internal
standard, or analytes are below the working range, recheck the
system performance. If necessary, concentrate the sample and re-
analyze.
10.5 Record all data on a digital storage device (magnetic disk, tape,
etc.) for qualitative and quantitative data reduction as discussed
below.
10.6 The instrumental performance must be monitored from run-to-run.
The areas of internal standards must be consistent (e.g., ± 20%).
If a low area is encountered, the injection may be suspect.
The resolution and peak shape of the internal standards, surrogates,
and other peaks should be monitored during or immediately after
data acquisition. Poor chromatography may indicate a bad injec-
tion, matrix interferences, or column degradation.
10.7 If a "dirty" sample is encountered, the analyst must employ ap-
propriate measures to demonstrate that there is no memory or
carry-over to subsequent samples. To assess the system cleanli-
ness, a standard, blank sample, or solvent blank may be run.
If the system is contaminated, remedial efforts may include (a)
changing or cleaning the syringe, (b) cleaning the injector, (c)
baking out the column at its maximum temperature, (d) changing
to a new column, or (e) cleaning the ion source.
11.0 Qualitative Identification
11.1 Full scan data
11.1.1 The peak must elute within the retention time windows
set for that homolog (as described in Section 7.5).
43
-------
11.1.2 The unknown spectrum should be compared to that of an
authentic PCB. The intensity of the three largest ions
in the molecular cluster (two largest for monochlorobi-
phenyls) must match the ratio observed for a standard
within ± 20%. Fragment clusters with proper intensity
ratios should also be present. System noise at low con-
centration or interferences may skew the ion ratio be-
yond the ± 20% criteria. If the analyst's best judgement
is that a peak, which does not meet the qualitative cri-
teria, is a PCB, the peak may be included in the calcu-
lation, with a footnote explaining the data and the rea-
son for relaxing the criteria.
11.1.3 Alternatively, a spectral search may be used to auto-
matically reduce the data. The criteria for acceptable
identification include a high index of similarity.
11.2 Selected ion monitoring (SIM) or limited mass scan (IMS) data -
The identification of a compound as a given PCB homolog requires
that two criteira be met:
11.2.1 (1) The peak must elute within the retention time window
set for that homolog (Seciton 7.5); and (2) the ratio of
two ions obtained by LMS (Table 10) or by SIM (Table 11)
must match the ratio observed for a standard within ± 20%.
The analyst must search the higher mass windows, in par-
ticular M+70, to prevent mi sidentification of a PCB frag-
ment ion cluster as the parent. System noise at low
concentration or interferences may skew the ion ratio
beyond the ± 20% criteria. If the analyst's best judge-
ment is that a peak, which does not meet the qualitative
criteria, is a PCB, the peak may be included in the cal-
culation, with a footnote explaining the data and the
reason for relaxing the criteria.
11.2.2 If one or the other of these criteria is not met, inter-
ferences may have affected the results, and a reanalysis
using full scan EIMS conditions is recommended.
11.3 Disputes in interpretation - Where there is reasonable doubt as
to the identity of a peak as a PCB, the analyst must either
identify the peak as a PCB or proceed to a confirmational analysis
(see Section 13.0).
12.0 Quantitative Data Reduction
12.1 After a chromatographic peak has been identified as a PCB, the
compound is quantitated based either on the integrated abund-
ance of the EICP or the SIM data for the primary characteristic
ion in Tables 10 and 11. If interferences are observed for the
44
-------
primary ion, use the secondary and then tertiary ion for quanti-
tation. If interferences in the parent cluster prevent quantita-
tion, an ion from a fragment cluster (e.g., M-70) may be used.
Whichever ion is used, the RF must be determined using that ion.
The same criteria should be applied to the internal standard
compounds (Table 14).
Note: With the higher homologs, the mass defect from unity is
significant. For instance, the mass of the most intense peak
for decachlorobiphenyl is 497.6830. Areas, EICPs, etc., must
be based on the true mass, not the nominal mass, or erroneous
results may be obtained. In addition, the tuning of some quadru-
poles may be less stable at high masses. The data quality must
be monitored especially carefully for the higher homologs.
12.2 Using the appropriate response factor (RF ) as determined in Sec-
tion 7.4, calculate the mass of each PCB peak (M ) using Equation
12-1. . P
An ,
M = _E . _±_ . M
p Ais RFp is Eq' 12~1
where A = area of the characteristic ion for the analyte PCB
p peak
A. = area of the characteristic ion for the internal
standard peak
RF = response factor of a given PCB congener
M. = mass of internal standard added to sample extract
(micrograms)
12.3 If a peak appears to contain non-PCB interferences which cannot
be circumvented by a secondary or tertiary ion, either:
12.3.1 Reanalyze the sample on a different column which sepa-
rates the PCB and interferents;
12.3.2 Perform additional chemical cleanup (Section 9) and then
reanalyze the sample; or
12.3.3 Quantitate the entire peak as PCB.
12.4 Sum all of the peaks for each homolog and then sum those to yield
the total PCB mass, MT, in the sample. If a concentration-per-
peak or concentration-per-homolog reporting format is desired,
carry each value through the calculations in an appropriate manner.
For example, if the analysis is being conducted to satisfy regu-
latory requirements for by-product PCBs, results may need to be
reported on a per resolvable peak basis. One rule1 states that
PCBs in air releases to air must be below the practical limit of
45
-------
Table 14. Characteristic Ions for Internal Standards
and 13C-Labeled PCB Surrogates
Ion (relative intensity)
Compound
dg-3,4,31 ,4'-Tetra-
chlorobiphenyl
1-Iodonaphthalene
d12-Chrysene
13C612C6H9C1
13C12H6C14
13C12H2C18
13Ci2Clio
Abbreviation
d6-C!4
INAP
DCRY
13C-Cli
13C-C14
13c-ci8
13c-ci10
Primary
298 (100)
254
240
194 (100)
304 (100)
442 (100)
510 (100)
Secondary
300 (49)
127
-
196 (33)
306 (49)
444 (65)
512 (87)
Tertiary
296 (78)
-
-
-
302 (78)
440 (89)
514 (50)
46
-------
quantitation, defined as "10 micrograms per cubic meter [roughly
0.01 part per million (ppm)] per resolvable chromatographic peak."
For regulatory purposes, only peaks greater than the 100 ug/m3
cutoff may need to be reported. A second rule2 requires that
inadvertently generated and recycled PCBs "vented to the ambient
air are limited to less than 10 ppm" total PCBs.
12.5 Calculation of air sample volume6
12.5.1 Nomenclature
M = Mass of PCB represented by a chromatographic peak
P micrograms
MT = Total mass of PCBs in sample, micrograms
C = Concentration of PCBs in air, micrograms per cubic
a
meter, corrected to standard conditions of 20°C,
760 mm Hg (68°F, 29.92 in. Hg) on dry basis
A = Cross-sectional area of nozzle, square meter (square
n feet)
B = Water vapor in the gas stream, proportion by volume
ws
I = Percent of isokinetic sampling
MW = Molecular weight of water, 18 g/g-mole (18 lb/
w Ib-mole)
P. = Barometric pressure at the sampling site, mm Hg
bar (in. Hg)
P = Absolute stack gas pressure, mm Hg (in. Hg)
P tH = Standard absolute pressure, 760 mm Hg (29.92 in
Std Hg)
R = Ideal gas constant, 0.06236 mm Hg-m3/K-g-mole (21.83 in.
Hg-ft3/°R-lb-mole)
T = Absolute average dry gas meter temperature °K (°R)
T = Absolute average stack gas temperature °K (°R)
Tstd = standard absolute temperature, 293°K (528°R)
V, = Total volume of liquid collected in impingers and
silica gel, milliliters. Volume of water col-
lected equals the weight increase in grams times
1 mL/g
47
-------
V = Volume of gas sample as measured by dry gas meter,
m dcm (dcf)
V , . ,N = Volume of gas sample measured by the dry gas
m^s ' meter corrected to standard conditions,
dscm (dscf)
V /•std\ = Volume of water vapor in the gas sample cbr-
*• ' rected to standard conditions, scm (scf)
V. = Total volume of sample, milliliter
V = Stack gas velocity, calculated by EPA Method 2,7
s m/sec (ft/sec)
AH = Average pressure differential across the orifice
meter, mm H20 (in. H20)
p = Density of water, 1 g/mL (0.00220 Ib/mL)
w
6 = Total sampling time, minutes
13.6 = Specific gravity of mercury
60 = Seconds per minute
100 = Conversion to percent
12.5.2 Average dry gas meter temperature and average orifice
pressure drop - See data sheet (Figure 4).
12.5.3 Dry gas volume - Correct the sample volume measured by ,
the dry gas meter to standard conditions [20°C, 760 mm Hg
(68°F, 29.92 in. Hg)] by using Equation 12-2.
lii r i * -f *^ /• i i *
v std bar 13.6 .. v bar 13.6 Eq. 12-2
- Vm T -KV
where K = 0.3855°K/mm Hg for metric units
= 17.65 °R/in. Hg for English units
12.5.4 Volume of water vapor
Vw(std) = Vlc MVT P~ = K Vlc Eq' 12"3
where K = 0.00134 nrVmL for metric units
= 0.0472 ft3/mL for English units
48
-------
12.5.5 Moisture content
R = Vstd) _ E ,2_4
ws Vstd) + Vw(std) ' ^
If the liquid droplets are present in the gas stream, as-
sume the stream to be saturated and use a psychrometric
chart to obtain an approximation of the moisture per-
centage.
12.6 Concentration of PCBs in stack gas - Determine the concentration
of PCBs in the air according to Equation 12-5 and report in micro-
grams per cubic meter using Table 15. If an alternate reporting
format (e.g., concentration per peak) is desired, a different
report form may be used.
MT
C = K y— ! - ' Eq. 12-5
a Vm(std)
where K = 35.31 ft3/m3
12.7 I so kinetic variation
12.7.1 Calculations from raw data.
T 100 Ts EK Vlc +
-------
Table 15. Analysis Report
BY-PRODUCT PCBs IN AIR
Internal Sample No.
Notebook No.
Data File Code
Volume Collected [V , t-
Amount Extracted
External Sample No.
Sample Source
Sample Matrix
Final Volume
Analyte
homolog
1-C1
Mass M (ug)
Analyte
homolog
6-C1
Mass Mp (|jg)
2-C1
7-C1
3-C1
8-C1
4-C1
9-C1
5-C1
10-C1
Estimated Method LOQ
Total (MT)
Concentration (C.)
Highest concentration per resolvable chromatographic peak
|jg/m
M9/H3
Reported by:
Internal Audit:
EPA Audit:
Name
Name
Name
Signature/Date
Signature/Date
Signature/Date
Organization
Organization
Organization
50
-------
13.0 Confirmation
If there is significant reason to question the qualitative identifica-
tion (Section 11), the analyst may choose to confirm that a peak is not
a PCB. Any technique may be chosen provided that it is validated as
having equivalent or superior selectivity and sensitivity to GC/EIMS.
Some candidate techniques include alternate GC columns (with EIMS de-
tection), GC/CIMS, GC/NCIMS, high resolution EIMS, and MS/MS techniques.
Each laboratory must validate confirmation techniques to show equivalent
or superior selectivity between PCBs and interferences and sensitivity
(limit of quantisation, LOQ).
If a peak is confirmed as being a non-PCB, it may be deleted from the
calculation (Section 12). If a peak is confirmed as containing both PCB
and non-PCB components, it must be quantitated according to Section 12.3.
14.0 Quality Assurance
Each participating laboratory must develop a quality assurance plan
(QAP) according to EPA guidelines.13. Additional guidance is also avail-
able.14 The quality assurance plan must be submitted to the Agency (re-
gional QA officer) for approval prior to analysis of samples.
The elements of a QAP include:
• Title Page
• Table of Contents
• Project Description
• Project Organization and Responsibility
• QA Objectives for Measurement Data in Terms of Precision, Accuracy,
Completeness, Representativeness, and Comparability
Sampling Procedures
• Sample Custody
• Calibration Procedures and Frequency
• Analytical Procedures
Data Reduction, Validation and Reporting
Internal Quality Control Checks
Performance and System Audits
51
-------
• Preventive Maintenance
• Specific Routine Procedures Used to Assess Data Precision, Accuracy
and Completeness
• Corrective Action
• Quality Assurance Reports to Management
15.0 Quality Control
15.1 Each laboratory that uses this method must operate a formal qual-
ity control (QC) program. The minimum requirements of this pro-
gram consist of an initial and continuing demonstration of lab-
oratory capability by the analysis of check samples. The labora-
tory must maintain performance records to define the quality of
data that are generated.
15.2 Certification and performance checks - Prior to the analysis of
samples, the laboratory must define its routine performance. At
a minimum, this must include demonstration of acceptable response
factor precision with at least three replicate analyses; and anal-
ysis of a blind QC check sample (e.g., the response factor cali-
bration solution at unknown concentration submitted by the QA
officer). Acceptable criteria for the response factor precision
and the accuracy of the QC check sample analysis must be pre-
sented in the QA plan.
Ongoing performance checks should consist of periodic repetition
of the initial demonstration or more elaborate measures. More
elaborate measures may include control charts and analysis of QC
check samples consisting of other congeners or with matrix inter-
ferences.
15.3 Procedural QC - The various steps of the analytical procedure
should have quality control measures. These include but are not
limited to:
15.3.1 GC performance - See Section 7.2 for performance criteria.
15.3.2 MS performance - See Section 7.3 for performance criteria.
15.3.3 Qualitative identification - At least 10% of the PCB
identifications, as well as any questionable results,
should be confirmed by a second mass spectrometrist.
15.3.4 Quantisation - At least 10% of all manual calculations,
including peak area calculations, must be checked. After
changes in computer quantitation routines, the results
should be manually checked.
52
-------
15.4 Sample QC - Each sample and each sample set must have QC measures
applied to it to establish the data quality for each analysis re-
sult. The responses of the internal standards, general spectral
data quality, and consistency of the internal standard area are all
measures of the data quality on individual samples. Within a
sample set, analysis of replicates and standard addition samples
are measures of the precision and accuracy, respectively.
15.4.1 The general spectral data quality is indicative of the
overall reliability of the data for a sample. The levels
of the background, intensity ratios within chlorine clus-
ters, etc., must all be evaluated. If the data quality
is marginal, the analyst may footnote results with an
explanation regarding any doubts about the data quality.
If the data are unacceptable (see Section 11.0), either
the GC/MS determination or the entire analysis must be
repeated.
15.4.2 An easy and significant assessment of the data quality
is the consistency of the internal standard areas. If
the internal standard area is consistent, the injection
volume was correct and the system is operating within
general tolerances (i.e., the chromatography column is
transmitting compounds and the spectrometer is detecting
them). If the internal standard area does not meet the
criteria specified in the QAP, e.g., ± 20% of other in-
jections, the data must be reviewed. If the injection
or the GC/MS performance is suspect, the sample must be
reanalyzed, or other corrective action taken.
15.4.3 QC for small sample sets - For small sample sets (1-10
samples), the minimum QC requirements can be a heavy bur-
den. Analysts are encouraged to be efficient and bunch
similar samples to increase the size of a set. A set is
defined as a group of samples analyzed together by the
same extraction/cleanup technique and determined on the
GC/MS system on the same day or successive days under
the same conditions.
At least one method blank must be run. The blank must
be exposed to the same sources of contamination—solvent,
glassware, etc.--as the samples. If conditions change,
additional blanks must be generated. An example would
be a new lot of solvent, or a change in dishwashing pro-
tocol .
At least one sample must be run in replicate. Tripli-
cates are preferable, but duplicates may be acceptable.
The acceptable precision among replicates must be speci-
fied in the QAP.
53
-------
At least one sample must be analyzed by the standard
addition technique. The analyst may select the most
difficult sample, based on prior knowledge of the sample
set, or a random sample. Two aliquots of the sample are
analyzed, one "as is" and one spiked (surrogate spiking
and equilibration techniques are described in Section
9.2) with Solution FSxxx ng PCB or SIMxxx pg PCB. If
the analyst has no prior knowledge of the sample, the
spiking level should be in the middle of the calibrated
range for the mass spectrometer. If the concentrations
of PCBs are known to be high or low, the amount added
should be adjusted so that the spiking level is 1.5 to
4 times the measured PCB level in the unspiked sample.
The samples should be analyzed together and the quanti-
tative results calculated. The recovery of the spiked
compounds (calculated by difference) must be 70-130%.
If the sample is known to contain specific PCB isomers,
these isomers may be substituted for solution FSxxx ng
PCB or SIMxxx pg PCB.
15.4.4 QC for intermediate sample sets - With intermediate
(approximately 10-100 samples) sample sets, the number
of method blanks, replicates, and standard addition sam-
ples must comprise at least 10% each. For example, if
23 samples are to be analyzed as a set, 3 blanks, 3 dup-
licates, and 3 standard addition samples would be added
in to give a total of 32 samples, at a minimum.
15.4.5 QC for large sample sets - When a large sample analysis
program is being planned, the QA plan may propose spe-
cific QC measures. If no'ne are proposed, the guidelines
for intermediate sets may be followed. One QC measure
which may increase efficiency is the use of control
charts. If, for example,- the control charts establish
that there is no blank problem over the long term, the
percent of blanks may be reduced. Any changes in the
procedure (e.g., a new lot of solvent) will still, of
course, require a blank.
15.5 It is recommended that the participating laboratory adopt addi-
tional QC practices for use with this method. The specific prac-
tices that are most productive depend upon the needs of the lab-
oratory and the nature of the samples. Field duplicates or trip-
licates may be analyzed to monitor the precision of the sampling
technique. Whenever possible, the laboratory should perform
analysis of standard reference materials and participate in rele-
vant performance evaluation studies.
54
-------
16.0 Method Performance
The method performance has not been evaluated. Limits of quantisation;
average intralaboratory recoveries, precision, and accuracy; and inter-
laboratory recoveries, precision, and accuracy will be presented when
available.
17.0 Documentation and Records
Each laboratory is responsible for maintaining full records of the analy-
sis. Laboratory notebooks should be used for handwritten records. GC/MS
data must be archived on magnetic tape, disk, or a similar device. Hard
copy printouts may be kept in addition if desired. QC records should
be maintained separately from sample analysis records.
The documentation must describe completely how the analysis was performed.
Any variances from the protocol must be noted and fully described. Where
the protocol lists options (e.g., sample cleanup), the option used and
specifies (solvent volumes, digestion times, etc.) must be stated.
The remining samples and extracts should be archived for at least 2
months or until the analysis report is approved, whichever is longer,
and then disposed unless other arrangements are made. The magnetic tapes
of the analysis and hardcopy spectra, quantisation reports, work sheets,
etc., must be archived for at least 3 years.
55
-------
REFERENCES
1. USEPA. 1982. 40 CRF 761, Polychlorinated biphenyls (PCBs); manufactur-
ing, precision, distribution in commerce, and use prohibitions; use in
closed and controlled waste manufacturing processes. 47 FR 46980-46996.
2. USEPA. 1984. 40 CFR Part 761, Polychlorinated biphenyls (PCBs); manu-
facturing, processing, distribution commerce and use prohibitions; re-
sponse to individual and class petitions for exemptions. 49 FR 28154-
28209.
3. Erickson MD, Stanley JS, Radolovich G, Turman K, Bauer K, Onstot J,
Rose D, Wickham M. 1982. Analytical methods for by-product PCBs--
preliminary validation and interim methods. Washington, DC: Office of
Toxic Substances, U.S. Environmental Protection Agency. EPA-500/5-82-
006; NTIS No. PB83 127 696.
4. Erickson MD, Stanley JS, Radolovich G, Blair RB. 1983. Analytical
method: the analysis of by-product chlorinated biphenyls in commercial
products and product wastes. Revision 1, Prepared by Midwest Research
Institute for Office of Toxic Substances, U.S. Environmental Protection
Agency, Washington, DC, under Subcontract No. A-3044(8149)-271, Work As-
signment No. 17 to Battelle, Washington, DC, August 15, 1983.
5. Erickson MD, Stanley JS. 1982. Methods of analysis for incidentally
generated PCBs--literature review and preliminary recommendations.
Washington, DC: Office of Toxic Substances, U.S. Environmental Pro-
tection Agency. EPA-560/5-82-005; NTIS No. PB83 126573.
6. Haile CL, Baladi E. November 1977. Methods for determining the poly-
chlorinated biphenyl emissions from incineration and capacitor and tras-
former filling plants. U.S. Environmental Protection Agency. EPA-600/4-
77-048, 90 pp. NTIS No. PB-276 745/7G1.
7. U.S. Environmental Protection Agency. August 1977. Federal Register.
42:160.
8. Martin RM. Construction details of isokinetic source sampling equipment.
Environmental Protection Agency. Air Pollution Control Office. Publica-
tion No. APTD-0581.
9. ASME. January 1984. Test protocol: sampling for the determination of
chlorinated organic compounds in stack emissions. Prepared from Environ-
mental Standards Workshop. Sponsored by The American Society of Mechan-
ical Engineers, U.S. Department of Energy, and U.S. Environmental Pro-
tection Agency. Unpublished report available from Mr. C. 0. Velzy,
Charles R. Velzy Associates, Inc., Consulting Engineers, 355 Main Street,
Armonk, New York 10504.
10. Bellar TA, Lichtenberg JJ. 1981. The determination of polychlorinated
biphenyls in transformer fluid and waste oils. Prepared for U.S. Environ-
mental Protection Agency, EPA-600/4-81-045.
56
-------
11. Erickson MD. November 1984. USEPA. Analytical method: the analysis
of by-product chlorinated biphenyls in commercial products and product
wastes, revision 2. Draft special report no. 1. U.S. Environmental
Protection Agency Contract No. 68-02-3938. Work Assignment No. 6
12. Erickson MD. November 1984. USEPA. Analytical method: the analysis
of by-product chlorinated biphenyls in water, revision 2. Draft special
report no. 3. U.S. Environmental Protection Agency, Contract No. 68-
02-3938. Work Assignment No. 6.
13. USEPA. 1980. Guidelines and specifications for preparing quality as-
surance project plans. Office of Monitoring Systems and Quality Assur-
ance, QAMS-005/80.
14. USEPA. 1983. Quality assurance program plan for the Office of Toxic
Substances. Office of Pesticides and Toxic Substances, U.S. Environ-
mental Protection Agency, Washington, D.C.
57
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-560/5-85-011
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Analytical Method: The Analysis of By-product
Chlorinated Biphenyls in Air, Revision 2
5. REPORT DATE
May 1985
6. PERFORMING ORGANIZATION CODE
8201A06
7. AUTHOR(S)
Mitchell D. Erickson
8. PERFORMING ORGANIZATION REPORT NO.
Special Report No. 2
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Midwest Research institute
425 Volker Boulevard
Kansas City, MO 64110
10. PROGRAM ELEMENT NO.
Work Assignment No. 6
11. CONTRACT/GRANT NO.
68-02-3938
12. SPONSORING AGENCY NAME AND ADDRESS
Field Studies Branch (TS-798), Office of Toxic Substance:
U.S. Environmental Protection Agency
401 M Street, SW
Washington, DC 20460 ' .
13. TYPE OF REPORT AND PERIOD COVERED
Special (September 84 - May 85)
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
The EPA Work Assignment Manager is Daniel T. Heggem (202) 382-3990.
The EPA Project Officer is Joseph J. Breen (202) 382-3569.
16. ABSTRACT
This is a gas chromatographic/electron impact mass spectrometric (GC/EIMS) method
applicable to the determination of chlorinated biphenyls (PCBs) in air emitted from
commercial production through stacks, as fugitive emissions, or static (room, other
containers, or outside) air. The PCBs present may originate either as synthetic by-
products or as contaminants derived from commercial PCB products (e.g., Aroclors). The
PCBs may be present as single isomers or complex mixtures and may include all 209 con-
geners from monochlorobiphenyl through decachlorobiphenyl.
A variety of general and specific sample preparation options are presented in this
method. This method takes a different approach from those which rely on Aroclor mix-
tures for calibration and quantitation. In this method PCBs are detected and quanti-
tated by homolog group. The results can be summed to give a total PCB value comparable
to results generated by other methods or they may be presented as 10 individual homolog
values. This homolog distribution can provide additional quantitative information on
the composition and source of the PCBs.
The method performance is assessed for each sample. A set of four 13C-labeled PCBs is
employed as recovery surrogates. If the surrogates are recovered and other QC param-
eters are within acceptable limits, then the data may be considered valid.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
PCBs
Polychlorinated biphenyls
Chlorinated biphenyls
Analytical methods
Determination
By-products
GC/MS
Air
Stack gas
c. COSATl Field/Group
18. DISTRIBUTION STATEMENl
19. SECURITY CLASS (This Report)
Unlimited
21. NO. OF PAGES
63
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION is OBSOLETE
------- |