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

-------